Writing assignment
memo3/Dumbaugh.pdf
http://jpl.sagepub.com
Journal of Planning Literature
DOI: 10.1177/0885412208318559 2008; 23; 17 originally published online May 12, 2008; Journal of Planning Literature
Eric Dumbaugh Designing Communities to Enhance the Safety and Mobility of Older Adults: A Universal Approach
http://jpl.sagepub.com/cgi/content/abstract/23/1/17 The online version of this article can be found at:
Published by:
http://www.sagepublications.com
can be found at:Journal of Planning Literature Additional services and information for
http://jpl.sagepub.com/cgi/alerts Email Alerts:
http://jpl.sagepub.com/subscriptions Subscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
http://jpl.sagepub.com/cgi/content/refs/23/1/17 Citations
at Ebsco Electronic Journals Service (EJS) on December 30, 2009 http://jpl.sagepub.comDownloaded from
Journal of Planning Literature Volume 23 Number 1
August 2008 17-36 © 2008 Sage Publications
10.1177/0885412208318559 http://jpl.sagepub.com
hosted at http://online.sagepub.com
17
Designing Communities to Enhance the Safety and Mobility of Older Adults
A Universal Approach
Eric Dumbaugh Texas A&M University
By 2030, seventy-two million people, or fully one of every five Americans, will be aged sixty-five or older, raising important concerns about how our cities, communities, and transportation facilities will safety accommodate their travel needs. The current policy solution is to increase driver testing, provide senior-oriented paratransit services, and ultimately, to house older, nondriving adults in senior care communities where more mobile individuals can provide basic household-related services for them. These practices, while perhaps well intentioned, effectively segregate older adults from the larger community. We can do better. This article summarizes the available literature on the travel- related needs, abilities, and preferences of older adults. Through a synthesis of the current knowledge, it identifies four strategies that can be used to design communities and transportation systems to address the safety and mobility needs of an aging population in an inclusive, universal manner.
Keywords: older adults; traffic safety; mobility; community design; universal design
By 2030, seventy-two million people, or fully one of every five Americans, will be sixty-five or
older (U.S. Census Bureau 2004), raising important concerns about how cities and communities will address the safety and mobility needs of an aging population. Age-related declines in visual acuity, coordination, flexibility, and reaction times all com- bine to reduce an individual’s ability to operate a motor vehicle (Owsley 2004). In environments designed principally for automobile mobility, the result is that many otherwise capable older adults find themselves increasingly unable to accomplish basic, household-sustaining travel, resulting in a diminished capacity for independent living, an increased isola- tion from friends, relatives, and the broader commu- nity, and a corresponding deterioration in their psychological well-being (Bailey 2004; Coughlin 2001; Kerschner and Aizenberg 1999; Marottoli et al. 1997; Ragland, Satariano, and MacLeod 2005; Straight 1997).
Traffic safety is also an important concern. In terms of fatal crashes per mile traveled, adults aged eighty and older are roughly seven times more likely to be killed in a traffic crash than are individuals aged twenty-five to seventy. By 2030, the number of adults aged sixty-five or older killed on U.S. roadways is
projected to triple to more twenty-four thousand each year (Hakamies-Blomqvist 2004). Older adults who shift from driving to walking also confront substan- tial safety risks. Pedestrians are 20 times more likely to be killed on a per-mile-traveled basis than are motorists (Surface Transportation Policy Project 2004), and pedestrians older than sixty-five are fully twice as likely to be killed as are members of the pop- ulation as a whole (Fatality Analysis Reporting System 2006). While transit is often cited as a solu- tion to the hazards experienced by older pedestrians and motorists, transit accounts for less than 2 percent of all trips undertaken by older adults (Rosenbloom 2004; Straight 1997) and still poses at least some degree of exposure to injury or death, since the use of transit entails at least some walking to access transit stops and destinations.
The conventional policy prescription for address- ing these issues is to establish more stringent licens- ing requirements for older drivers to screen out those deemed “unsafe”1 and to supply nondriving older adults with senior-oriented paratransit services. While such programs may enhance the safety and mobility of older adults, their functional effect is to segregate, and potentially stigmatize, older road users. Is this the best that can be done?
at Ebsco Electronic Journals Service (EJS) on December 30, 2009 http://jpl.sagepub.comDownloaded from
Scope and Organization of this Review
This article departs from the conventional treatment of senior travel to examine the issue from the perspec- tive of universal design, which calls for solutions that integrate the needs and preferences of persons with a diverse array of abilities into a single, inclusive design (Mace, Hardie, and Place 1997). Rather than identify- ing countermeasures that minimally address the basic safety and mobility needs of older adults, this article instead seeks to develop a more thorough understand- ing of the actual travel barriers encountered by older adults, as well as outline specific design solutions that may meaningfully overcome them.
This article is organized into three principal sec- tions. The first examines how the travel of older adults differs from that of younger cohorts, paying attention to both the needs and preferences of older adults, as well as to how aging may affect the use of specific travel modes in different environments. The second section considers the safety challenges encountered by older adults, identifying the specific roadway and community design factors that result in their injury or death. The third section considers older adults’ needs, preferences, and abilities holistically and identifies four universal design strategies that can be applied to address their safety and mobility needs.
Aging, Mobility, and Travel
Not surprisingly, the travel behavior of older adults differs from that of younger cohorts. A recent exami- nation of data collected as part of the Nationwide Personal Transportation Survey reveals that work trips comprise an ever-declining share of travel as people exceed the age of sixty-five. Vehicle miles traveled (VMT) declines as people age as well. This is undoubtedly attributable to retirement and the elimination of work trips, the trip type that generates the majority of person- and household-level VMT. While this, by itself, is not a particularly notable find- ing, what is notable is that VMT among older adults has roughly doubled since 1983 (Rosenbloom 2004).
On a superficial level, it is tempting to view this increase in VMT as an indication that older adults have become more mobile in recent years. Yet, changes in the built environment that have necessitated longer trips are at least partially responsible, creating a grow- ing mobility disadvantage for older adults who do not or cannot drive. In an examination of the travel patterns of persons aged sixty-five and older, Rosenbloom
(2004) found that older nondrivers travel between 25 and 50 percent fewer miles than older drivers, and more important, make only half the number of trips, regardless of length. This disparity increases with age, with nondrivers aged seventy-five and older making only a third of the trips made by older drivers of the same age (Straight 1997). When compared with older drivers, older nondrivers make 15 percent fewer trips to the doctor, 60 percent fewer shopping and dining trips, and 65 percent fewer trips to visit friends and family. In short, non-driving older adults make fewer trips, interact less often with the broader community, and as stated in the title of a recent report by the Surface Transportation Policy Project, find themselves increasingly “stranded without options” (Bailey 2004).
Older Drivers and Driving Cessation
Given their convenience and flexibility, personal automobiles are the preferred mode of travel for older adults. Roughly 90 percent of all trips taken by older adults are accomplished as either a driver or a pas- senger of a private automobile (Rosenbloom 2004). Focus groups conducted with older adults revealed that they view personal automobiles as the primary means of accomplishing their travel objectives, and view the ability to drive as being synonymous with personal independence (Coughlin 2001; Kerschner and Aizenberg 1999). Older adults are extremely reluctant to give up driving, and adopt a host of com- pensatory strategies to offset age-related declines in driving performance and prolong their ability to drive. A recent study sponsored by the American Association of Retired Persons surveyed drivers over the age of seventy-five to understand their driving behavior. This study found that 63 percent of older drivers avoided driving at night, and 51 percent avoided driving on congested routes or during con- gested time periods. Older adults also reported avoid- ing driving on routes they viewed as unsafe, with 30 percent avoiding freeways and Interstates, and 9 per- cent avoiding multilane roads (Straight 1997).
But what happens when older adults forgo driving? In general, older adults do not replace driving with other means of independent travel, such as using tran- sit or walking, but instead begin to rely on friends and family members to chauffer them to their desired des- tinations. This reliance on others is not without costs. In a series of focus groups, older nondrivers consis- tently reported feelings of guilt, shame, and obliga- tion associated with their reliance on others, as well as a diminished sense of personal independence
18 Journal of Planning Literature
at Ebsco Electronic Journals Service (EJS) on December 30, 2009 http://jpl.sagepub.comDownloaded from
(Coughlin 2001; Kerschner and Aizenberg 1999). Studies examining the psychological effects of dri- ving cessation have further found that the shift from driver to nondriver is associated with significant increases in clinical depression, even after controlling for an individual’s health status and demographic characteristics (Marottoli et al. 1997; Ragland, Satariano, and MacLeod 2005).
Older nondrivers generally report a decreased sat- isfaction with their ability to accomplish their travel objectives, although communities that provide higher quality transit service appear to minimize the effects of driving cessation. In a study that compared the travel patterns of older drivers and nondrivers in met- ropolitan Boston, Coughlin (2001) found that older nondrivers living in the city were generally satisfied with their ability to access desired destinations, and actually reported making more trips than both subur- ban drivers and nondrivers, due to the quality and proximity of transit service. The increased travel of urban nondrivers consisted of more social visits to friends and relatives as well as more frequent shopping trips—the two trip types that older adults typically forgo when they cease driving.
Shifting to Transit
Coughlin’s findings suggest that transit can pro- vide an important alternative to the use of a personal automobile. Nevertheless, only about 2 percent of all trips undertaken by seniors are accomplished using public transportation (Rosenbloom 2004; Straight 1997). A survey of individuals 50 and older sought to identify the barriers that prevented older adults from using transit. Major barriers included the perception of crime, followed closely by the lack of accessible destinations and increased travel times associated with transit use (Ritter, Straight, and Evans 2002).
Nevertheless, the low rates of transit use among older adults are in part attributable to the characteris- tics of the built environment. Seventy-five percent of older adults live in areas where developmental densi- ties are too low to support conventional transit ser- vices (Freund 2004), and a third live in areas where no transit service is available (Burkhardt 2002). Further, half of adults over the age of seventy-five reported that they did not have a bus or transit stop of any kind within walking distance of where they live (Straight 1997). Many authors have noted that con- temporary land use patterns and the correspondingly low quality of transit service are major factors that limit transit use among older adults. Nevertheless, the perception that development patterns are difficult to
change leads these authors to advocate for flexible- route paratransit services (Alsnih and Hensher 2003; Freund 2004; Giuliano 2004; Straight 1997; Suen and Sen 2004), despite the fact that older adults view paratransit even less favorably than they do conven- tional, fixed-route, public transportation service (Coughlin 2001).
Once one accounts for the types of environments in which older adults live, older adults in higher den- sity areas use transit at much higher rates than those living in less dense environments. While transit is used for about 2 percent of the total trips undertaken by adults over the age of sixty-five, it accounts for between 9 and 12 percent of their travel in environ- ments where the density is ten thousand persons per square mile, or roughly eight dwelling units per acre2
(Giuliano 2004). Interestingly, older adults who use transit regularly are not particularly concerned with crime but instead with the limited service frequencies during off-peak periods and weekends, which are the times when they are most likely to use transit (Coughlin 2001).
Shifts to Walking
While automobile travel among older adults has increased dramatically during the last two decades, walking continues to constitute a relatively small share of their total travel. Data from the 1995 Nationwide Personal Transportation Survey indicates that walking accounts for only 4 to 5 percent of all trips made by persons over the age of sixty-five (Rosenbloom 2004). As with transit, the likelihood that an older adult will walk to accomplish their travel objectives is highly dependent on the characteristics of the built environment. And like other age cohorts, older individuals in higher density, mixed-use envi- ronments report higher overall rates of walking, with adults over the age of seventy-five in environments with densities of ten thousand persons per acre or greater accomplishing more than 20 percent of their total trips by walking (Giuliano 2004).
Such findings are not surprising. Research on the subject of walking has consistently found that higher residential densities and the presence of mixed land uses are associated with increased rates of walking (Cervero 1989; Cervero and Kockelman 1997; Dunphy and Fisher 1996; Frank and Pivo 1994). From a functional perspective, however, what matters is not simply that communities are dense or contain multiple land uses but that meaningful destinations are located within walking distance from where older adults live, preferably with sidewalks and intersections that permit
Dumbaugh / Designing Communities for Older Adults 19
at Ebsco Electronic Journals Service (EJS) on December 30, 2009 http://jpl.sagepub.comDownloaded from
them to walk there safely (Cambridge Systematics 1994; Cervero, 1996; Cervero and Gorham 1995; Friedman, Gordon, and Peers 1994; Handy 1996; Kitamura, Laidet, and Mokhtarian 1997; Lund 2003).
A survey sponsored by the American Association of Retired Persons recently asked adults aged forty- five and older about the community characteristics they viewed as being important as they aged. Safe neighborhoods were identified as the most important community characteristic, selected by 97 percent of respondents as a desired community trait. An over- whelming majority also indicated that having doc- tor’s offices (92 percent), shopping centers (84 percent), grocery stores (83 percent), and drug stores (80 percent) near their homes was either important or highly important (Matthew Greenwald & Associates 2003). While these survey results cannot be used to infer that older adults will walk to these destinations, they do indicate that individuals value having a mix of supportive nonresidential uses within their com- munity, and they provide a clear indication of which uses individuals view as being most important.
Walking Speeds and Acceptable Walking Distances
How far will older adults walk before shifting to another mode? The conventional rule of thumb used by many planners is that most individuals will walk to destinations that are located within a quarter mile of their home (Nelessen 1994; Duany, Plater-Zyberk and Speck 2000). This quarter-mile threshold is based on average walking speeds; the average pedestrian walks at a rate of roughly 4 feet per second (ft/s), which corresponds to the ability to travel a quarter of a mile in roughly 5.5 minutes. Yet, older adults often walk at speeds much lower than these averages, par- ticularly when they experience mobility impairments such as rheumatoid or hip arthritis, or when they require the assistance of a cane or walker (see Table 1). At 2.6 ft/s, the average speed at which a cane or crutch-assisted individual walks, the amount of time required to travel a quarter mile increases to roughly 9 minutes. These findings are particularly relevant in that 17 percent of individuals between the ages of seventy-five and seventy-nine require an assistive device to walk, increasing to more than a third of persons aged eighty-five and older (Ritter, Straight, and Evans 2002).
There has been little formal examination into the walking thresholds that older adults will accept before shifting to another mode, but it is likely that
older adults are even more sensitive to walking dis- tances than are younger cohorts. A recent survey by Ritter, Straight, and Evans (2002) found that exces- sive walking distances were identified by individuals aged fifty and older as the single greatest barrier to increased walking, with 40 percent of surveyed respondents indicating that destinations were too far to permit walking. Other major barriers included poor sidewalks (37 percent), the absence of resting places (33 percent), and dangerous intersections.
Perceptual Barriers to Walking
While environmental factors such as travel distances and the absence of pedestrian facilities can create sub- stantive barriers to walking, social and cultural factors also appear to influence an older adult’s likelihood of viewing walking as a viable means of travel. Individuals currently aged sixty-five and older came of age during the postwar boom in automobile ownership and travel (Jackson 1985; Kay 1997), and as revealed from focus groups, principally view walking as a recre- ational pastime rather than as a meaningful travel option (Coughlin 2001; Kerschner and Aizenberg 1999). Thus, while 65 percent of older adults report having a food store within a half-mile of their home, and half report having a pharmacy within a half-mile (Ritter, Straight, and Evans 2002), fewer than 5 percent of all trips undertaken by adults sixty-five and older are accomplished by walking (Rosenbloom 2004). This suggests that there may be some unrealized opportuni- ties for enhancing senior mobility by simply increasing their awareness that walking is an available option for traveling to nearby destinations.
Interventions that have sought to encourage physical activity among older adults have been suc- cessful at increasing walking, at least for the purposes of health promotion (Barnard, et al. 2004; Conn et al. 2003; King, Rejeski, and Buchner 1998; Tanaka, Reiling, and Seals 1998; Taylor et al. 2003). These
20 Journal of Planning Literature
Table 1 Average Walking Speed for Mobility Impairments Associated with Aging
Disability/Assistive Device Walking Speed (ft/s)
Cane or crutch 2.62 Hip arthritis 2.24 to 3.66 Rheumatoid arthritis 2.46 Walker 2.07 Wheelchair 3.55
Source: Adapted from Oxley, Fildes, and Dewar 2004.
at Ebsco Electronic Journals Service (EJS) on December 30, 2009 http://jpl.sagepub.comDownloaded from
findings strongly suggest that programs aimed at enhancing the perception of walking as a viable travel mode, combined with the provision of facilities aimed at making walking safe and comfortable, may help encourage higher rates of walking among older adults. Safe Routes to Schools programs, which have successfully increased rates of walking and bicycling among children by combining infrastructure improve- ments with education and outreach programs (Staunton, Hubsmith, and Kallins 2003), may serve as a useful model for developing a similar program to increase utilitarian walking among seniors.3
Older Adults and Traffic Safety
A second important consideration is the safety of older adults as they travel through their communities. The United States has the highest traffic fatality rate of any developed country in the world, experiencing roughly two- to three-times the number of per-capita traffic-related fatalities as its international peers (World Health Organization 2004). And beyond the population- level risks faced by travelers in the United States, older adults are more likely than other age cohorts to be killed in a traffic crash. Of the roughly forty thousand annual traffic fatalities that occur in the United States each year, adults aged sixty-five and older account for 17 percent of them (Fatality Analysis Reporting System 2006), despite accounting for only 12 percent of the total population, making fewer trips, and travel- ing substantially fewer miles than their younger cohorts. Based on conservative estimates of licensure and mileage, one author estimated that by 2030, more than twenty-four thousand persons aged sixty-five and older will be killed annually in traffic fatalities—fully three times the number of older adults killed in 2000 (Hakamies-Blomqvist 2004).
Crashes Involving Older Motorists
Statistics such as those presented above are often used to argue that older drivers are less safe than other age groups, as well as to advocate for more stringent driver licensing requirements for older motorists. Aging clearly has an effect on an individ- ual’s ability to operate a motor vehicle. The negative effects that age-related declines in visual acuity, coor- dination, and reaction times have on the ability to operate a motor vehicle have been well documented in the recent traffic safety literature (for recent reviews, see Dewar 2002; Owsley 2004). Yet, before one concludes that older drivers are more dangerous
than other motorists, several factors warrant consid- eration. Older bodies are more fragile than younger ones, making older adults more likely to be injured or killed when a crash occurs (Hauer 1988; Hakamies- Blomqvist 2004). Yet older drivers are no more likely than younger drivers to injure or kill other road users on a per-mile-traveled basis (Dulisse 1997; Maycock 1997), and are actually less likely to injure or kill a pedestrian (Hakamies-Blomqvist 2004).4 The current reliance on fatal crash statistics to evaluate the safety of specific subpopulations results in a “frailty bias” against older motorists. Older adults are overrepre- sented in fatal crash statistics not because they are more hazardous or more crash-prone than other age groups, but instead because they are more likely to be killed when a crash occurs—and thus to have their crashes recorded in annual fatal crash statistics.
Rather than being disproportionately prone to crashes, older adults generally drive more cautiously than younger age groups and are much less likely to be involved in crashes associated with irresponsible or reckless driving, such as single-vehicle, run-off- roadway crashes, crashes involving excessive speeds, or crashes involving a driver following another vehi- cle too closely (Hakamies-Blomqvist 2004; Federal Highway Administration 1993; Stamatiadis, Taylor, and McKelvey 1991). Such findings are consistent with what is known about the driving patterns of older adults. Older adults compensate for declining driving abilities by driving more carefully and at lower speeds. The result is a substantial decline in crashes associated with irresponsible or reckless dri- ving behavior (Hakamies-Blomqvist 2004).
Despite their more cautious behavior, older drivers are nevertheless more likely to be killed in a traffic crash than younger drivers and are particularly at risk when attempting to turn across multiple lanes of traffic on higher speed roads. More than half of all crashes involving older motorists are angle crashes, a crash type associated with a driver being struck by an oncoming vehicle when attempting a left turn or a U-turn at a dri- veway or intersection (Abdel-Aty, Chen, and Radwan 1999; Federal Highway Administration 1995; Hauer 1988; Kloeppel et al. 1995; Lyles and Staplin 1991; Maleck and Hummer 1986; Partyka 1983; Preusser et al. 1998; Ulfarsson, Kim, and Lentz 2006). The relative risk of involvement in an intersection-related crash increases substantially with age. Stamatiadis, Taylor, and McKelvey (1991) found that at age sixty-five, older drivers were no more likely to be involved in crashes at intersections than were other age groups. But by the age of seventy-five, older adults’ rate of being involved in
Dumbaugh / Designing Communities for Older Adults 21
at Ebsco Electronic Journals Service (EJS) on December 30, 2009 http://jpl.sagepub.comDownloaded from
an intersection-related crash is double that of younger cohorts. And by the age of eighty, older drivers are 2.5 times more likely to be involved in a crash at a signal- ized intersection and four times more likely to be involved in a crash at an unsignalized intersection. A study conducted by Chandraratna, Mitchell, and Stamatiadis (2002) estimated that a driver’s risk of being involved in a crash involving a left turn at an intersection increased by roughly 8 percent each year that the driver exceeded the age of sixty-five.
What makes intersections so hazardous for older adults? The key problem appears to be that declines in visual acuity associated with aging lead older adults to underestimate available gaps in oncoming traffic, and thus to attempt turning maneuvers in front of oncom- ing vehicles (Hakamies-Blomqvist 2004; Hallmark and Mueller 2004; Smiley 2004; Straight 1997; Wasielewski 1984). The tendency to misjudge traffic gaps is further evidenced by police citations at crash locations, with drivers older than sixty-five being twice as likely to be cited for failing to yield to oncoming traffic than are younger drivers (Matthias, De Nicholas, and Thomas 1996). The problem with iden- tifying safe gaps in oncoming traffic is exacerbated by higher vehicle operating speeds. Older drivers are gen- erally able to identify safe gaps in traffic when oncom- ing vehicles are traveling at speeds of 30 miles per hour or less, but they have increasing difficulty doing so when vehicles are traveling at higher speeds (Chandraratna, Mitchell, and Stamatiadis 2002; Scialfa et al. 1991; Staplin 1995; Yi 1996).
The design of roadways and road networks appears to increase older adults’ exposure to turn- related crashes. Conventional design practices attempt to develop a “functional hierarchy” of roads, unloading all nonlocal traffic onto higher speed, mul- tilane arterial thoroughfares. The resulting street net- work forces older drivers to travel on roadways that necessitate turns across multiple lanes of higher speed, through-moving traffic—the precise traffic conditions in which they are most likely to be involved in a crash. Thus, while the cause of these crashes may be partly attributable to the effects of aging, road designs that encourage high-speed travel, combined with an absence of a lower-speed road net- work that older adults can use, is also partly to blame.
Considering Motorist Safety Countermeasures
Strategies to limit roadway access, such as the installation of a raised median and the consolidation
or elimination of driveways, are the conventional means for addressing turn-related crashes. Studies examining the effectiveness of access management techniques generally report a 30 to 35 percent reduc- tion in crashes associated with the provision of a raised median, and a 4 percent reduction in crashes for each driveway that is eliminated (Bowman and Vecellio 1994; Bretherton et al. 1990, Gluck, Levinson, and Stover 1999). While these safety gains are impressive, it is important to note that these stud- ies consider crash reductions at the population level and do not examine how they affect the incidence of crashes among older adults.
This review was unable to identify a single study that examined the specific effects of access manage- ment techniques on the safety of older drivers, but such strategies would likely have a limited effect on reducing their exposure to turn-related crashes, since their principal effect is to relocate turning movements to signalized intersections and designated median breaks, rather than eliminating a driver’s need to turn across multiple lanes of traffic. To the extent that access management benefits older motorists at all, it is principally through the relocation of turning move- ments from unsignalized to signalized locations. Nevertheless, one of the principal effects of consoli- dating driveways and installing medians is to increase operating speeds for through-moving vehicles (Bonneson and McCoy 1996; Gluck, Levinson, and Stover 1999; Transportation Research Board 2000), suggesting that access management may have a mixed effect, or possibly even a negative effect, on the safety of older drivers.
Authors examining the safety of older drivers gen- erally conclude that protected-only left-turn phasing should be substituted for unprotected turns or protected/permitted left-turn phasing at signalized intersections, particularly along higher-speed thor- oughfares (Chandraratna, Mitchell, and Stamatiadis 2002; Hallmark and Mueller 2004; Matthias, De Nicholas, and Thomas 1996; Staplin et al. 1998; 2001; Ulfarsson, Kim, and Lentz 2006). While pro- tected-only left turns will undoubtedly enhance the safety of older motorists, it is a solution that comes at the expense of a roadway’s operational performance, since the shift to protected-only phasing results in a reallocation of signal time from the through-moving to the left-turn portions of the signal phase, thus increasing a roadway’s control delay,5 which is the principal determinant of level of service in urban environments (Transportation Research Board 2000). Such solutions are likely to be politically unpopular
22 Journal of Planning Literature
at Ebsco Electronic Journals Service (EJS) on December 30, 2009 http://jpl.sagepub.comDownloaded from
in areas where there is a mandate to address traffic congestion.
Crashes Involving Older Pedestrians
As with their behavior as drivers, older adults are generally more cautious as pedestrians. Older pedes- trians are much less likely than younger pedestrians to be involved in crashes associated with unsafe or reck- less pedestrian behavior, such as midblock “dart outs” or other improper crossings. Older adults also appear to be more vigilant in searching out opportunities for protected crossing locations than other age groups. While between 65 and 75 percent of all pedestrian fatalities occur at nonintersection locations, only about 35 percent of the fatal crashes involving older pedestrians do (Harkley and Zegeer 2004; Insurance Institute for Highway Safety 2006). Nevertheless, these statistics also reveal an alarming trend: despite searching for safe crossing opportunities and gener- ally behaving more cautiously than other age groups, older pedestrians are twice as likely to be killed, per capita, than are members of the population as a whole (Harkey and Zegeer 2004; Insurance Institute for Highway Safety 2006; Zegeer et al. 2002.
Most of the safety problems experienced by older pedestrians can be attributed to three principal fac- tors. First, conventional traffic control practices result in pedestrian intervals that are inadequate for older pedestrians. Second, even where pedestrian intervals are adequately timed, permitted left and right turns during the pedestrian phase result in pedestrians being struck by a vehicle while in the crosswalk. Finally, both issues are strongly related to a third and overarching safety issue, which is that roadways are generally designed for speeds that are too high to safely accommodate either older pedestrians or older motorists. Each of these issues is addressed in the sections below.
Signalized Intersection Timing
Because of the conventional policy mandate to alleviate traffic congestion, most traffic signals along arterial thoroughfares are designed to maximize vehi- cle through-put. This has resulted in practices that attempt to minimize the amount of time allocated to pedestrian crossing movements across arterial thor- oughfares. Standard engineering practice currently recommends timing pedestrian intervals for 4 ft/s— the average pedestrian walking speed—and then only providing enough time for the pedestrian to reach the center of the furthest lane of traffic, rather than
providing enough time for the pedestrian to fully complete the crossing and arrive at the adjacent curb.6
Given the obvious difficulties such timings will have on less mobile segments of the population, 3.5 ft/s has been increasingly encouraged to fulfill the requirements of the Americans with Disabilities Act (Dorfman 1997; Oxley, Fildes, and Dewar 2004; U.S. Department of Transportation 2003). The use of this lower crossing speed has been acknowledged (although not required) in the most recent edition of the Manual on Uniform Traffic Control Devices (Federal Highway Administration 2004).
While 3.5 ft/s is preferable to the 4 ft/s standard, most older adults typically travel at speeds well below even this more conservative design value. Observational studies examining the intersection crossing speeds of older pedestrians found that older adults, on average, crossed at speeds between 3.2 ft/s (Guerrier and Jolibois 1998) and 2.8 ft/s (Hoxie and Rubenstein 1994). Applying the 3.5 ft/s standard to a conventional 60-foot, 5-lane roadway (two 12-foot through lanes in each direction plus a 12-foot center turn lane), and applying the conventional timing methodology results in a pedestrian clearance phase of 15 seconds. While this timing is generally suffi- cient for the average person using a wheelchair to enter the center of the furthest lane of traffic, it is inadequate for the majority of older pedestrians. At 3.2 ft/s—the higher of the two average observed crossing values for older pedestrians—this timing will allow an older pedestrian to travel only about 48 feet before the signal changes and the opposing stream of traffic is released. The older pedestrian must still travel an additional 12 feet—and across a through-moving lane of traffic—to complete the crossing and arrive at the adjacent curb.
It is important to further point out that these calcu- lations are based on population averages. Those indi- viduals traveling at below-average speeds will find themselves required to travel even farther through oncoming traffic to fully cross the street. Further com- plicating matters for older pedestrians, most jurisdic- tions do not even use the more conservative of the two available design values. A survey of signal timing practices found that 85 percent of municipalities simply apply the Manual on Uniform Traffic Control Devices default value of 4 ft/s when establishing the pedestrian interval at signalized intersections, without conducting location-specific studies to determine the appropriateness of these intervals (TranSafety 1997).
The effects that conventional signal timing practices have on the comfort and safety of older adults is
Dumbaugh / Designing Communities for Older Adults 23
at Ebsco Electronic Journals Service (EJS) on December 30, 2009 http://jpl.sagepub.comDownloaded from
obvious, particularly to the older adults who use them. A series of focus groups conducted in Florida, California, and Michigan, sponsored jointly by the National Highway Traffic Safety Administration, the American Automobile Association, and the Beverly Foundation, asked adults aged sixty-five and older to identify the policy actions that would best enhance the mobility of seniors. Eighty percent listed safer intersec- tions and crosswalks, and 70 percent specifically rec- ommended lengthening the pedestrian crossing phase at intersections (Kerschner and Aizenberg 1999).
Permitted Turning Maneuvers
Establishing appropriate signal intervals solves only part of the problem. In addition to the hazards posed by oncoming vehicles, older pedestrians are also placed at risk from turning vehicles (Zegeer et al. 2002). In this case, the problem is created by permit- ted right and left turns during the pedestrian interval, which results in vehicles turning into the crosswalk while a pedestrian is present. Authors generally attribute this problem to a combination of inadequate sight distance at intersections and vision and mobility impairments that make older adults unable to quickly identify and respond to the hazards posed by turning motorists (Davies 1999; Guerrier and Jolibois 1998; Staplin et al. 2001; Zegeer et al. 2002).
While aging can impair an individual’s ability to identify and quickly respond to the hazards posed by turning vehicles, it is important to observe that the purpose of pedestrian signals is to provide a protected interval for pedestrians to cross the street. Permitting left and right turns during the pedestrian interval encourages conflicts between motorists and pedestri- ans. Contrary to conventional wisdom, there is no empirical evidence indicating that insufficient sight distances are responsible for these crashes, nor that improving sight distances would have any effect on reducing them. Most motorists simply assume that it is the pedestrian’s responsibility to yield to oncoming and turning vehicles, regardless of the intersection sight distance, and education programs attempting to instruct drivers to the contrary have been found to be largely ineffective at changing a driver’s subsequent driving behavior (Christie 2001; Cochrane Injuries Group Driver Education Reviewers 2001; Insurance Institute for Highway Safety 2001; Michael 2004; Mohan and Roberts 2001; Sarkar and Andreas 2004). In the absence of vigorous and continuous police enforcement of crosswalk laws, the only practicable solution to these conflicts is to either provide an
all-red pedestrian interval, or else eliminate both per- mitted right turns on red and permitted left turns (Britt, Bergman, and Moffat 1995; Zegeer et al. 2002).
Vehicle Speeds and the Safety of Older Adults
While lengthening the pedestrian interval and restricting turning movements at intersections may go a long way toward addressing the safety of older pedestrians, the more fundamental problem encoun- tered by older pedestrians—and pedestrians of all ages—is that contemporary design practice priori- tizes higher-speed vehicle travel and operational per- formance, typically at the expense of safety. Several studies that have been widely cited in the recent pedestrian safety literature have reported a direct cor- relation between vehicle speeds and pedestrian crash severity, with crash severity increasing exponentially as a function of vehicle speed (Anderson et al. 1997; Ashton 1982; Durkin and Pheby 1992).7 Equally as important—and less commented on—is a recent study by Garder (2004), which found that higher vehicle speeds were also associated with higher rates of pedestrian crashes. Using European models to derive a baseline estimate of the number of expected pedestrian crashes for 122 road segments in Maine, the author compared the predicted number of crashes for each location against the actual reported number. While most of the streets had posted speeds of 25 mph, actual operating speeds varied, and the author categorized the studied locations based on their observed operating speeds. Low speed roadways, defined as those having operating speeds of less than 20 mph, reported roughly half the number of crashes as predicted by the models. Moderate speed road- ways (20-25 mph) reported three times as many crashes as predicted, and high speed roadways (over 25 mph) reported fully five times as many crashes as predicted. In general, low-speed, “main street”-type designs reported the lowest rates of vehicle-pedestrian crashes, while downtown areas with wide travel lanes and higher operating speeds reported the highest rates.
The finding that roadways with lower operating speeds would report lower rates of pedestrian crashes is not surprising. Lower operating speeds provide motorists with more time to respond to potentially hazardous conditions, such as an older pedestrian entering the street, as well as reducing the distance needed to bring a vehicle to a stop once the hazardous condition is identified (i.e., stopping sight distance).
24 Journal of Planning Literature
at Ebsco Electronic Journals Service (EJS) on December 30, 2009 http://jpl.sagepub.comDownloaded from
Further, motorists traveling at lower speeds appear to be more inclined to accommodate crossing pedestri- ans. An observational study of crossing behavior con- ducted in Orono, Maine, found that where average vehicle speeds were less than 11 mph, 100 percent of motorists yielded to pedestrians at unsignalized crosswalks; as average vehicle speeds exceeded 20 mph, only 17 percent did (Garder 2001).
The problem with conventional roadway design practice is that it begins from the perspective that “every effort should be made to use as high a design speed as practical to attain a desired degree of safety” (AASHTO 2001, 67). This approach, derived from an erroneous interpretation of crash statistics in the 1960s (Dumbaugh 2005a), directs the current approach to safe roadway design, despite a growing body of evidence indicating that conventional higher- speed designs negatively affect safety for pedestrians and motorists alike (Dumbaugh 2006; Miles-Doan and Thompson 1999; Noland 2001; Noland and Oh 2004; Ossenbruggen, Pendharkar and Ivan 2001; Pucher and Dijkstra 2000, 2003). Attempts to reduce operating speeds to safe levels solely though the adoption of speed limits is largely ineffective in the absence of aggressive police enforcement (Armour 1986; Beenstock, Gafni, and Goldin 2001; Britt, Bergman, and Moffat 1995; Zaal 1994), with more than 75 percent of drivers in urban environments being observed to exceed posted speed limits, often by substantial margins (Chowdhury et al. 1998; Fitzpatrick et al. 2003; Fitzpatrick, Shamburger, and Fambro 1996; Tarris, Mason, and Antonucci 2000).
Unlike the United States, most other developed countries adopt reduced design speeds in developed areas where pedestrians are present or expected in order to promote traffic safety (Dumbaugh 2005a; Federal Highway Administration 2001; Lamm, Psarianos, and Mailaender 1999; Pucher and Dijkstra 2000, 2003; OECD 1998; Skene 1999). Indeed, while 30 mph is the minimum recommended design speed for arterial and collector thoroughfares in urbanized areas in the United States (AASHTO 2004), European countries view 30 mph (50 km/h) as the maximum acceptable design speed in environments with any degree of roadside development or pedes- trian activity (Lamm, Psarianos, and Mailaender 1999). While anecdotal, such practices may neverthe- less help to explain why European counties have not only much higher rates of walking when compared with the United States but also much lower rates of both pedestrian and motorist traffic fatalities, regard- less of whether safety is measured in terms of traffic
fatalities per capita or traffic fatalities per vehicle mile traveled (Transportation Research Board 2005; World Health Organization 2004).
Reconsidering the Garden City
While it is tempting to cite aging as the principal cause of the reduced mobility and heightened crash risks experienced by older adults, such an approach overlooks the designed-in barriers that older adults face. Many, if not most, of these barriers can be directly attributed to conventional community design practices, which emerged during the first half of the twentieth century as planners and engineers sought to balance the livability of cities and neighborhoods with the rapid increase in motor vehicle use. Based on influential works by individuals such as Lewis Mumford, Benton MacKaye, and Clarence Stein, and illustrated by “garden cities” such as Radburn and Chatham Village, the approach that emerged was to design communities to curb unwanted vehicle traffic in residential areas (Southworth and Ben-Joseph 1995). This was accomplished by strategies now familiar throughout the United States. First, planners and community designers sought to eliminate unwanted cut-through traffic by limiting the number of external access points in residential road networks, with the optimal design solution being a community that has exactly one external connection—thus elimi- nating cut-through traffic entirely. To further reduce neighborhood traffic volumes, nonresidential uses were moved outside of a community’s boundaries, typically onto arterial thoroughfares. These practices were encoded into development guidelines published and promoted by the Federal Housing Administration in the 1930s, and have since become institutionalized through local development codes and contemporary development and lending practices. The result is the single-use, disconnected subdivision design that characterizes most contemporary residential develop- ments in the United States (See Figure 1; Garvin 1998; Meyer and Dumbaugh 2004; Miles-Doan and Thompson 1999; Southworth and Ben-Joseph 1995).
While these development practices have been extremely effective at their intended purpose— eliminating unwanted vehicle traffic in residential communities—they have had three principal and largely unintended side effects. First, and perhaps most ironic, the decision to disconnect residential street net- works to prevent cut-through traffic has all but ensured that personal automobiles are the only meaningful
Dumbaugh / Designing Communities for Older Adults 25
at Ebsco Electronic Journals Service (EJS) on December 30, 2009 http://jpl.sagepub.comDownloaded from
form of transportation for people living in these com- munities. The disconnected road networks intended to prevent cut-through traffic also prevent pedestrians from taking direct routes to desired destinations, and further require transit vehicles to spend a great deal of their total revenue hours circling neighborhoods to col- lect riders prior to embarking on the line-haul portion of the trip. The result is that the distances between households and household-supportive retail attractions are typically too far to support walking (Frank, Engelke, and Schmid 2003; Frank, Stone, and Bachman 2000), and that conventional transit service is unattractive in the measure that matters most: travel time8 (Ewing 1996a; Friedman, Gordon, and Peers 1994; Parsons Brinkerhoff Quade and Douglas, Inc. 1996; Thompson and Frank 1995).
Second, the elimination of multiple network con- nections within residential communities has resulted in the need for high-volume, multilane roads to carry displaced local traffic. The results are increased traf- fic complexity, a need to traverse multiple lanes of traffic to accomplish a turning maneuver, and increased crossing distances for pedestrians, all devel- opmental characteristics associated with increases in both pedestrian and motorist crashes involving older
adults. These designs also diminish operational per- formance of the arterial roadways, since they funnel trips that might otherwise have been accommodated by local streets onto the arterial system. The result is additional demand for turning movements at signal- ized intersections and a corresponding reallocation of signal time, which in turn leads to an increase in control delay and a decline in a roadway’s level of service (Kulash 1990).
A third major side effect of conventional commu- nity design practice is the advent of the modern com- mercial strip. To eliminate nonresidential traffic in neighborhoods, municipal development codes require retail uses to locate along arterial thoroughfares, rather than within the communities they serve (Duany, Plater-Zyberk, and Speck 2000; Kay 1997; Kunstler 1994, 1998). While much is made of the negative visual quality of commercial strips (Duany, Plater- Zyberk, and Speck 2000; Kunstler 1994, 1998; Nelessen 1994), the major problem with these devel- opments for older adults is not aesthetic but func- tional. As either drivers or pedestrians, older adults are more likely to be injured or killed along high-speed, multilane thoroughfares. Planning practices that locate these uses along arterial thoroughfares place
26 Journal of Planning Literature
Figure 1 The New Garden City
at Ebsco Electronic Journals Service (EJS) on December 30, 2009 http://jpl.sagepub.comDownloaded from
older adults in the position of either traveling along roadways that are poorly suited to their abilities, or else forgoing the ability to independently accomplish their basic, household-supporting travel.
Overcoming Travel Barriers Experienced by Older Adults: A Universal Approach
Most of the safety and mobility barriers currently encountered by older adults are a direct product of con- ventional design practice. While authors examining the subject have recognized these barriers, the conven- tional solutions that emerge—more stringent licensing requirements and senior-oriented paratransit services—do nothing to eliminate the designed-in bar- riers that complicate older adults’ travel. Increased licensing requirements may effectively screen out potentially unsafe drivers but do not ensure that older adults can meaningfully accomplish their travel needs once they forgo driving. Instead, the default solution is to provide senior-oriented paratransit services that will shuttle older adults from their home to their desired destinations.9 While paratransit may enable older adults to safely accomplish their travel needs, it does so by adding an additional layer of complexity to a community’s transportation system, and while leaving the fundamental travel barriers they face unresolved.
Rather than providing services that minimally address basic needs, universal design instead chal- lenges designers to develop solutions that eliminate the safety and mobility barriers older adults face, preferably with solutions that address multiple needs with a single, inclusive design. With respect to older adults, this means moving beyond basic paratransit service to meaningfully rethink—and eliminate—the designed-in barriers that are responsible for the safety and mobility challenges older adults face.
Considering these barriers holistically, four strate- gies emerge. Yet to be truly “universal,” it is not enough for these strategies to simply address the needs of one user group, no matter what their abilities, at the expense of others, including those whose needs are already well served under contemporary practice. As such, the following strategies were identified for their ability to not only eliminate the safety and mobil- ity barriers encountered by older adults but to do so in a manner that is broadly compatible with the needs and concerns of road users of all ages and abilities.
Strategy 1: Complement the arterial system with a network of lower-speed, two-lane through-routes. While many older drivers avoid interstates and other higher-order roadways, the overwhelming majority
(90 percent) report that they feel comfortable driving on lower-speed, two-lane routes (Straight 1997). That older adults should feel more comfortable using these routes is not surprising. Lower-speed, two-lane routes allow older adults to more accurately identify safe gaps in oncoming traffic, and they eliminate the need to cross multiple lanes of through-moving traffic to accomplish a turn. In short, they are more compatible to older adults’ driving abilities. Yet rather than pro- viding lower-speed routes that older adults can safely use, contemporary community design practice is instead oriented toward unloading local traffic onto arterial thoroughfares, forcing older drivers to travel on roadways that are poorly matched to their needs and abilities.
Providing a complementary network of lower- speed roads has benefits that extend well beyond older drivers. First, lower operating speeds translate into a reduction in stopping sight distances, thus pro- viding through-moving motorists with the improved ability to avoid a crash should an older motorist—or any other road user—enter the roadway in front of them. They further reduce the likelihood that an older adult will turn inappropriately in front of an oncom- ing vehicle in the first place, since older adults are better able to identify safe gaps in oncoming traffic that travels at lower speeds (Scialfa et al. 1991; Staplin 1995). And perhaps most important for all age groups, roadways designed for lower operating speeds report fewer crashes, injuries, and fatalities for all age groups than do higher-speed, conventionally designed arterial roads (Dumbaugh 2005b, 2006; Ewing and Dumbaugh 2005; Lamm, Psarianos, and Mailaender 1999; Noland 2001; Noland and Oh 2004; OECD 1998; Pucher and Dijkstra 2000, 2003; Skene 1999). Finally, the inclusion of multiple through-routes enhances operational performance, since it allows traffic to be diffused along multiple routes rather than loaded exclusively on the arterial system (Alba and Beimborn 2005; Ewing 1996a; Kulash 1990, 2001; Transportation Research Board 2000; U.S. Environmental Protection Agency 2004).
Adding a network of secondary through-routes can also enhance the viability of conventional transit ser- vices as well. A rule of thumb used in transit service planning is that through-routes should be spaced at distances of no more than half a mile apart, which allows transit vehicles to eliminate the collection por- tion of the trip, thus enhancing the efficiency of line- haul services (Ewing 1996a; Thompson and Frank 1995). The conceptual framework illustrated in Figure 2 spaces households out at no more than one
Dumbaugh / Designing Communities for Older Adults 27
at Ebsco Electronic Journals Service (EJS) on December 30, 2009 http://jpl.sagepub.comDownloaded from
quarter mile from the nearest transit stop, provided the local street network is well connected (see Strategy 2). While residential densities of at least four units per acre or greater must be provided to ade- quately support basic bus service (Parsons Brinkerhoff Quade and Douglas, Inc. 1996), most conventional developments already achieve these densities by reducing individual lot sizes and apply- ing zero-lot line configurations.
The provision of a network of lower-speed road- ways need not entail a radical departure from conven- tional design practice. Instead, connected networks can be provided by simply modifying existing subdivision regulations to require the provision of secondary streets that connect across adjacent developments. Thus, rather than having future developments unload all of their traffic at a single point on the arterial sys- tem, collector streets can be aligned to create a sec- ondary road network, often at little or no additional cost to either municipalities or developers. By further relocating retail development from arterial thorough- fares to secondary routes (see Strategy 3), older drivers can accomplish their basic household-supporting travel without ever traveling on the arterial system.
Strategy 2: Enhance the connectivity of the local street network within communities, but ensure that vehicle speeds and volumes remain low. Connecting the local street network is increasingly on everyone’s
list of smart development practices.10 Connecting the local street network encourages walking because it permits more direct trip routing and reduces walking distances to transit stops, shopping, and other attrac- tions (Cervero and Gorham 1995; Frank, Stone, and Bachman 2000; Friedman, Gordon, and Peers 1994; Handy 1995, 1996; Lund 2003). It also appears to reduce automobile travel. Authors examining the effects of street connectivity have consistently found that better-connected street networks reduce both trip lengths and VMT, with the magnitude of the effect ranging from 5 percent to 57 percent (Cervero and Kockelman 1997; Kulash 1990; McNally and Ryan 1993; Rabiega and Howe 1994; Stone, Foster, and Johnson 1992). To the extent that there is disagreement among authors on the benefits of connecting local street networks, it is not about whether better con- nected streets decrease trip lengths—studies have uni- formly found that they do—but instead whether the reductions in VMT are truly meaningful, since the reduction in individual trip lengths may be offset with increased trip making (Crane 1999). From the per- spective of older adults, however, such debates are purely academic. For older adults—and particularly for those who do not drive—reduced trip lengths and increased trip frequencies are both desirable outcomes.
The principal objection to increasing local street connectivity is that it can encourage cut-through traf- fic on residential streets—the problem that conven- tional disconnected networks were intended to solve. Yet the problem is not so much traffic itself as it is traffic speeds and volumes that are incompatible with normal residential use. Diffusing traffic across a net- work reduces the individual traffic load that any indi- vidual street must carry (Alba and Beimborn 2005; Kulash 1990), and the introduction of traffic calming devices can reduce vehicle speeds to levels that are compatible with residential neighborhoods—ideally 20 mph or less (Burden 2000; Ewing 1999). To fur- ther discourage speeding and cut-through traffic, four-way stops or traffic circles11 can be introduced at local intersections, and long, uninterrupted streets can be broken into shorter segments, which limit a vehicle’s ability to accelerate to undesirable speeds (Ewing 1996a).
Strategy 3: Balance system capacity with opportu- nities for protected left turns and safe pedestrian crossings. Under conventional design practice, sig- nalized intersections pose a unique challenge for balancing safety and mobility. Signalized inter- sections determine the effective capacity of most
28 Journal of Planning Literature
Figure 2 Connected Higher- and Lower-Order Networks
1/2Mile 1/2Mile
Transit Stop
Arterial
Collector
Local
at Ebsco Electronic Journals Service (EJS) on December 30, 2009 http://jpl.sagepub.comDownloaded from
thoroughfares, yet they are also the locations where older pedestrians and motorists are most likely to be involved in a crash. The reason intersections are not presently more accommodating to older adults is that strategies to enhance older adults’ safety, such as increasing the pedestrian interval or adopting turn restrictions, also increase signal cycle lengths, lead- ing to a deterioration of a roadway’s operational per- formance (Gluck, Levinson, and Stover 1999; Lomax et al. 1997; Transportation Research Board 2000). Under conventional community design practice, there is no way resolve the impasse between safety and operational performance; increasing pedestrian inter- vals or eliminating permitted turns necessarily increases signal lengths. What solutions are available to enhance the safety of older road users while main- taining acceptable levels of operational performance?
Connected street networks are again the answer. Two principal factors are involved. First, because connected streets disperse traffic more evenly throughout the street network, the traffic load at any individual intersection is reduced, thereby reducing the signal time that must be allocated to any individ- ual movement (Transportation Research Board 2000). When local street networks are well con- nected, traffic can be diffused to levels that eliminate the need for multiple through-lanes (Alba and Beimborn 2005; Kulash 1990), which in turn reduces pedestrian crossing distances and the corresponding signal time that must be allocated to the pedestrian interval. Using a senior-appropriate timing of 2.6 ft/s along a conventional, seven-lane arterial (Six 12-foot through-lanes plus a center median) results in a pedestrian interval of 32 seconds. Reducing the road to a five-lane configuration drops the timing down to 23 seconds. If a two- or three-lane configuration can be used, the pedestrian interval can be reduced down to 9.5 or 14 seconds, respectively, while still fully accommodating the crossing needs of older or slower-moving pedestrians. Better still, if traffic vol- umes and vehicle speeds are kept low on two- or three- lane roadways, it may be possible to eliminate the need for protected left turns entirely, since the safety problems experienced by older adults are not left turns or street crossings per se, but instead the need to traverse multiple lanes of higher-speed, through-moving traffic.
The benefit of such an approach extends to all road users—not just the elderly or those experiencing mobility limitations that reduce their walking speeds. Shortening street crossing distances and ensuring that
pedestrian intervals are adequately timed enhances the safety and comfort of all pedestrians. Likewise, the shortening of the pedestrian interval permits more signal time to be allocated to through-moving traffic, which both increases the effective capacity of road- ways and road networks and enables drivers of all abilities to more quickly and comfortably reach their destinations.
Strategy 4: Encourage household-serving retail and services to locate in community-oriented centers, rather than in strip developments along arterials. While connected networks will go a long way toward enhancing the safety and mobility of older adults, multilane arterial thoroughfares will still be needed to address regional mobility. But if there is one thing that pedestrian advocates and traffic engineers can agree on, it is that arterial thoroughfares are poor locations for commercial and retail developments. New urbanists, “complete streets” advocates, and other groups concerned about the safety and comfort of pedestrians and bicyclists regularly call for solu- tions that reduce vehicle operating speeds on com- mercial arterial thoroughfares to levels found on lower-order streets (Duany, Plater-Zyberk, and Speck 2000; Ewing 1996b; McCann 2005; Skene 1999). Engineers, on the other hand, dislike strip develop- ments along arterial thoroughfares because they undermine the operational integrity of the arterial system and lead to higher crash rates due to the intro- duction of lower-speed, access-related traffic into the through-moving vehicle stream (Dumbaugh 2006; Gluck, Levinson, and Stover 1999). To the extent that these groups disagree, it is about the appropriate function of arterial thoroughfares that are forced to balance local and pedestrian access with the need to provide regional mobility (Duany, Plater-Zyberk, and Speck 2000; Dumbaugh 2005b; Ewing 2001; Ewing and King 2002; Gattis 2005; Institute of Transportation Engineers 2006). A solution that addresses the con- cerns of both groups is to simply prevent arterial strip developments from occurring in the first place. Rather than spreading commercial and retail develop- ment out along arterial roadways, these uses can instead be clustered into community centers along the network of secondary routes that traverse communi- ties. Such designs provide older motorists with the ability to access household-supporting retail uses without traveling on the arterial system, and can fur- ther enable older pedestrians to safely and comfort- ably access these uses as well.
Dumbaugh / Designing Communities for Older Adults 29
at Ebsco Electronic Journals Service (EJS) on December 30, 2009 http://jpl.sagepub.comDownloaded from
As with the other strategies outlined in this review, relocating retail uses from arterials to secondary routes benefits all road users, not just older adults. The inclu- sion of neighborhood-oriented retail within communi- ties has been found, time and again, to reduce VMT and to increase walking among all age groups (Cambridge Systematics, Inc. 1994; Cervero 1989, 1996; Cervero and Kockelman 1997; Frank and Pivo 1994; Frank, Stone, and Bachman 2000; Kockelman 1997; Lund 2003; Parsons Brinkerhoff Quade and Douglas, Inc. 1996. When configured into lower-speed, “main street”-type designs, they are also substantially safer for motorists and pedestrians alike than are conventional strip developments along arterials (Dumbaugh 2005a, 2005b, 2006; Garder 2004; Miles-Doan and Thompson 1999; Naderi 2003; Ossenbruggen, Pendharkar, and Ivan 2001). Finally, relocating commercial and retail developments away from arterial thoroughfares frees these roadways to serve the higher-speed, through- moving travel for which they are intended.
The implementation of neighborhood-oriented com- mercial developments need not entail the wholesale reinvention of contemporary engineering practice—a current project of many urban advocates (Duany, Plater-Zyberk, and Speck 2000; Duany Plater-Zyberk and Co. 2002; Forbes 2000; Institute of Transportation Engineers 2006; Kubilins 2000)—but can instead be achieved through the more thoughtful application of local land use regulations. More permissive zoning designations, coupled with ordinances that regulate form rather than use, can accomplish this end, par- ticularly when combined with architectural and design guidance that specifies desired development patterns (Nelessen 1994; Sperber 2005; Tracy 2003). This approach, termed “form-based coding” is already gaining currency within the planning profes- sion and provides a key means for encouraging the development of communities that better address the needs of older adults—and their younger counterparts as well.
Conclusion
While they remain comfortably able to drive, many older adults report being relatively satisfied with conventional communities (Ritter, Straight, and Evans 2002). Yet, aging naturally leads to declines in coordination, flexibility, reaction times, and visual acuity, all of which are essential components of the driving task. While declining driving abilities are a natural result of aging, conventional design practice
presupposes that residents will accomplish the major- ity of their travel as drivers. The result is the systematic creation of safety and mobility barriers to older adults.
This article has examined the extant literature on the safety and mobility of older adults to identify the specific barriers these individuals face. These barriers include a need to travel on higher-order streets to accomplish basic, household-sustaining travel objec- tives, the absence of accessible destinations, and insufficient pedestrian treatments along roadways and at intersections, among others. Rather than elim- inating these barriers, however, the conventional pol- icy solution is to instead provide senior-oriented paratransit services, which may enable older adults to accomplish their basic travel objectives but which does so only by segregating older adults from the broader community and by increasing the complexity of a community’s transportation system. For a solu- tion to be “universal,” it is not enough for it to merely address the minimum needs of individuals experienc- ing disabilities, whether temporary or permanent. These solutions should instead strive to eliminate the core barriers that persons with differing abilities face, preferably with integrated solutions that enable everyone’s needs to be accommodated within a sin- gle, inclusive design. From a universal design per- spective, the current reliance on paratransit services should be viewed as a stopgap solution at best.
The current emphasis on paratransit services stems not from an absence of concern for the safety and mobil- ity needs of older adults but instead from the perspective that most contemporary developments are unaccommo- dating to older adults, and that development practices are difficult to change. This is a highly pessimistic view that assumes that the future must be an extension of the very recent past. It also ignores latent opportunities for reshaping the built environment into forms that better address the needs of older adults. According to a recent report published by the Brookings Institution, fully half of the built environment existing in 2030 will have been built since 2000, with the majority of this development occurring in the “sunbelt states” that are attractive to retirees (Nelson 2004, 2006). Combined with the dou- bling of the retirement-aged population that is projected to occur during the same time period, this presents an unprecedented opportunity to redesign communities into more accommodating forms.
The four design solutions presented in this article—providing secondary through-routes, connect- ing the local street network, providing balanced inter- sections, and eliminating strip developments—are strategies that take advantage of these opportunities.
30 Journal of Planning Literature
at Ebsco Electronic Journals Service (EJS) on December 30, 2009 http://jpl.sagepub.comDownloaded from
They enhance the safety of older adults while they con- tinue to drive and provide them with meaningful alter- natives to driving if they ultimately forgo it. Yet, these strategies are also broadly beneficial to other individu- als, including not only those experiencing mobility impairments or other disabilities but also those who remain healthy and continue to drive. Safety, mobility, and livability need not be competing design objectives, nor must designs that benefit one group come at the expense of others. Through the adoption of a universal approach to community design, it is possible to ensure that the nation’s communities will be able to address the needs of an aging population in a manner that is inclusive, integrated, and broadly beneficial to persons with a full spectrum of abilities.
Notes
1. More than half of all states have adopted some combination of the following conditions for issuing licenses to older drivers: more frequent driver’s license tests, mandatory vision testing dur- ing the licensing process, mandatory in-person license renewals, and mandatory road tests (U.S. Government Accountability Office 2007).
2. Units per acre was determined by converting persons per square mile to persons per acre and assuming there is an average of two persons per dwelling unit.
3. There is a growing body of evidence indicating that increased walking can produce a number of health-related bene- fits for older adults, including improved cardio-respiratory fit- ness, balance, agility, and relief from arthritis suffering (Cunningham et al. 1987; Ettinger et al. 1997; Kovar et al. 1992; Shin 1999; Tanaka, Reiling, and Seals 1998; Taylor et al. 2003).
4. This may be at least partially attributable to the fact that older adults travel at lower speeds than do younger-aged cohorts. Since pedestrians are much less likely to be injured or killed when struck at lower speeds (Durkin and Pheby 1992; U.K. Department of Transport 1997; Zegeer et al. 2002), the tendency of older adults to travel at lower speeds would appear to lead to less severe crashes with pedestrians. It is unclear whether older drivers are also less likely to crash into pedestrians than are other age groups.
5. Control delay is the vehicle delay associated with the need to stop at a signalized intersection. In the absence of intersection control, most highway lanes would have the same capacity as freeway lanes, which can theoretically carry about 2,400 vehicles per hour per lane. Yet, the need to decelerate, stop, and wait for a traffic signal, and then accelerate back up to the roadway’s intended operating speed results in reductions in the number of vehicles that can travel along a given roadway in an hour, thereby leading to substantial reductions in the roadway’s actual capacity. Control delay is a measure that captures these effects, and it com- prises both the capacity lost while vehicles are stopped as well as the time lost while vehicles decelerate on the approach to a stop and accelerate after leaving it. Readers interested in a more thor- ough discussion of the calculation and application of control delay should refer to the Highway Capacity Manual (Transportation Research Board 2000), which is the leading
professional reference on analyzing the operational performance of roadways.
6. The rationale for this practice is that a stopped motorist will not deliberately drive into a pedestrian who is located in front of their vehicle. By timing the pedestrian signal to get the pedestrian only to the center of the outside travel lane, it is possible to release stopped vehicles on the inside lanes 1.5 seconds earlier, assuming that the roadway has both 12-foot lanes and is timed for a pedestrian crossing speed of 4 feet per second.
7. At 15 mph, most pedestrians will survive a crash, often sustaining only minor injuries. At 25 mph, almost all crashes result in severe injuries, and roughly half are fatal. At 40 mph, fully 90 percent of crashes are fatal.
8. While community circulators have been proposed to increase the efficiency of the line-haul transit services, they are unlikely to do much to boost overall transit ridership because they still require riders to circulate through a community before embarking on the line-haul portion of their trip. The use of circu- lators has the further disadvantage of adding a transfer—and thus additional wait times—to the transit trip.
9. Transit providers dislike providing these services due to their high operating costs. Paratransit serves only about 1 percent of the nation’s transit trips yet accounts for fully 9 percent of its operational expenditures. As a result, many transit service providers place specific limitations on who is eligible to use para- transit service, typically including only the elderly and those des- ignated as disabled, and who must thus be provided with paratransit service to meet the requirements of the Americans with Disabilities Act (American Public Transportation Association 2006; Millar 2005).
10. Many municipalities are beginning to establish street con- nectivity requirements as part of their subdivision regulations. A cursory scan of local subdivision regulations revealed that requirements specifying minimum levels of street connectivity or maximum block lengths have been adopted by not only the usual “smart growth” suspects in Oregon and Maryland but also by the cities of Austin, Charlotte, Durham, Gainesville, Nashville, Omaha, and Orlando, among others.
11. From an operational perspective, roundabouts and traffic circles are often preferred to four-way stop control, since they mandate compliance and enable vehicles to travel through an intersection without stopping (Ewing 1999). Yet, requiring vehi- cles to stop at intersections may be desirable along local streets. The type of intersection control used should be a matter of com- munity choice.
References
AASHTO. See American Association of State Highway and Transportation Officials.
Abdel-Aty, Mohamed A., Chien L. Chen, and A. Essam Radwan. 1999. Using conditional probability to find driver age effect in crashes. Journal of Transportation Engineering 125 (6): 502-7.
Alba, C. A., and Beimborn, E. 2005. Analysis of the effects of local street connectivity on arterial traffic. Transportation Research Board 84th Annual Conference Proceedings [CD- ROM]. Washington, DC: Transportation Research Board.
Alsnih, R., and D. A. Hensher. 2003. The mobility and accessi- bility expectations of seniors in an aging population. Transportation Research Part A, 37:903-16.
Dumbaugh / Designing Communities for Older Adults 31
at Ebsco Electronic Journals Service (EJS) on December 30, 2009 http://jpl.sagepub.comDownloaded from
American Association of State Highway and Transportation Officials. 2001. A policy on geometric design of highways and streets, 4th ed. Washington, DC: Author.
———. 2004. A policy on the geometric design of highways and streets, 5th ed. Washington, DC: Author.
American Public Transportation Association. 2006. Public trans- portation fact book, 57th ed. Washington, DC: Author.
Anderson, R. W. G., A. J. McLean, M. J. B. Farmer, H. Lee, and C. G. Brooks. 1997. Vehicle travel speeds and the incidence of fatal pedestrian crashes. Accident Analysis and Prevention 29 (5): 667-74.
Armour, M. 1986. The effect of police presence on urban driving speeds. ITE Journal February: 40-45.
Ashton, S. A. 1982. Vehicle design and pedestrian injuries. In Pedestrian accidents, edited by A. J. Chapman, F. M. Wade, and H. C. Foot, 169-202. New York: John Wiley and Sons.
Bailey, L. 2004. Aging Americans: Stranded without options. Washington, DC: Surface Transportation Policy Project.
Barnard, S., S. Dunn, E. Reddic, K. Rhodes, J. Russel, T. Tuitt, B. Velde, J. Walden, P. Wittman, and K. White. 2004. Wellness in Tillery: A community-built program. Family and Community Health 47: 151-57.
Beenstock, M., D. Gafni, and E. Goldin. 2001. The effect of traf- fic policing on road safety in Israel. Accident Analysis and Prevention 33 (1): 73-80.
Bonneson, J. A., and P.T. McCoy. 1996. Capacity and opera- tional effects of midblock left-turn lanes—final report, NCHRP Project 3-49. Washington, DC: Transportation Research Board, National Research Council.
Bowman, B. L., and R. L. Vecellio. 1994. Effect of urban and sub- urban median types on both vehicular and pedestrian safety. Transportation Research Record, Journal of the Transportation Research Board 1445:169-79. Washington, DC: Transportation Research Board, National Research Council.
Bretherton, W., J. Womble, P. Parsonson, and G. W. Black. 1990. One suburban county: Policy for selecting median treatments for arterials. Compendium of Technical Papers: 197-201. Washington, DC: Institute of Transportation Engineers.
Britt, J. W., A. B. Bergman, and J. Moffat. 1995. Law enforce- ment, pedestrian safety, and driver compliance with crosswalk laws: evaluation of a four-year campaign in Seattle. Transportation Research Record: Journal of the Transportation Research Board 1485:160-167.
Burden, D. 2000. Streets and sidewalks, people and cars: The cit- izen’s guide to traffic calming. Sacramento, CA: Local Government Commission.
Burkhardt, J. 2002. TCRP Report 82: Improving public transit options for older persons. Washington, DC: National Research Council.
Cambridge Systematics, Inc. 1994. The effects of land use and travel demand management strategies on commuting behav- ior. Technology Sharing Program, U.S. Department of Transportation. Washington, DC: U.S. Department of Transportation.
Cervero, R. 1989. America’s suburban centers—The land use–transportation link. Boston: Unwin Hyman.
———. 1996. Mixed land uses and commuting: Evidence from the American Housing Survey. Transportation Research A 30:361-77.
Cervero, R., and R. Gorham. 1995. Commuting in transit versus automobile neighborhoods. Journal of the American Planning Association 61:210-25.
Cervero, R., and K. Kockelman. 1997. Travel demand and the 3Ds: Density, diversity and design. Transportation Research D 2:199-219.
Chandraratna, S., L. Mitchell, and N. Stamatiadis. 2002. Evaluation of the transportation safety needs of older drivers. Lexington: University of Kentucky.
Chowdhury, M. A., D. L. Warren, H. Bissell, and S. Taori. 1998. Are the criteria for setting advisory speeds on curves still rel- evant? ITE Journal February: 32-45.
Christie, R. 2001. The effectiveness of driver training as a road safety measure: A review of the literature. Noble Park, Victoria, Australia: Royal Automobile Club of Victoria.
Cochrane Injuries Group Driver Education Reviewers. 2001. Evidence-based road safety: The driving standards agency’s schools programme. Lancet 358:230-32.
Conn, V. S., A. M. Marian, K. J. Burks, M. J. Rantz, and S. H. Pomeroy. 2003. Integrative review of physical activity inter- vention research with aging adults. Journal of the American Geriatrics Society 51:1159-68.
Coughlin, Joseph. 2001. Transportation and older persons: Perceptions and preferences. Washington, DC: AARP.
Crane, R. 1999. The impacts of urban form on travel: A critical review (Working Paper WP99RC1). Washington, DC: Lincoln Institute for Land Policy.
Cunningham, D. A., P. A. Rechnitzer, J. H. Howard, and A. P. Donner. 1987. Exercise training of men at retirement: A clin- ical trial. Journal of Gerontology 42 (1): 17-23.
Davies, D. 1999. Research, development, and implementation of pedestrian safety facilities in the United Kingdom (Report No. FHWA-RD-99-089). Washington, DC: Federal Highway Administration.
Dewar, R. E. 2002. Age differences: Drivers old and young. In Human factors in traffic safety, edited by R. E. Dewar and P. L. Olsen, 209-33. Tucson: Lawyers and Judges Publishing.
Dorfman, R. A. 1997. Taking a walk: No longer safe for elders in urban America and Asia. Journal of Aging and Identity 2 (2): 139–42.
Duany, A., E. Plater-Zyberk, E., and J. Speck. 2000. Suburban nation: The rise of sprawl and the decline of the American dream. New York: North Point Press.
Duany Plater-Zyberk and Co. 2002. The lexicon of the new urbanism, Version 3.2. Miami, FL: Duany Plater-Zyberk and Co.
Dulisse, B. 1997. Older drivers and risk to other road users. Accident Analysis & Prevention 29:573-82.
Dumbaugh, E. 2005a. Safe streets, livable streets. Journal of the American Planning Association 71 (3): 283-98.
———. 2005b. Safe streets, livable streets: A positive approach to urban roadside design. PhD diss., Georgia Institute of Technology, Department of Civil and Environmental Engineering.
———. 2006. The design of safe urban roadsides: An empirical analysis. Transportation Research Record: Journal of the Transportation Research Board 1961:74-82.
Dunphy, R. T., and K. Fisher. 1996. Transportation, congestion and density: New insights. Transportation Research Record 1552: 89-96.
32 Journal of Planning Literature
at Ebsco Electronic Journals Service (EJS) on December 30, 2009 http://jpl.sagepub.comDownloaded from
Durkin, M., and T. Pheby. 1992. York: Aiming to be the UK’s first traffic calmed city. Traffic Management and Road Safety. London: PTRC Education and Research Services.
Ettinger , W. H., Jr., R. Burns, S. P. Messier, W. Applegate, W. J. Rejeski, T. Morgan, S. Shumaker, M. J. Berry, M. O’Toole, J. Monu, and T. Craven. 1997. A randomized trial comparing aerobic exercise and resistance exercise with a health educa- tion program in older adults with knee osteoarthritis: The fit- ness arthritis and seniors trial (FAST). Journal of the American Medical Association 277 (1): 64-66.
Ewing, R. 1996a. Best development practices. Chicago: Planners Press.
———. 1996b. Pedestrian and transit-friendly design. Tallahassee: Public Transit Office, Florida Department of Transportation.
———. 1999. Traffic calming: State of the practice. Washington, DC: Institute of Transportation Engineers.
———. 2001. From highway to my way. Planning. January: 22-27.
Ewing, R., and E. Dumbaugh. 2005. The built environment and traffic safety. Paper presented at the 46th Annual Conference of the American Collegiate Schools of Planning, October.
Ewing, R., and M. King. 2002. Flexible design of New Jersey’s main streets. New Brunswick, NJ: Alan M. Voorhees Transportation Center.
Fatality Analysis Reporting System. 2006. Data collected and distributed by the National Highway Traffic Safety Administration (NHTSA). http://www-fars.nhtsa.dot.gov/ (accessed May 8, 2007).
Federal Highway Administration. 1993. Traffic maneuver prob- lems of older drivers: Final technical report (Report FHWA- RD-92-092). Washington, DC: U.S. Department of Transportation.
———. 1995. Traffic operations control for older drivers (FHWA-RD-94-119). Washington, DC: U.S. Department of Transportation.
———. 2001. Geometric design practices for European roads (Pub. No. FHWA-PL-01-026). Washington, DC: U.S. Department of Transportation.
———. 2004. Manual for uniform traffic control devices. Washington, DC: U.S. Department of Transportation.
Fitzpatrick, K., P. Carlson, M. Brewer, and M. D. Wooldridge. 2003. Design speed, operating speed, and posted speed limit practices. Proceedings of the Transportation Research Board 2003 Annual Meeting [CD-ROM]. Washington, DC: Transportation Research Board.
Fitzpatrick, K., B. Shamburger, and D. Fambro. 1996. Design speed, operating speed, and posted speed survey. Transportation Research Record: Journal of the Transportation Research Board 1523:55-60.
Forbes, G. 2000. Urban roadway classification: Before the design begins. Transportation Research E-Circular Number E-C019. December.
Frank, L., P. O. Engelke, and T. L. Schmid. 2003. Health and community design. Washington, DC: Island Press.
Frank, L. D., and G. Pivo. 1994. Impacts of mixed-use and den- sity on the utilization of three modes of travel: Single-occu- pant vehicle, transit, walking. Transportation Research Record 1466:44-52.
Frank, L. D., B. Stone, and W. Bachman. 2000. Linking land use with household emissions in central Puget Sound:
Methodological framework and findings. Transportation Research D 5:173-96.
Freund, K. 2004. Surviving without driving: Policy options for safe and sustainable senior mobility. In Transportation in an aging society: A decade of experience, 114-124. Washington, DC: National Academy of Sciences, Transportation Research Board.
Friedman, B., S. P. Gordon, and J. B. Peers. 1994. Effect of neo- traditional neighborhood design on travel characteristics. Transportation Research Record 1466:63-70.
Garder, P. E. 2001. Pedestrian safety in Maine. Augusta: Maine Department of Transportation.
———. 2004. The impact of speed and other variables on pedes- trian safety in Maine. Accident Analysis and Prevention 36:533-42.
Garvin, A. 1998. Are garden cities still relevant? Proceedings of the 1998 National Planning Conference. http://www.asu.edu/ caed/proceedings98/Garvin/garvin.html (accessed March 13, 2008).
Gattis, J. L. 2005. Counterpoint to “Safe Streets, Livable Streets.” Journal of the American Planning Association 71 (3): 16-18.
Giuliano, Genevieve. 2004. Land use and travel patterns among the elderly. In Transportation in an aging society: A decade of experience, 192-212. Washington, DC. National Academy of Sciences, Transportation Research Board.
Gluck, J., H. S. Levinson, and V. Stover. 1999. Impacts of access management techniques (NCHRP Report 420). Washington, DC: Transportation Research Board, National Academy of the Sciences.
Guerrier, J. H., and S. C. Jolibois. 1998. The safety of elderly pedestrians at five urban intersections in Miami. Proceedings of the Human Factors and Ergonomics Society 42nd Annual Meeting. Chicago.
Hakamies-Blomqvist, L. 2004. Safety of older persons in traffic. In Transportation in an aging society: A decade of experience, 22-35. Washington, DC. National Academy of Sciences, Transportation Research Board.
Hallmark, S. L., and K. Mueller. 2004. Impact of left-turn phas- ing on older and younger drivers at high-speed intersections. Ames: Iowa Department of Transportation.
Handy, S. 1995. Understanding the link between urban form and travel behavior. Paper presented at the 74th Annual Meeting of the Transportation Research Board. Washington, DC.
———. 1996. Urban form and pedestrian choices: Study of Austin neighborhoods. Transportation Research Record 1552:135-44.
Harkey, D. L., and C. V. Zegeer. 2004. PEDSAFE: Pedestrian safety guide and countermeasure selection system (Report No. FHWA-SA-04-003). Washington, DC: U.S. Department of Transportation.
Hauer, E. 1988. The safety of older persons at intersections. In Special Report 218. Transportation in an aging society: Improving mobility and safety for older persons Vol. 2, 194- 252.Washington, DC: Transportation Research Board.
Hoxie, R. E., and L. Z. Rubenstein. 1994. Are older pedestrians allowed enough time to cross intersections safely? Journal of American Geriatric Sociology 42 (11): 1219-20.
Institute of Transportation Engineers. 2006. Context-sensitive solutions in designing major urban thoroughfares for walka- ble communities. Washington, DC: Institute of Transportation Engineers.
Dumbaugh / Designing Communities for Older Adults 33
at Ebsco Electronic Journals Service (EJS) on December 30, 2009 http://jpl.sagepub.comDownloaded from
Insurance Institute for Highway Safety. 2001. Education alone won’t make drivers safer. Status Report 36:5.
———. 2006. Fatality facts 2005: Older people. http://www.iihs .org/research/fatality_facts/olderpeople.html#sec0 (accessed April 18, 2007).
Jackson, K. T. 1985. Crabgrass frontier: The suburbanization of the United States. New York: Oxford University Press.
Kay, J. H. 1997. Asphalt nation. Berkeley: University of California Press.
Kerschner, H., and R. Aizenberg. 1999. Transportation in an aging society: Focus group project. Washington, DC: AAA Foundation for Traffic Safety and the Beverly Foundation.
King, A. C., J. Rejeski and D. M. Buchner. 1998. Physical activ- ity interventions targeting older adults: A critical review and recommendations. American Journal of Preventive Medicine 15(4): 316-33.
Kitamura, R., L. Laidet, and P. Mokhtarian. 1997. A micro-analy- sis of land use and travel in five neighborhoods in the San Francisco Bay area. Transportation 24:51-65.
Kloeppel, E., R. D. Peters, C. James, J. E. Fox, and E. Alicandri. 1995. Comparison of older and younger driver responses to emergency driving events (Report FHWA-RD-95-056). McLean, VA: Office of Safety and Traffic Operations Research and Development, Federal Highway Administration, U.S. Department of Transportation.
Kockelman, K. M. 1997. Travel behavior as a function of acces- sibility, land use mixing, and land use balance: Evidence from San Francisco Bay area. Transportation Research Record 1607:116-25.
Kovar, P., J. Allegrante, C. Mackenzie, M. Peterson, B. Gutin, and M. Carlson. 1992. Supervised fitness walking in patients with arthritis of the knee: A randomized, controlled trial. Annals of Internal Medicine 116 (7): 529-34.
Kubilins, M. A. 2000. Designing functional streets that contribute to our quality of life. Urban Street Symposium Conference Proceedings. Transportation Research E-Circular Number E-C019. December.
Kulash, W. 1990. Traditional neighborhood development: Will the traffic work? Paper presented at the 11th Annual Pedestrian Conference, Bellevue WA, October. http:// user.gru.net/domz/kulash.htm (accessed March 13, 2008.).
Kulash, W. 2001. Can’t get there from here—Or can we? Forum for Applied Research and Public Policy 16 (2): 13-18.
Kunstler, J. H. 1994. The geography of nowhere: The rise and decline of America’s man-made landscape. New York: Free Press.
———. 1998. Home from nowhere: Remaking our everyday world for the 21st century. New York: Touchstone Press.
Lamm, R., B. Psarianos, and T. Mailaender. 1999. Highway design and traffic safety engineering handbook. New York: McGraw Hill.
Lomax, T. J., S. M. Turner, G. Shrunk, H. S. Levinson, R. H. Pratt, P. N. Bay, and G. B. Douglas. 1997. Quantifying con- gestion: Final report (NCHRP Report No. 398). Washington, DC: National Cooperative Highway Research Program.
Lund, H. 2003. Testing the claims of new urbanism: Local access, pedestrian travel and neighboring behaviors. Journal of the American Planning Association, 69 (4): 414-29.
Lyles, R. W., and L. Staplin. 1991. Age differences in motion per- ception and specific traffic maneuver problems. Transportation Research Record: Journal of the Transportation Research Board 1325:23–33.
Mace, R. L., G. J. Hardie, and J. P. Place. 1997. Accessible envi- ronments: Towards universal design. Raleigh, NC: Center for Universal Design.
Maleck, T. L., and J. E.Hummer. 1986. Driver age and highway safety. Transportation Research Record 1059:6-12.
Marottoli, R. A., C. R. Mendes de Leon, T. A. Glass, C. S. Williams, L. M. Cooney Jr., L. F. Berkman, and M. E. Tinetti. 1997. Driving cessation and increased depressive symptoms: Prospective evidence from the New Haven EPESE. Journal of the American Geriatrics Society 45:202-6.
Matthew Greenwald & Associates, Inc. 2003. These four walls . . . Americans 45+ Talk about home and community. Washington, DC: AARP.
Matthias, J. D., M. E. De Nicholas, and G. B .Thomas. 1996. A study of the relationship between left turn accidents and dri- ver age in Arizona (AZ-SP-9603). Phoenix: Arizona Department of Transportation.
Maycock, G. 1997. The safety of older car drivers in the European Union. Basingstroke, UK: European Road Safety Federation, Automobile Association Foundation for Road Safety Research.
McCann, B. 2005. Complete the streets! Planning, May: 18-23. Chicago: American Planning Association.
McNally, M. G., and S. Ryan. 1993. Comparative assessment of travel characteristics for neotraditional designs. Transportation Research Record: Journal of the Transportation Research Board, 1400:67-77.
Meyer, M. D., & E. Dumbaugh. 2004. Institutional and regulatory factors related to non-motorized travel and walkable commu- nities. Report commissioned by the Transportation Research Board and the Institute of Medicine, July. Incorporated in Special Report 282: Does the built environment influence physical activity? 2005. Washington, DC: Transportation Research Board, National Research Council.
Michael, S. 2004. What is the effect of driver education programs on traffic crash and violation rates?( Report No. FHWA-AZ- 04-546). Arizona: Arizona Department of Transportation.
Miles-Doan, R., and G. Thompson. 1999. The planning profes- sion and pedestrian safety: Lessons from Orlando. Journal of Planning Education and Research 18:211-20.
Millar, W. 2005. Testimony before the House Committee on Government Reform on the peformance and funding of the Washington Metropolitan Area Transit Authority’s METRO System. February 18.
Mohan, D., and I. Roberts. 2001. Global road safety and the con- tribution of big business: Road safety policies must be based on evidence. BMJ 323 (7314): 648.
Naderi, J. R. 2003. Landscape design in the clear zone: The effects of landscape variables on pedestrian health and driver safety. Transportation Research Board 82nd Annual Conference Proceedings [CD-ROM]. Washington, DC: Transportation Research Board.
Nelessen, A. C. 1994. Visions for a new American dream: Process, principles and an ordinance to plan and design small communities. Chicago: Planners Press.
Nelson, A. C. 2004. Toward a new metropolis: The opportunity to rebuild America. Washington, DC: The Brookings Institution.
———. 2006. Leadership in a new era. Journal of the American Planning Association 72 (4): 393-407.
Noland, R. B. 2001. Traffic fatalities and injuries: Are reductions the result of “improvements” in highway design standards?
34 Journal of Planning Literature
at Ebsco Electronic Journals Service (EJS) on December 30, 2009 http://jpl.sagepub.comDownloaded from
Transportation Research Board 80th Annual Conference Proceedings [CD-ROM]. Washington, DC: Transportation Research Board.
Noland, R. B., and L. Oh. 2004. The effect of infrastructure and demographic change on traffic-related fatalities and crashes: A case study of Illinois county-level data. Accident Analysis and Prevention 36:525-32.
OECD. See Organisation for Economic Co-Operation and Development.
Organisation for Economic Co-Operation and Development. 1998. Safety of Vulnerable Road Users (Publication No. DSTI/DOT/RTR/RS7(98)1/FINAL). Paris: Author.
Ossenbruggen, P. J., J. Pendharkar, and J. Ivan. 2001. Roadway safety in rural and small urbanized areas. Accident Analysis and Prevention 33:485-98.
Owsley, C. 2004. Driver capabilities. In Transportation in an aging society: A decade of experience, 44-55. Washington, DC: National Academy of Sciences, Transportation Research Board.
Oxley, J., B. N. Fildes, and R. E. Dewar. 2004. Safety of older pedestrians. In Transportation in an aging society: A decade of experience, 167-191. Washington, DC: National Academy of Sciences, Transportation Research Board.
Parsons Brinkerhoff Quade and Douglas, Inc. 1996. TCRP Report 16: Transit and urban form, Vol. 1. Washington, DC: Transportation Research Board.
Partyka, S. 1983. Comparison by age of drivers in two-car fatal crashes. Washington, DC: National Highway Traffic Safety Administration, U.S. Department of Transportation.
Preusser, D. F., A. F. Williams, S. A. Ferguson, R. G. Ulmer, and H. B. Weinstein. 1998. Fatal crash risk for older drivers at intersections. Accident Analysis & Prevention 30 (2): 151-9.
Pucher, J., and L. Dijkstra. 2000. Making walking and bicycling safer: Lessons from Europe. Transportation Quarterly 53 (3): 25-50.
———. 2003. Promoting safe walking and bicycling to improve public health: Lessons from the Netherlands and Germany. American Journal of Public Health 93 (9): 1509-16.
Rabiega, W. A. and D. A. Howe. 1994. Shopping travel efficiency of traditional, neotraditional, and cul-de-sac neighborhoods with clustered and strip commercial. Paper presented at the Association of Collegiate Schools of Planning 36th Annual Conference, Tempe, Arizona.
Ragland, D. R., W. A. Satariano, and K. E. MacLeod. 2005. Driving cessation and increased depressive symptoms. The Journals of Gerontology A: Biological Sciences and Medical Sciences 60 (3): 399-403.
Ritter, A. S., A. Straight, and E. Evans. 2002. Understanding senior transportation: Report and analysis of a survey of con- sumers age 50+. Washington, DC: AARP.
Rosenbloom, S. 2004. Mobility of the elderly: Good news and bad news. In Transportation in an aging society: A decade of experience, 3-21. Washington, DC: National Academy of Sciences, Transportation Research Board.
Sarkar, S., and M. Andreas. 2004. Drivers’ perception of pedes- trian rights and walking environments. In Transportation Research Record: Journal of the Transportation Research Board 1878: 75-82.
Scialfa, C. T., L. T. Guzy, H. W. Leibowitz, P. M. Garvey, and R. A.Tyrrell. 1991. Age differences in estimating vehicle velocity. Psychology and Aging 6 (1): 60-66.
Shin, Y. (1999). The effects of a walking exercise program on physical function and emotional state of elderly Korean women. Public Health Nursing 16 (2): 146–154.
Skene, M. 1999. Traffic calming on arterial roadways? Institute of Transportation Engineers compendium of technical papers. Washington, DC: Institute of Transportation Engineers.
Smiley, A. 2004. Adaptive strategies of older drivers. In Transportation in an aging society: A decade of experience, 36-43. Washington, DC. National Academy of Sciences, Transportation Research Board.
Southworth, M., and E. Ben-Joseph. 1995. Street standards and the shaping of suburbia. Journal of the American Planning Association 61 (1): 65-81.
Sperber, B. 2005. Function follows form. Professional Builder, September. http://www.housingzone.com/probuilder/article/ CA6253706.html (accessed March 13, 2008).
Stamatiadis, N., W. Taylor, and F. McKelvey. 1991. Elderly drivers and intersection accidents. Transportation Quarterly 45 (3): 377-90.
Staplin, L. 1995. Simulator and field measures of driver age differences in left-turn gap judgments. Transportation Research Record: Journal of the Transportation Research Board 1485:49-55. Washington, DC: National Research Council.
Staplin, L., K. Lococo, S. Byington, and D. Harkey. 2001. Highway design handbook for older drivers and pedestrians (Report No. FHWA-RD-01-103). Washington, DC: U.S. Department of Transportation.
Staplin, L. K. H. Lococo, J. A. McKnight, A. S. McKnight, and G. L. Odenheimer. 1998. Intersection negotiation problems of older drivers. Washington, DC: National highway Traffic Safety Administration. September.
Staunton, C. E., D. Hubsmith, and W. Kallins. 2003. Promoting safe walking and biking to school: The Marin County success story. American Journal of Public Health 93 (9): 1431-34.
Stone, J. R., M. S. Foster, and C. E. Johnson. 1992. Neo- traditional neighborhoods: A solution to traffic congestion? Site impact traffic assessment 72-76. New York: American Society of Civil Engineers.
Straight, A. 1997. Community transportation survey. Washington, DC: AARP.
Suen, S. L., and L. Sen. 2004. Mobility options for seniors. In Transportation in an aging society: A decade of experience, 97-113. Washington, DC. National Academy of Sciences, Transportation Research Board.
Surface Transportation Policy Project. 2004. Mean streets: How far have we come? Washington, DC: Surface Transportation Policy Project.
Tanaka, H., M. J. Reiling, and D. R. Seals. 1998. Regular walk- ing increases peak limb vasodilatory capacity of older hyper- tensive humans: Implications for arterial structure. Journal of Hypertension 16:423-28.
Tarris, J. P., J. M. Mason Jr., and N. Antonucci. 2000. Geometric design of low-speed urban streets. Transportation Research Record: Journal of the Transportation Research Board 1701:95-103.
Taylor, L., F. Whittington, C. Hollingsworth, M. Ball, S. King, V. Patterson, S. Diwan, C. Rosenbloom, and A. Neel. 2003. Assessing the effectiveness of a walking program on physical function of residents living in an assisted living facility. Journal of Community Health Nursing 20 (1): 15-26.
Dumbaugh / Designing Communities for Older Adults 35
at Ebsco Electronic Journals Service (EJS) on December 30, 2009 http://jpl.sagepub.comDownloaded from
Thompson, G., and J. Frank. 1995. Transit patronage as a prod- uct of land use potential and connectivity: The Sacramento case. Washington, DC: U.S. Department of Transportation.
Tracy, S. 2003. Smart growth zoning codes: A resource guide. Sacramento: Local Government Commission.
TranSafety. 1997. Researchers study the walking speeds of older pedestrians. Road Management and Engineering Journal. http://www.usroads.com/journals/rej/9704/re970404.htm (accessed April 18, 2007).
Transportation Research Board. 2000. Highway capacity manual. Washington, DC: National Academy of the Sciences.
———. 2005. Critical issues in transportation. Washington, DC: Transportation Research Board.
Ulfarsson, G. F., S. Kim, and E. T. Lentz. 2006. Factors affecting common vehicle-to-vehicle collision types: Road safety prior- ities in an aging society. Transportation Research Record: Journal of the Transportation Research Board 1980:70-78.
U.K. Department of Transport. 1997. Killing speed and saving lives. London: U.K. Department of Transport.
U.S. Census Bureau. 2004. U.S. interim projections by age, sex, race, and Hispanic Origin. Data published by the U.S. Census Bureau. http://www.census.gov/ipc/www/usinterimproj/ (accessed May 8, 2007).
U.S. Department of Transportation. 2003. Safe mobility for a maturing society: Challenges and opportunities. Washington, DC: U.S Department of Transportation. November.
U.S. Environmental Protection Agency. 2004. Characteristics and performance of regional transportation systems. Washington, DC: U.S. Environmental Protection Agency.
U.S. Government Accountability Office. 2007. Older driver safety: Knowledge sharing should help states prepare for
increase in older driver population. Report to the Special Committee on Aging, U.S. Senate. Washington, DC: Government Accountability Office, April.
Wasielewski, P. 1984. Speed as a measure of driver risk: Observed speeds versus driver and vehicle characteristics. Accident Analysis & Prevention 16:89-103.
World Health Organization. 2004. World report on road traffic injury prevention. Geneva: Author.
Yi, P. 1996. Gap acceptance for elderly drivers on rural highways. ITE 1996 compendium of technical papers, 299-303. Washington, DC: Institute of Transportation.
Zaal, D. 1994. Traffic law enforcement: A review of the literature. Netherlands: Institute of Road Safety Research.
Zegeer, C. V., C. Seiderman, P. Lagerwey, M. Cynecki, M. Ronkin, and R. Schnieder. 2002. Pedestrian facilities users guide: Providing safety and mobility (Report No. FHWA-RD-102-01). Washington, DC: Federal Highway Administration.
Eric Dumbaugh is an assistant professor in the Department of Landscape Architecture and Urban Planning at Texas A&M University. His research examines strategies for integrating mobility, traffic safety, and community livability into a holis- tic, context-based approach to roadway and community design. His recent publications include “Safe Streets, Livable Streets,” published in the Journal of the American Planning Association, and “The Design of Safe Urban Roadsides: An Empirical Analysis,” which received the Transportation Research Board’s Award for Outstanding Paper in Geometric Design in 2007.
36 Journal of Planning Literature
at Ebsco Electronic Journals Service (EJS) on December 30, 2009 http://jpl.sagepub.comDownloaded from
<< /ASCII85EncodePages false /AllowTransparency false /AutoPositionEPSFiles true /AutoRotatePages /None /Binding /Left /CalGrayProfile (Dot Gain 20%) /CalRGBProfile (sRGB IEC61966-2.1) /CalCMYKProfile (U.S. Web Coated \050SWOP\051 v2) /sRGBProfile (sRGB IEC61966-2.1) /CannotEmbedFontPolicy /Error /CompatibilityLevel 1.3 /CompressObjects /Off /CompressPages true /ConvertImagesToIndexed true /PassThroughJPEGImages true /CreateJobTicket false /DefaultRenderingIntent /Default /DetectBlends true /DetectCurves 0.1000 /ColorConversionStrategy /LeaveColorUnchanged /DoThumbnails false /EmbedAllFonts true /EmbedOpenType false /ParseICCProfilesInComments true /EmbedJobOptions true /DSCReportingLevel 0 /EmitDSCWarnings false /EndPage -1 /ImageMemory 1048576 /LockDistillerParams true /MaxSubsetPct 100 /Optimize false /OPM 1 /ParseDSCComments true /ParseDSCCommentsForDocInfo true /PreserveCopyPage true /PreserveDICMYKValues true /PreserveEPSInfo true /PreserveFlatness true /PreserveHalftoneInfo false /PreserveOPIComments false /PreserveOverprintSettings true /StartPage 1 /SubsetFonts true /TransferFunctionInfo /Apply /UCRandBGInfo /Remove /UsePrologue false /ColorSettingsFile () /AlwaysEmbed [ true /ACaslon-Ornaments /AGaramond-BoldScaps /AGaramond-Italic /AGaramond-Regular /AGaramond-RomanScaps /AGaramond-Semibold /AGaramond-SemiboldItalic /AGar-Special /AkzidenzGroteskBE-Bold /AkzidenzGroteskBE-BoldIt /AkzidenzGroteskBE-It /AkzidenzGroteskBE-Light /AkzidenzGroteskBE-LightOsF /AkzidenzGroteskBE-Md /AkzidenzGroteskBE-MdIt /AkzidenzGroteskBE-Regular /AkzidenzGroteskBE-Super /AlbertusMT /AlbertusMT-Italic /AlbertusMT-Light /Aldine401BT-BoldA /Aldine401BT-BoldItalicA /Aldine401BT-ItalicA /Aldine401BT-RomanA /Aldine401BTSPL-RomanA /Aldine721BT-Bold /Aldine721BT-BoldItalic /Aldine721BT-Italic /Aldine721BT-Light /Aldine721BT-LightItalic /Aldine721BT-Roman /Aldus-Italic /Aldus-Roman /AlternateGothicNo2BT-Regular /Anna /AntiqueOlive-Bold /AntiqueOlive-Compact /AntiqueOlive-Italic /AntiqueOlive-Roman /Arcadia /Arcadia-A /Arkona-Medium /Arkona-Regular /AssemblyLightSSK /AvantGarde-Book /AvantGarde-BookOblique /AvantGarde-Demi /AvantGarde-DemiOblique /BakerSignetBT-Roman /BaskervilleBE-Italic /BaskervilleBE-Medium /BaskervilleBE-MediumItalic /BaskervilleBE-Regular /BaskervilleBook-Italic /BaskervilleBook-MedItalic /BaskervilleBook-Medium /BaskervilleBook-Regular /BaskervilleBT-Bold /BaskervilleBT-BoldItalic /BaskervilleBT-Italic /BaskervilleBT-Roman /BaskervilleMT /BaskervilleMT-Bold /BaskervilleMT-BoldItalic /BaskervilleMT-Italic /BaskervilleMT-SemiBold /BaskervilleMT-SemiBoldItalic /BaskervilleNo2BT-Bold /BaskervilleNo2BT-BoldItalic /BaskervilleNo2BT-Italic /BaskervilleNo2BT-Roman /Bauhaus-Bold /Bauhaus-Demi /Bauhaus-Heavy /BauhausITCbyBT-Bold /BauhausITCbyBT-Medium /Bauhaus-Light /Bauhaus-Medium /BellCentennial-Address /BellGothic-Black /BellGothic-Bold /Bell-GothicBoldItalicBT /BellGothicBT-Bold /BellGothicBT-Roman /BellGothic-Light /Bembo /Bembo-Bold /Bembo-BoldExpert /Bembo-BoldItalic /Bembo-BoldItalicExpert /Bembo-Expert /Bembo-ExtraBoldItalic /Bembo-Italic /Bembo-ItalicExpert /Bembo-Semibold /Bembo-SemiboldItalic /Berkeley-Black /Berkeley-BlackItalic /Berkeley-Bold /Berkeley-BoldItalic /Berkeley-Book /Berkeley-BookItalic /Berkeley-Italic /Berkeley-Medium /Berling-Bold /Berling-BoldItalic /Berling-Italic /Berling-Roman /BernhardModernBT-Bold /BernhardModernBT-BoldItalic /BernhardModernBT-Italic /BernhardModernBT-Roman /Bodoni /Bodoni-Bold /Bodoni-BoldItalic /Bodoni-Italic /Bodoni-Poster /Bodoni-PosterCompressed /Bookman-Demi /Bookman-DemiItalic /Bookman-Light /Bookman-LightItalic /Boton-Italic /Boton-Medium /Boton-MediumItalic /Boton-Regular /Boulevard /BremenBT-Black /BremenBT-Bold /CaflischScript-Bold /CaflischScript-Regular /Carta /Caslon224ITCbyBT-Bold /Caslon224ITCbyBT-BoldItalic /Caslon224ITCbyBT-Book /Caslon224ITCbyBT-BookItalic /Caslon540BT-Italic /Caslon540BT-Roman /CaslonBT-Bold /CaslonBT-BoldItalic /CaslonTwoTwentyFour-Black /CaslonTwoTwentyFour-BlackIt /CaslonTwoTwentyFour-Bold /CaslonTwoTwentyFour-BoldIt /CaslonTwoTwentyFour-Book /CaslonTwoTwentyFour-BookIt /CaslonTwoTwentyFour-Medium /CaslonTwoTwentyFour-MediumIt /CastleT-Bold /CastleT-Book /Caxton-Bold /Caxton-BoldItalic /Caxton-Book /Caxton-BookItalic /Caxton-Light /Caxton-LightItalic /CelestiaAntiqua-Ornaments /Centennial-BlackItalicOsF /Centennial-BlackOsF /Centennial-BoldItalicOsF /Centennial-BoldOsF /Centennial-ItalicOsF /Centennial-LightItalicOsF /Centennial-LightSC /Centennial-RomanSC /CenturyOldStyle-Bold /CenturyOldStyle-Italic /CenturyOldStyle-Regular /CheltenhamBT-Bold /CheltenhamBT-BoldItalic /CheltenhamBT-Italic /CheltenhamBT-Roman /Christiana-Bold /Christiana-BoldItalic /Christiana-Italic /Christiana-Medium /Christiana-MediumItalic /Christiana-Regular /Christiana-RegularExpert /Christiana-RegularSC /Clarendon /Clarendon-Bold /Clarendon-Light /ClassicalGaramondBT-Bold /ClassicalGaramondBT-BoldItalic /ClassicalGaramondBT-Italic /ClassicalGaramondBT-Roman /CMTI10 /CommonBullets /ConduitITC-Bold /ConduitITC-BoldItalic /ConduitITC-Light /ConduitITC-LightItalic /ConduitITC-Medium /ConduitITC-MediumItalic /CooperBlack /CooperBlack-Italic /CopperplateGothicBT-Bold /CopperplateGothicBT-BoldCond /CopperplateGothicBT-Heavy /CopperplateGothicBT-Roman /CopperplateGothicBT-RomanCond /Copperplate-ThirtyThreeBC /Copperplate-ThirtyTwoBC /Coronet-Regular /Courier /Courier-Bold /Courier-BoldOblique /Courier-Oblique /Critter /CS-Special-font /DextorD /DextorOutD /DidotLH-OrnamentsOne /DidotLH-OrnamentsTwo /DINEngschrift /DINEngschrift-Alternate /DINMittelschrift /DINMittelschrift-Alternate /DINNeuzeitGrotesk-BoldCond /DINNeuzeitGrotesk-Light /Dom-CasItalic /Dom-CasualBT /Ehrhard-Italic /Ehrhard-Regular /EhrhardSemi-Italic /EhrhardtMT /EhrhardtMT-Italic /EhrhardtMT-SemiBold /EhrhardtMT-SemiBoldItalic /EhrharSemi /ElectraLH-Bold /ElectraLH-BoldCursive /ElectraLH-Cursive /ElectraLH-Regular /EnglischeSchT-Bold /EnglischeSchT-Regu /ErasContour /ErasITCbyBT-Bold /ErasITCbyBT-Book /ErasITCbyBT-Demi /ErasITCbyBT-Light /ErasITCbyBT-Medium /ErasITCbyBT-Ultra /EUEX10 /EUFB10 /EUFB5 /EUFB7 /EUFM10 /EUFM5 /EUFM7 /EURB10 /EURB5 /EURB7 /EURM10 /EURM5 /EURM7 /EuropeanPi-Four /EuropeanPi-One /EuropeanPi-Three /EuropeanPi-Two /Eurostile /Eurostile-Bold /Eurostile-BoldExtendedTwo /Eurostile-ExtendedTwo /EUSB10 /EUSB5 /EUSB7 /EUSM10 /EUSM5 /EUSM7 /ExPonto-Regular /Fenice-Bold /Fenice-BoldOblique /FeniceITCbyBT-Bold /FeniceITCbyBT-BoldItalic /FeniceITCbyBT-Regular /FeniceITCbyBT-RegularItalic /Fenice-Light /Fenice-LightOblique /Fenice-Regular /Fenice-RegularOblique /Fenice-Ultra /Fenice-UltraOblique /FlashD-Ligh /Folio-Bold /Folio-BoldCondensed /Folio-ExtraBold /Folio-Light /Folio-Medium /FontanaNDEeOsF /FontanaNDEeOsF-Semibold /FormalScript421BT-Regular /Formata-Bold /Formata-MediumCondensed /FournierMT-Ornaments /FrakturBT-Regular /FranklinGothic-Book /FranklinGothic-BookItal /FranklinGothic-BookOblique /FranklinGothic-Condensed /FranklinGothic-Demi /FranklinGothic-DemiItal /FranklinGothic-DemiOblique /FranklinGothic-Heavy /FranklinGothic-HeavyItal /FranklinGothic-HeavyOblique /FranklinGothic-Medium /FranklinGothic-MediumItal /FranklinGothic-Roman /FrizQuadrataITCbyBT-Bold /FrizQuadrataITCbyBT-Roman /Frutiger-Black /Frutiger-BlackCn /Frutiger-BlackItalic /Frutiger-Bold /Frutiger-BoldCn /Frutiger-BoldItalic /Frutiger-Cn /Frutiger-ExtraBlackCn /Frutiger-Italic /Frutiger-Light /Frutiger-LightCn /Frutiger-LightItalic /Frutiger-Roman /Frutiger-UltraBlack /Futura /FuturaBlackBT-Regular /Futura-Bold /Futura-BoldOblique /Futura-Book /Futura-BookOblique /FuturaBT-Bold /FuturaBT-BoldCondensed /FuturaBT-BoldCondensedItalic /FuturaBT-BoldItalic /FuturaBT-Book /FuturaBT-BookItalic /FuturaBT-ExtraBlack /FuturaBT-ExtraBlackCondensed /FuturaBT-ExtraBlackCondItalic /FuturaBT-ExtraBlackItalic /FuturaBT-Heavy /FuturaBT-HeavyItalic /FuturaBT-Light /FuturaBT-LightCondensed /FuturaBT-LightItalic /FuturaBT-Medium /FuturaBT-MediumCondensed /FuturaBT-MediumItalic /Futura-ExtraBold /Futura-ExtraBoldOblique /Futura-Heavy /Futura-HeavyOblique /Futura-Light /Futura-LightOblique /Futura-Oblique /GalliardITCbyBT-Italic /GalliardITCbyBT-Roman /Garamond-Antiqua /Garamond-BoldCondensed /Garamond-BoldCondensedItalic /Garamond-BookCondensed /Garamond-BookCondensedItalic /Garamond-Halbfett /GaramondITCbyBT-Bold /GaramondITCbyBT-BoldCondensed /GaramondITCbyBT-BoldCondItalic /GaramondITCbyBT-BoldItalic /GaramondITCbyBT-BoldNarrow /GaramondITCbyBT-BoldNarrowItal /GaramondITCbyBT-Book /GaramondITCbyBT-BookCondensed /GaramondITCbyBT-BookCondItalic /GaramondITCbyBT-BookItalic /GaramondITCbyBT-Light /GaramondITCbyBT-LightCondensed /GaramondITCbyBT-LightCondItalic /GaramondITCbyBT-LightItalic /GaramondITCbyBT-LightNarrow /GaramondITCbyBT-LightNarrowItal /GaramondITCbyBT-Ultra /GaramondITCbyBT-UltraCondensed /GaramondITCbyBT-UltraCondItalic /GaramondITCbyBT-UltraItalic /Garamond-Kursiv /Garamond-KursivHalbfett /Garamond-LightCondensed /Garamond-LightCondensedItalic /GaramondThree /GaramondThree-Bold /GaramondThree-BoldItalic /GaramondThree-Italic /GaramondThreeSMSspl /GaramondThreespl /GaramondThreeSpl-Bold /GaramondThreeSpl-Italic /GarthGraphic /GarthGraphic-Black /GarthGraphic-Bold /GarthGraphic-BoldCondensed /GarthGraphic-BoldItalic /GarthGraphic-Condensed /GarthGraphic-ExtraBold /GarthGraphic-Italic /Geometric231BT-HeavyC /GeometricSlab712BT-BoldA /GeometricSlab712BT-ExtraBoldA /GeometricSlab712BT-LightA /GeometricSlab712BT-LightItalicA /GeometricSlab712BT-MediumA /GeometricSlab712BT-MediumItalA /Giddyup /Giddyup-Thangs /GillSans /GillSans-Bold /GillSans-BoldCondensed /GillSans-BoldItalic /GillSans-Condensed /GillSans-ExtraBold /GillSans-Italic /GillSans-Light /GillSans-LightItalic /GillSans-UltraBold /GillSans-UltraBoldCondensed /Gill-Special /Giovanni-Bold /Giovanni-BoldItalic /Giovanni-Book /Giovanni-BookItalic /Glypha /Glypha-Bold /Glypha-BoldOblique /Glypha-Oblique /Goudy /Goudy-Bold /Goudy-BoldItalic /Goudy-ExtraBold /Goudy-Italic /GoudyOldStyleBT-Bold /GoudyOldStyleBT-BoldItalic /GoudyOldStyleBT-ExtraBold /GoudyOldStyleBT-Italic /GoudyOldStyleBT-Roman /GoudySans-Bold /GoudySans-BoldItalic /GoudySansITCbyBT-Bold /GoudySansITCbyBT-BoldItalic /GoudySansITCbyBT-Medium /GoudySansITCbyBT-MediumItalic /GoudySans-Medium /GoudySans-MediumItalic /Granjon /Granjon-Bold /Granjon-BoldOsF /Granjon-Italic /Granjon-ItalicOsF /Granjon-SC /GreymantleMVB-Ornaments /Helvetica /Helvetica-Black /Helvetica-BlackOblique /Helvetica-Black-SemiBold /Helvetica-Bold /Helvetica-BoldOblique /Helvetica-Condensed /Helvetica-Condensed-Black /Helvetica-Condensed-BlackObl /Helvetica-Condensed-Bold /Helvetica-Condensed-BoldObl /Helvetica-Condensed-Light /Helvetica-Condensed-LightObl /Helvetica-Condensed-Oblique /Helvetica-Light /Helvetica-LightOblique /Helvetica-Narrow /Helvetica-Narrow-Bold /Helvetica-Narrow-BoldOblique /Helvetica-Narrow-Oblique /HelveticaNeue-BlackCond /HelveticaNeue-BlackCondObl /HelveticaNeue-Bold /HelveticaNeue-BoldCond /HelveticaNeue-BoldCondObl /HelveticaNeue-BoldExt /HelveticaNeue-BoldExtObl /HelveticaNeue-BoldItalic /HelveticaNeue-Condensed /HelveticaNeue-CondensedObl /HelveticaNeue-ExtBlackCond /HelveticaNeue-ExtBlackCondObl /HelveticaNeue-Extended /HelveticaNeue-ExtendedObl /HelveticaNeue-Heavy /HelveticaNeue-HeavyCond /HelveticaNeue-HeavyCondObl /HelveticaNeue-HeavyExt /HelveticaNeue-HeavyExtObl /HelveticaNeue-HeavyItalic /HelveticaNeue-Italic /HelveticaNeue-Light /HelveticaNeue-LightCond /HelveticaNeue-LightCondObl /HelveticaNeue-LightItalic /HelveticaNeueLTStd-Md /HelveticaNeueLTStd-MdIt /HelveticaNeue-Medium /HelveticaNeue-MediumCond /HelveticaNeue-MediumCondObl /HelveticaNeue-MediumExt /HelveticaNeue-MediumExtObl /HelveticaNeue-MediumItalic /HelveticaNeue-Roman /HelveticaNeue-ThinCond /HelveticaNeue-ThinCondObl /HelveticaNeue-UltraLigCond /HelveticaNeue-UltraLigCondObl /Helvetica-Oblique /HelvLight /Humanist521BT-Bold /Humanist521BT-BoldCondensed /Humanist521BT-BoldItalic /Humanist521BT-ExtraBold /Humanist521BT-Italic /Humanist521BT-Light /Humanist521BT-LightItalic /Humanist521BT-Roman /Humanist521BT-RomanCondensed /Humanist521BT-UltraBold /Humanist521BT-XtraBoldCondensed /Humanist777BT-BlackB /Humanist777BT-BlackItalicB /Humanist777BT-BoldB /Humanist777BT-BoldItalicB /Humanist777BT-ItalicB /Humanist777BT-LightB /Humanist777BT-LightItalicB /Humanist777BT-RomanB /ICMEX10 /ICMMI8 /ICMSY8 /ICMTT8 /ILASY8 /ILCMSS8 /ILCMSSB8 /ILCMSSI8 /Imago-Book /Imago-BookItalic /Imago-ExtraBold /Imago-ExtraBoldItalic /Imago-Medium /Imago-MediumItalic /Industria-Inline /Industria-InlineA /Industria-Solid /Industria-SolidA /Insignia /Insignia-A /IPAExtras /IPAHighLow /IPAKiel /IPAKielSeven /IPAsans /JoannaMT /JoannaMT-Bold /JoannaMT-BoldItalic /JoannaMT-Italic /KlangMT /Kuenstler480BT-Black /Kuenstler480BT-Bold /Kuenstler480BT-BoldItalic /Kuenstler480BT-Italic /Kuenstler480BT-Roman /KunstlerschreibschD-Bold /KunstlerschreibschD-Medi /Lapidary333BT-Black /Lapidary333BT-Bold /Lapidary333BT-BoldItalic /Lapidary333BT-Italic /Lapidary333BT-Roman /LASY10 /LASY5 /LASY6 /LASY7 /LASY8 /LASY9 /LASYB10 /LatinMT-Condensed /LCIRCLE10 /LCIRCLEW10 /LCMSS8 /LCMSSB8 /LCMSSI8 /LDecorationPi-One /LDecorationPi-Two /Leawood-Black /Leawood-BlackItalic /Leawood-Bold /Leawood-BoldItalic /Leawood-Book /Leawood-BookItalic /Leawood-Medium /Leawood-MediumItalic /LegacySans-Bold /LegacySans-BoldItalic /LegacySans-Book /LegacySans-BookItalic /LegacySans-Medium /LegacySans-MediumItalic /LegacySans-Ultra /LegacySerif-Bold /LegacySerif-BoldItalic /LegacySerif-Book /LegacySerif-BookItalic /LegacySerif-Medium /LegacySerif-MediumItalic /LegacySerif-Ultra /LetterGothic /LetterGothic-Bold /LetterGothic-BoldSlanted /LetterGothic-Slanted /Life-Bold /Life-Italic /Life-Roman /LINE10 /LINEW10 /Lithos-Black /Lithos-Regular /LOGO10 /LOGO8 /LOGO9 /LOGOBF10 /LOGOSL10 /LOMD-Normal /LubalinGraph-Book /LubalinGraph-BookOblique /LubalinGraph-Demi /LubalinGraph-DemiOblique /LucidaMath-Symbol /LydianBT-Bold /LydianBT-BoldItalic /LydianBT-Italic /LydianBT-Roman /LydianCursiveBT-Regular /Marigold /MathematicalPi-Five /MathematicalPi-Four /MathematicalPi-One /MathematicalPi-Six /MathematicalPi-Three /MathematicalPi-Two /Melior /Melior-Bold /Melior-BoldItalic /Melior-Italic /MercuriusCT-Black /MercuriusCT-BlackItalic /MercuriusCT-Light /MercuriusCT-LightItalic /MercuriusCT-Medium /MercuriusCT-MediumItalic /MercuriusMT-BoldScript /Meridien-Medium /Meridien-MediumItalic /Meridien-Roman /Minion-Black /Minion-Bold /Minion-BoldCondensed /Minion-BoldCondensedItalic /Minion-BoldItalic /Minion-Condensed /Minion-CondensedItalic /MinionExp-Italic /MinionExp-Semibold /MinionExp-SemiboldItalic /Minion-Italic /Minion-Ornaments /Minion-Regular /Minion-Semibold /Minion-SemiboldItalic /MonaLisa-Recut /MSAM10 /MSAM10A /MSAM5 /MSAM6 /MSAM7 /MSAM8 /MSAM9 /MSBM10 /MSBM10A /MSBM5 /MSBM6 /MSBM7 /MSBM8 /MSBM9 /MTEX /MTEXB /MTEXH /MTGU /MTGUB /MTMI /MTMIB /MTMIH /MTMS /MTMSB /MTMUB /MTMUH /MTSY /MTSYB /MTSYH /MTSYN /MusicalSymbols-Normal /Myriad-Bold /Myriad-BoldItalic /Myriad-CnBold /Myriad-CnBoldItalic /Myriad-CnItalic /Myriad-CnSemibold /Myriad-CnSemiboldItalic /Myriad-Condensed /Myriad-Italic /Myriad-Roman /Myriad-Sketch /Myriad-Tilt /NeuzeitS-Book ] /NeverEmbed [ true ] /AntiAliasColorImages false /CropColorImages true /ColorImageMinResolution 150 /ColorImageMinResolutionPolicy /OK /DownsampleColorImages true /ColorImageDownsampleType /Bicubic /ColorImageResolution 300 /ColorImageDepth -1 /ColorImageMinDownsampleDepth 1 /ColorImageDownsampleThreshold 1.50000 /EncodeColorImages true /ColorImageFilter /DCTEncode /AutoFilterColorImages true /ColorImageAutoFilterStrategy /JPEG /ColorACSImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /ColorImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /JPEG2000ColorACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /JPEG2000ColorImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /AntiAliasGrayImages false /CropGrayImages true /GrayImageMinResolution 150 /GrayImageMinResolutionPolicy /OK /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 300 /GrayImageDepth -1 /GrayImageMinDownsampleDepth 2 /GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /GrayImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /JPEG2000GrayACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /JPEG2000GrayImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /AntiAliasMonoImages false /CropMonoImages true /MonoImageMinResolution 1200 /MonoImageMinResolutionPolicy /OK /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 1200 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict << /K -1 >> /AllowPSXObjects false /CheckCompliance [ /None ] /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox false /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile (U.S. Web Coated \050SWOP\051 v2) /PDFXOutputConditionIdentifier () /PDFXOutputCondition () /PDFXRegistryName (http://www.color.org) /PDFXTrapped /Unknown /CreateJDFFile false /SyntheticBoldness 1.000000 /Description << /FRA <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> /JPN <FEFF3053306e8a2d5b9a306f30019ad889e350cf5ea6753b50cf3092542b308000200050004400460020658766f830924f5c62103059308b3068304d306b4f7f75283057307e30593002537052376642306e753b8cea3092670059279650306b4fdd306430533068304c3067304d307e305930023053306e8a2d5b9a30674f5c62103057305f00200050004400460020658766f8306f0020004100630072006f0062006100740020304a30883073002000520065006100640065007200200035002e003000204ee5964d30678868793a3067304d307e30593002> /DEU <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> /PTB <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> /DAN <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> /NLD <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> /ESP <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> /SUO <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> /ITA <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> /NOR <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> /SVE <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> /ENU <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> >> >> setdistillerparams << /HWResolution [2400 2400] /PageSize [612.000 792.000] >> setpagedevice
memo3/Elvik_Laws_of_accident_causation.pdf
Accident Analysis and Prevention 38 (2006) 742–747
Laws of accident causation
Rune Elvik ∗ Institute of Transport Economics, P.O. Box 6110, Etterstad, N-0602 Oslo, Norway
Received 30 October 2005; received in revised form 8 January 2006; accepted 16 January 2006
Abstract
This paper suggests that the influence of a number of important risk factors on road accidents can be described in terms of a few highly general statistical regularities that determine the shape of the relationship between the risk factors and accident occurrence. The statistical regularities are referred to as “laws of accident causation”. The following “laws” are proposed:
1. The universal law of learning, which states that the ability to detect and control traffic hazards increases uniformly as the amount of travel increases. This law implies that accident rate per unit of exposure will decline as the amount of exposure increases.
2. The law of rare events, which states that the more rarely a certain risk factor is encountered the larger is its effect on accident rate. This law
3
4
m c ©
K
1
s o c c p S 1 ( t n
0 d
implies that a risk factor encountered on, for example 5% of all trips, will be associated with a greater increase in accident rate than an otherwise identical risk factor encountered on 50% of all trips.
. The law of complexity, which states that the more units of information per unit of time a road user must attend to, the higher becomes the probability that an error will made. This law implies that accident rate will increase the more elements of information the traffic environment contains.
. The law of cognitive capacity, which states that the more cognitive capacity approaches its limits, the higher the accident rate. This law implies that impairments affecting mental functions will have a greater effect on accident rate than impairments affecting physical functioning only.
Instances of all these laws, as well as a discussion of to how the laws can be tested empirically, are given. It is hoped that proposing a few basic echanisms that can summarise the impact of a number of risk factors will stimulate research that may lead to a more general theory of accident
ausation. 2006 Elsevier Ltd. All rights reserved.
eywords: Accident; Causation; Scientific law; Explanation; Empirical testing
. Introduction
Attempts at explaining accidents are as old as the scientific tudy of accidents itself. Bortkiewicz (1898), whose work is ften regarded as the start of modern accident research, con- luded that accidents occurred at random and were thus inexpli- able. Later contributions have attributed accidents to individual roneness (Shaw and Sichel, 1971), human errors (Sabey and taughton, 1975; Treat et al., 1979), system failures (Perrow, 999), or a desire in road users to seek a target level of risk Wilde, 1982). This paper will not try to review these contribu- ions to the theory of accident causation, nor try to develop a ew theory. Its objective is much more limited.
∗ Tel.: +47 22 57 3800; fax: +47 22 57 0290. E-mail address: [email protected].
Empirical research has identified a very large number of risk factors that are statistically associated with road accident occur- rence, i.e. factors whose presence increases the probability of accidents. In principle, one might try to “explain” road acci- dents by listing these factors, perhaps adding information on their relative importance. This would at best be only the begin- ning of a theory of accident causation. A list of risk factors can be informative and useful, but it begs more basic questions, like: Why is factor X a risk factor for accidents? Why does risk factor Y appear to be more important in explaining accidents than risk factor X? What we need is, in other words, an account of mech- anisms that explain why a certain factor becomes a risk factor. This paper will propose a few such mechanisms, stated in terms of general (i.e. not restricted to any particular country or road traffic system) hypotheses that are empirically testable. The main question this paper seeks to answer is: Can the effects of risk fac- tors on the probability of road accidents be explained in terms of
001-4575/$ – see front matter © 2006 Elsevier Ltd. All rights reserved. oi:10.1016/j.aap.2006.01.005
R. Elvik / Accident Analysis and Prevention 38 (2006) 742–747 743
a few underlying mechanisms that generate the statistical rela- tionships between risk factors and accidents and determine the shape of these relationships?
2. Mechanisms underlying risk factors
The number of risk factors that influences accidents is vast. Nobody can enumerate all these risk factors, yet their effects on accidents may display striking regularities. The ability of a road user to recognise risk factors and prevent them from leading to accidents is likely to be strongly influenced by the experi- ences made when using the transport system. This suggests the following law:
The universal law of learning: The ability to detect and con- trol traffic hazards improves continuously as a result of exposure to these hazards.
Traffic hazards denote any potential risk factor. The term refers, broadly speaking, to anything that can go wrong. Expo- sure refers to the amount of travel, the number of kilometres covered per month, year or during lifetime. The main implica- tion of the universal law of learning is that road user accident rate per kilometre travelled tends to decline as the number of kilome- tres travelled increases. This tendency is likely to be most clearly evident among novice drivers, but it is suggested that it applies to all road users throughout life. Learning never ends, because things once learnt may be forgotten and need to be re-learnt later o a
a T d t f
i
e t d w w a r
o a a s l n t
o h
c
difficult manoeuvres (like turning left across several lanes of opposing traffic) or frequent changes in the cognitive demands placed on road users. Complexity can partly be controlled by road users, by regulating speed or by concentrating more strictly on the driving task. Inherent variability in complexity is, however, likely to exceed the possibilities for compensatory behaviour. Complexity is thus a characteristic of the traffic envi- ronment, external to the road user.
The ability to compensate for inherent variations in task demands depends to a major extent on the mental state of the road user. The more the ability to concentrate attention, to make decisions in a short time, and to carry out appropriate action is reduced, the higher become the chances of accident involve- ment. In short, the more reduced our cognitive capacity gets, the less able we are to detect and control traffic hazards. Thus, the law of cognitive capacity states:
The law of cognitive capacity: The more cognitive capac- ity approaches its limits, the greater the increase in the rate of accidents.
This implies that impairments affecting mental functions have a greater effect on accident rate than physical impairments, not affecting mental functions. Moreover, the more strongly affected mental functions are, the greater the effect on accident rate.
It is proposed that these four mechanisms describe in general terms the association between a number of risk factors and acci- d t a b a t
t o a I a p t s t o s t t t
3
3
t p F d
n. The law is a hypothesis only; if it is correct one would expect ccident rate to decline uniformly as travel exposure increases.
Exposure to traffic hazards is not uniform. Some traffic haz- rds are encountered more irregularly or more rarely than others. he more rarely a certain traffic hazard is encountered, the more ifficult it is to predict its occurrence, and the less are the oppor- unities for learning how to control the hazard. This implies the ollowing regularity:
The law of rare events: The more rarely a certain traffic hazard s encountered the greater is its effect on accident rate.
Thus, a traffic hazard encountered on 5% of all trips is xpected to be associated with a greater increase in accident rate han a traffic hazard encountered on 50% of all trips. This ten- ency is likely to apply to environmental hazards (rainfall, snow, ild animals) as well as to highly surprising features of road- ay design (unexpected sharp curves, for example). It may even
pply to the relative proportions made up by different groups of oad users in mixed traffic.
Traffic hazards may be perfectly controllable if they appear ne at a time. Modern traffic systems are, however, complex nd sometimes present several traffic hazards at the same time, ll of which the road user ought ideally to attend to. If a traffic ituation becomes very complex, our cognitive capacity may no onger be able to keep up with the task demands, leading to the eglect of certain traffic hazards or an inadequate response to hem. This suggests the following law of complexity:
The law of complexity: The more potentially relevant items f information a road user must attend to per unit of time the igher the probability of accidents.
A complex traffic environment is one that, for example, is haracterised by dense and mixed traffic, the necessity to make
ents. In the following, the four mechanisms will be referred o as “laws of accident causation”. The four “laws” are meant s empirically testable hypotheses; thus, to the extent they are ased on theoretical notions, these need to be translated to oper- tional terms. Some preliminary suggestions about how to do his will be discussed later in the paper.
It is by no means suggested that the four “laws” proposed in his paper are the only mechanisms that may explain the effects f risk factors on accidents. Nor is it suggested that the “laws” re true scientific laws, in the sense of that term in epistemology. n epistemology, a distinction is made between scientific laws nd accidental generalisations. The latter category consists of henomena that have been observed a great many times, but hat do not represent invariant relationships the same way a true cientific law does. Thus, fresh water will always freeze when the emperature stays below 0 ◦C long enough; but no finite number f observations justifies the claim that “All swans are white” is a cientific law. The “laws” proposed in this paper are definitely of he “white swans” variety: they are to be regarded as hypotheses hat must be given up when most observations no longer support hem.
. Instances of the laws of accident causation
.1. The universal law of learning
Does the rate of accident involvement vary inversely with he amount of exposure? One data set showing this tendency is resented in Fig. 1, based on a study reported by Sagberg (1998). ig. 1 shows self reported accident rates for novice drivers (i.e. rivers who have held a licence for less than 18 months) in
744 R. Elvik / Accident Analysis and Prevention 38 (2006) 742–747
Fig. 1. Accident rates among novice drivers in Norway as a function of monthly driving distance. Based on Sagberg (1998).
Norway, depending on their monthly driving distance. There is a remarkable drop in accident rates as the monthly number of kilometres driven increases. This drop is seen both in men and in women. For all monthly driving distances greater than about 250 km, women have a lower accident rate than men. Their mean accident rate is, however, slightly higher than the mean accident rate for men, due entirely to their lower mean monthly driving distance.
Very similar relationships were found by Forsyth et al. (1995). One might think that the relationship shown in Fig. 1 does not apply to experienced drivers. Most studies find that driver acci- dent rates are fairly stable between the ages of 25 and 65 years, suggesting that the process of learning has been completed by the age of 25 years, and that a decrease in performance sets in at about the age of 65 years. By contrast, Fig. 2 shows the results of a study reported by Hakamies-Blomqvist et al. (2002), in which accident rates for drivers aged 26–40 years were compared to accident rates for drivers aged 65 years and above.
Accident rates decline remarkably as annual driving distance increases. This applies to both groups. The shape of the rela- tionship between annual driving distance and accident rate is virtually identical for middle-aged drivers and older drivers. Driving long annual distances is apparently associated with more success in avoiding accidents irrespective of age.
The fact that accident rates drop as annual driving dis- tance increases is consistent with the universal law of learning,
F f
but alternative explanations are clearly possible. High-mileage drivers may do a higher proportion of their driving on compar- atively safe roads, like motorways, than low-mileage drivers. They may also invest in safer cars, due to the greater need for driving. The high accident rate among low-mileage older drivers may in part be endogenous: drivers who justifiably feel unsafe (for example as a result of past accidents or near-misses) may restrict their driving for that reason.
To really test the universal law of learning, these alternative interpretations must be controlled for. The studies quoted above did not fully control for the confounding factors, yet the rela- tionships they found were very strong and ought to spur further research.
3.2. The law of rare events
Accidents tend to come as a surprise. They are never expected to happen. Accidents are, however, more likely to happen when something unexpected or unusual occurs, than when such events do not take place. This is the essence of the law of rare events. Risk factors experienced rarely or infrequently provide fewer opportunities for learning than risk factors encountered more often. Fig. 3 shows an instance of this tendency. It shows the relationship between the proportion of wintertime driving per- formed on snow- or ice-covered roads and the relative accident r i t s r
m t o m s n i c T
F f B
ig. 2. Accident rates among middle-aged and older drivers in Finland as a unction of annual driving distance. Based on Hakamies-Blomqvist et al. (2002).
ate on this type of road surface compared to a bare road. Fig. 3 s based on a study by Brüde and Larsson (1980). It is seen that he relative accident rate on snow- or ice-covered roads increases harply as the proportion of driving done on snow- or ice-covered oads is reduced.
The rarity of snow or ice on the roads of Southern Sweden eans that drivers in this part of the country rarely get the oppor-
unity to develop skills for safe driving on snow or ice. Moreover, n the few occasions there is snow or ice, it is likely to come ore as a surprise than in Northern Sweden, where snow or ice
tays on the road surface during the whole winter. It should be oted, however, that even drivers who are exposed to snow or ce for more than 50% of their wintertime driving do not fully ompensate for the reduced friction associated with snow or ice. he accident rate on snow or ice remains about twice as high
ig. 3. Relative accident rate (bare road surface = 1.0) on roads in Sweden as a unction of the proportion of driving on roads covered by snow or ice. Based on rüde and Larsson (1980).
R. Elvik / Accident Analysis and Prevention 38 (2006) 742–747 745
Fig. 4. Accident rate in curves in New Zealand as a function of the length of the straight section preceding the curve and curve radius. Adapted from Matthews and Barnes (1988).
as the accident rate on bare roads even when more than 50% of exposure is subject to snow or ice.
Another example of the law of rare events is given in Fig. 4. It shows the relationship between the length of a straight road section ahead of a curve and the accident rate in curves, depend- ing on the sharpness of the curve (Matthews and Barnes, 1988). The accident rate in sharp curves is seen to increase markedly as these curves become less frequent. A similar relationship does not seem to apply to gentle curves.
The explanation for these relationships is likely to be related to driver expectations and the attendant behavioural adaptation. Drivers have no trouble in safely negotiating sharp curves if they expect there to be a lot of them along a road, as is evidenced in Fig. 4 by the low accident rates in sharp curves that have no straight section of road in-between them. If, however, a driver has become accustomed to driving on a straight road, the appearance of a sharp curve violates expectations, and behaviour may be more poorly adapted to task demands than in the case of a long winding road.
3.3. The law of complexity
Complexity is both an inherent characteristic of the traffic system and an outcome of driver choices. A driver who adopts small safety margins makes the task of driving more complex t
o y p b g a o d s t
c 5
Fig. 5. Injury accident rate in junctions in Norway as a function of the type of traffic control, number of legs, and proportion of traffic entering from the minor road. Based on Tran (1999) and Sakshaug and Johannessen (2005).
portion of vehicles entering from the minor road. Complexity is greater in yield controlled junctions than in roundabouts; it is greater in four leg junctions than in three leg junctions; and it is greater the more traffic there is from the minor road. Broadly speaking, the effects of complexity are as expected; the more complex a junction, the higher its accident rate.
Another example of the effects of complexity is shown in Fig. 6. Fig. 6 shows the relationship between the number of driveways per kilometre of road entering national roads in Nor- way and the injury accident rate (Muskaug, 1985). The data are old, but the relationship is not likely to have changed very much.
The number of driveways per kilometre gives a fairly good indication of the level of roadside development. The more drive- ways there are, the more development there will be along the road, in terms of shops, residential areas or industry.
3.4. The law of cognitive capacity
In normal driving, most drivers will have ample spare cogni- tive capacity; i.e. cognitive capacity they can spend on other things, like talking to passengers, listening to the radio, or, increasingly, having a conversation on the mobile phone. Most of the time, devoting spare capacity to such tasks does not inter- fere with the task of driving. This does not mean, however, that
F n (
han one who allows a greater margin for errors. An example of inherent complexity is the design and control
f junctions. Roundabouts have become very popular in recent ears, and for good reasons. A roundabout reduces the com- lexity of a junction. The theoretical number of conflict points etween the traffic movements passing through a junction is reatly reduced, in particular in four leg junctions. Moreover, ll traffic inside the roundabout comes from the same direction; ne no longer has to keep track of traffic coming from several irections at the same time. Fig. 5, based on recent Norwegian tudies (Tran, 1999; Sakshaug and Johannessen, 2005), shows he effects of complexity in junctions on accident rates.
Accident rates are specified according to the type of traffic ontrol (roundabout versus yield), the number of legs (3, 4 or ), and, for junctions that are controlled by yield signs, the pro-
ig. 6. Injury accident rate on national roads in Norway as a function of the umber of access points (driveways) per kilometre of road. Based on Muskaug 1985).
746 R. Elvik / Accident Analysis and Prevention 38 (2006) 742–747
Fig. 7. Relative accident rate associated with various medical conditions. Derived from Vaa (2003).
reductions in the capacity to perform the driving task, physically or mentally, have no effect on accident rates. The law of cogni- tive capacity states that the more cognitive capacity approaches its limits (there being no spare capacity that can be devoted to the driving task in case of need), the higher becomes the accident rate.
Underlying this hypothesis is the view that safe driving is to a large extent a matter of mental capacity and less a matter of physical capacity. Physically impaired drivers are able to drive almost as safely as physically fit drivers. As long as these drivers remain mentally fit, they will try to compensate for their physical disability as best they can. Hence, studies have found that various physical impairments are not associated with a large increase in accident rate. However, the more affected the functions of the brain, the larger becomes the effects on accident rate. Fig. 7, based on a report by Vaa (2003) shows some examples of this.
It is seen that the increase in accident rate associated with the physical impairments is quite small. Epilepsy, which can lead to losses of consciousness, has a substantial effect on accident rate. The same applies to, for example, Alzheimer’s disease and sleep apnoea. The latter disease reduces the quality of sleep and results in chronic tiredness and an increased likelihood of falling asleep during the day.
There is perhaps no clearer case of the effects of reduced cognitive capacity on accident rate than the effects of alcohol. Fig. 8 shows the relative accident rate associated with various l
F a t
Accident rate is found to increase dramatically as the amount of alcohol in the blood increases. Clearly, once blood alcohol level passes about 0.5 mg/ml (0.05%), drivers are no longer able to compensate for its effects to such an extent as to prevent accident rate from increasing.
4. Discussion and conclusions
Can the existence and effects of the multitude of risk fac- tors that have been found to contribute to road accidents be understood in terms of a few basic mechanisms? This paper sug- gests that the existence and effects of many risk factors can be explained in terms of a few highly general mechanisms, which have been referred to as “laws of accident causation”.
The mechanisms proposed in this paper are intended both to explain the existence of certain risk factors and to describe the shape of their statistical relationship with accident rates. Exam- ples have been given of all the four mechanisms proposed; these examples have deliberately been chosen to show clearly how each mechanism operates and influences accident rates. For the purpose of this paper, this approach is defensible. If, on the other hand, one wants to test whether the mechanisms apply in gen- eral, merely collecting confirmatory instances of each of these mechanisms will not do. On the contrary, every attempt should be made to falsify the proposed “laws of accident causation”.
All these “laws” can, in principle, be falsified. The universal l n t h P s t b i s t c c i
r h l u p o p c a e n p i s c l
evels of blood alcohol content (mg/ml) found in various studies.
ig. 8. Relative accident involvement rate associated with various levels of blood lcohol content (mg/ml). Based on Elvik and Vaa (2004) and the sources quoted herein.
aw of learning would seem to be falsified if accident rate is found ot to decline when exposure increases. On the other hand, if his was to be found, a different mechanism may be operating aving a stronger effect on accident rate than learning per se. erhaps those who drive long distances tend to adopt smaller afety margins, believing that their greater experience allows hem to do. This does not mean that learning does not take place, ut that highly skilled drivers make use of those skills not to ncrease safety, but to make driving more fun by employing the kills more actively. It would then be wrong to conclude that he law of learning was falsified; it would be more correct to onclude that it, like so many other social regularities, requires a eteris paribus clause, whose contents would need to be specified n fairly great detail to test the law.
Likewise, the law of rare events would seem to be falsified if isk factors affecting a small share of exposure are found not to ave a greater effect on accident rate than risk factors affecting a arge share of exposure. Such a finding would suggest that road sers are never taken by surprise, which by any reasonable inter- retation ought to qualify as falsifying the law. To falsify the law f complexity, it is necessary to measure complexity. The exam- les given in this paper did not go into that issue, but referred to haracteristics of the traffic environment that are widely regarded s aspects of complexity. Complexity can be measured both in ngineering terms (number of traffic signs, number of junctions, umber of traffic movements permitted in a junction, etc.) and in sychological terms (usually by measuring success in perform- ng a secondary task; the idea is that a drop in performance of a econdary task shows that traffic is more complex). Measuring omplexity in engineering terms is probably most relevant to the aw as stated in this paper; the psychological measure does not
R. Elvik / Accident Analysis and Prevention 38 (2006) 742–747 747
show complexity as such, but driver adaptation to it in terms of devoting a greater share of mental capacity to the driving task. In principle, complexity would not be a problem if drivers fully compensated for it; the law suggests that this does not actually happen. Hence, if accident rate is found not to increase when complexity increases, the law of complexity is falsified.
Finally, reduction in cognitive capacity is also measurable. It can be measured, for example, by giving tasks that require mental resources for their solution, and noting how long it takes to solve the tasks and if the solutions are correct. If it is found that large reductions in cognitive capacity are not associated with an increased accident rate, the law of cognitive capacity is falsified.
The various mechanisms proposed in this paper are related to each other and may perhaps be further reduced to a smaller number of even more basic mechanisms. The main conclusion to be drawn from the research presented in this paper is that the existence and effects of many important risk factors for road accidents can be accounted for in terms of a small number of mechanisms that generate the risk factors by way of limiting the exercise of rationality in the detection and control of traffic hazards.
References
von Bortkiewicz, L., 1898. Das Gesetz der kleinen Zahlen. B. G. Teubner, Leipzig.
B
E
F
Hakamies-Blomqvist, L., Raitanen, T., O’Neill, D., 2002. Driver ageing does not cause higher accident rates per km. Transport. Res. Part F 5, 271–274.
Matthews, L.R., Barnes, J.W., 1988. Relation between road environment and curve accidents. In: Proceedings of 14th ARRB Conference, Part 4, 105–120, Australian Road Research Board, Vermont South, Victoria, Australia.
Muskaug, R., 1985. Risiko på norske riksveger. En analyse av risikoen for trafikkulykker med personskade på riks- og europaveger utenfor Oslo. avhengig av vegbredde, fartsgrense, trafikkmengde og avkjørselstettet. TØI-rapport. Transportøkonomisk institutt, Oslo.
Perrow, C., 1999. Normal Accidents. Living with High Risk Technologies, second ed. Princeton University Press, Princeton, NJ.
Sabey, B.E., Staughton, G.C., 1975. Interacting roles of road environment, vehicle and road user in accidents. In: Proceedings of the Fifth Inter- national Conference of the International Association for Accident and Traffic Medicine, London.
Sagberg, F., 1998. Month-by-month changes in accident risk among novice drivers. In: Paper presented at the 24th International Congress of Applied Psychology, San Francisco, August 9–14.
Sakshaug, K., Johannessen, S., 2005. Revisjon av håndbok 115, analyse av ulykkessteder: Verdier for normal ulykkesfrekvens ved normal og god standard. Notat datert 3. mai 2005. SINTEF teknologi og samfunn, Trans- portsikkerhet og -informatikk, Trondheim.
Shaw, L., Sichel, H.S., 1971. Accident Proneness. Research in the Occur- rence, Causation and Prevention of Road Accidents. Pergamon Press, Oxford.
Tran, T. Vegtrafikkulykker i rundkjøringer – 1999. En analyse av trafikku- lykker i rundkjøringer bygd før 1995 på Europa- og riksvegnettet. Rapport TTS 2, 1999. Statens vegvesen, Vegdirektoratet, Oslo.
Treat J.R., Tumbas, N.S., McDonald, S.T., Shinar, D., Hume, R.D., Mayer, R.E., Stansifer, R.L., Castellan, N.J., 1979. TRI-level study of the causes
V
W
rüde, U., Larsson, J., 1980. Samband vintertid mellan väderlek-väglag- trafikolyckor. Statistisk bearbetning och analys. VTI-rapport 210. Statens väg- och trafikinstitut (VTI), Linköping.
lvik, R., Vaa, T., 2004. The Handbook of Road Safety Measures. Elsevier Science, Oxford.
orsyth, E., Maycock, G., Sexton, B., 1995. Cohort study of learner and novice drivers. Part 3. Accidents, offences and driving experience in the first three years of driving. Project Report 111. Transport Research Lab- oratory, Crowthorne, Berkshire.
of traffic accidents: final report. Report DOT HS 805-085. US Depart- ment of Transportation, National Highway Traffic Safety Administration, Washington, DC.
aa, T., 2003. Impairment, diseases, age and their relative risks of accident involvement: results from meta-analysis. Report 690. Institute of Transport Economics, Oslo.
ilde, G.J.S., 1982. The theory of risk homeostasis: implications for safety and health. Risk Anal. 2, 209–225.
- Laws of accident causation
- Introduction
- Mechanisms underlying risk factors
- Instances of the laws of accident causation
- The universal law of learning
- The law of rare events
- The law of complexity
- The law of cognitive capacity
- Discussion and conclusions
- References
memo3/FriedmanJAMA_article.pdf
ORIGINAL CONTRIBUTION
Impact of Changes in Transportation and Commuting Behaviors During the 1996 Summer Olympic Games in Atlanta on Air Quality and Childhood Asthma Michael S. Friedman, MD Kenneth E. Powell, MD, MPH Lori Hutwagner, MS LeRoy M. Graham, MD W. Gerald Teague, MD
DESPITE ADVANCES IN ASTHMA
therapy, asthma remains a substantial public health problem. In the United
States, asthma is a leading cause of childhood morbidity, with an esti- mated prevalence of 6.9% in children and youth younger than 18 years.1 Nu- merous studies have documented a rise in the morbidity, mortality, and preva- lence of asthma in different popula- tions.2-8 The cause or causes of this trend remain controversial.9-11
Experimental, laboratory, and epi- demiologic studies in the last several years have linked high concentrations of known air pollutants to respiratory health problems, most notably exacer- bations of asthma.12-23 However, op- portunities to study the health effects of anthropogenic improvements in air quality are rare. One study found a de- crease in particulate pollution and res- piratory hospital admissions associ- ated with the closure of an industrial factory in that community.24 To our knowledge, no study has examined the impact of improved ozone pollution for an extended period of time on asthma exacerbations or other markers of asthma morbidity. Also, the extent to which moderate concentrations of
ozone (ie, daily peak of 50-100 ppb) during various exposure lengths af- fects asthma morbidity remains con- troversial.12-16
Author Affiliations are listed at the end of this article. Corresponding Author and Reprints: Michael S. Fried- man, MD, Air Pollution and Respiratory Health Branch, National Center for Environmental Health, Centers for Disease Control and Prevention, Atlanta, GA 30333 (e-mail: [email protected]).
Context Vehicle exhaust is a major source of ozone and other air pollutants. Al- though high ground-level ozone pollution is associated with transient increases in asthma morbidity, the impact of citywide transportation changes on air quality and childhood asthma has not been studied. The alternative transportation strategy implemented dur- ing the 1996 Summer Olympic Games in Atlanta, Ga, provided such an opportunity.
Objective To describe traffic changes in Atlanta, Ga, during the 1996 Summer Olym- pic Games and concomitant changes in air quality and childhood asthma events.
Design Ecological study comparing the 17 days of the Olympic Games ( July 19– August 4, 1996) to a baseline period consisting of the 4 weeks before and 4 weeks after the Olympic Games.
Setting and Subjects Children aged 1 to 16 years who resided in the 5 central coun- ties of metropolitan Atlanta and whose data were captured in 1 of 4 databases.
Main Outcome Measures Citywide acute care visits and hospitalizations for asthma (asthma events) and nonasthma events, concentrations of major air pollutants, me- teorological variables, and traffic counts.
Results During the Olympic Games, the number of asthma acute care events de- creased 41.6% (4.23 vs 2.47 daily events) in the Georgia Medicaid claims file, 44.1% (1.36 vs 0.76 daily events) in a health maintenance organization database, 11.1% (4.77 vs 4.24 daily events) in 2 pediatric emergency departments, and 19.1% (2.04 vs 1.65 daily hospitalizations) in the Georgia Hospital Discharge Database. The number of non- asthma acute care events in the 4 databases changed –3.1%, +1.3%, −2.1%, and +1.0%, respectively. In multivariate regression analysis, only the reduction in asthma events recorded in the Medicaid database was significant (relative risk, 0.48; 95% con- fidence interval, 0.44-0.86). Peak daily ozone concentrations decreased 27.9%, from 81.3 ppb during the baseline period to 58.6 ppb during the Olympic Games (P,.001). Peak weekday morning traffic counts dropped 22.5% (P,.001). Traffic counts were significantly correlated with that day’s peak ozone concentration (average r=0.36 for all 4 roads examined). Meteorological conditions during the Olympic Games did not differ substantially from the baseline period.
Conclusions Efforts to reduce downtown traffic congestion in Atlanta during the Olympic Games resulted in decreased traffic density, especially during the critical morn- ing period. This was associated with a prolonged reduction in ozone pollution and sig- nificantly lower rates of childhood asthma events. These data provide support for ef- forts to reduce air pollution and improve health via reductions in motor vehicle traffic. JAMA. 2001;285:897-905 www.jama.com
©2001 American Medical Association. All rights reserved. (Reprinted) JAMA, February 21, 2001—Vol 285, No. 7 897
The main sources of ambient air pol- lutants are vehicle exhaust, industry, power generation plants, and back- ground contamination. Compared with emissions from nonvehicle sources, the relative amounts of nitrogen oxides, car- bon monoxide, and small particulate matter emitted from vehicles have in- creased disproportionately due to the dramatic increase in worldwide automo- bile use in the past 30 years.12,15 Many studies have found positive associa- tions between traffic density on street of residence and either asthma events or asthma prevalence.25-29 However, the im- pact of citywide automobile use and traf- fic flow on ambient air pollution and asthma morbidity has not been studied.
The 1996 Summer Olympic Games in Atlanta, Ga, provided a unique op- portunity to study the relationship be- tween automobile traffic, air quality, and asthma morbidity. Preparations for the Olympic Games required a strat- egy for minimizing road traffic conges- tion and ensuring that spectators could reach Olympic events in a reasonable amount of time. Additionally, the more than 1 million visitors to Atlanta (re- sulting in increased regional transpor- tation demands) could have magni- fied the region’s existing air quality violations for ozone pollution that oc- cur each summer. Atlanta’s strategy was similar to that used in Los Angeles, Calif, for the 1984 Summer Olympic Games.30 It included the development and use of an integrated 24-hour-a- day public transportation system, the addition of 1000 buses for park-and- ride services, local business use of al- ternative work hours and telecommut- ing, closure of the downtown sector to private automobile travel, altered down- town delivery schedules, and public warnings of potential traffic and air quality problems.31-33
METHODS Study Design
We used an ecological study design and compared the 17 days of the 1996 Sum- mer Olympic Games (July 19–August 4) to a summertime baseline period de- fined as the 4-week periods before and
after the Olympic Games ( June 21– July 18 and August 5–September 1). We measured the number of asthma acute care events, number of nonasthma acute care events, air pollution, meteorologi- cal conditions, and amount of vehicu- lar traffic. The specific dates for study were determined before any data were analyzed. Four-week baseline periods were used to avoid the spring and fall seasons, which can affect ozone lev- els12,13 and asthma exacerbation rates.34
Medical Definitions and Data Collection Our primary outcome measure was the number of hospitalizations, emergency department visits, and urgent care cen- ter visits for asthma. The study popula- tion included persons aged 1 through 16 years residing in the 5 central counties of metropolitan Atlanta (ie, Fulton, DeKalb, Cobb, Gwinnett, and Clayton). Because of their central location, these densely populated counties were more likely to experience dramatic changes in air quality in response to changes in driv- ing and commuting behaviors (ie, greater access to public transportation as an alternative). Visitors to Atlanta were ex- cluded from the study.
Data regarding asthma acute care events were collected from Georgia’s Medicaid claims file, the patient data- base of a health maintenance organi- zation (HMO), computerized emer- gency department records from 2 of the 3 pediatric hospitals in Atlanta (com- bined to create a single emergency de- partment data source), and the Geor- gia Hospital Discharge Database, which includes hospitalization records from all metropolitan Atlanta hospitals. Medical records at the 1 publicly funded pediatric hospital in Atlanta were not available for review. However, 60% to 66% of children seeking emergency care at that hospital were receiving Medic- aid in 1996 (Robert J. Geller, MD, writ- ten communication, January 17, 2001). Therefore, we believe that the Medic- aid files captured clinical information on most of this population. For the Georgia Hospital Discharge Database, a child admitted to a metropolitan At-
lanta hospital during the study period with a primary diagnosis of asthma (In- ternational Classification of Diseases, Ninth Revision [ICD-9], code 493) met our study definition for an asthma acute care event. For the other 3 data sources, a child seen in an emergency depart- ment or urgent care center with a pri- mary diagnosis of asthma met our study definition for an asthma acute care event, regardless of whether the child was hospitalized.
To determine if the study popula- tion was more, less, or as likely to seek emergency services in general during the Olympic period compared with the periods before and after the Olympic Games, or if the size of the study popu- lation significantly changed during the Olympic Games, we collected and ana- lyzed data on the day-by-day total num- ber of all nonasthma-related acute care events during the study period among Atlanta residents aged 1 through 16 years from the same 4 data sources.
Air Quality Data Collection All data on primary pollutants (ie, par- ticulate matter of 10 µm or smaller [PM10], carbon monoxide, nitrogen di- oxide, and sulfur dioxide) and second- ary pollutants (ozone) were obtained jointly from the Environmental Pro- tection Agency (EPA) and the Envi- ronmental Protection Division of Geor- gia’s Department of Natural Resources. All air quality measurement sites used in the study were located in the 5 cen- tral counties of Atlanta and were state operated and EPA regulated (FIGURE 1).
The ozone concentration chosen to represent each day’s exposure was the average of the peak 1-hour ozone con- centrations from the 3 monitoring sites in the study area. The daily 1-hour peak ozone levels for the same time periods in 1997-1998 were used for compari- son. To see if other areas in the region with similar weather patterns experi- enced similar ozone patterns during the 1996 summer, we obtained daily 1-hour peak ozone concentrations at 3 other Georgia sites (Fannin County, Au- gusta, and Columbus), all 60 to 150 miles from Atlanta.
AIR QUALITY AND CHILDHOOD ASTHMA
898 JAMA, February 21, 2001—Vol 285, No. 7 (Reprinted) ©2001 American Medical Association. All rights reserved.
We used the EPA’s standard method of measuring the 4 major primary pol- lutants.35 Data on PM10 were collected at the 1 site capable of continuous 24- hour monitoring and expressed as the cumulative total for each 24-hour pe- riod. The mean 8-hour running daily peak carbon monoxide level, the 24- hour daily mean sulfur dioxide level, and the 1-hour daily peak nitrogen di- oxide level were collected at 1, 2, and 3 monitoring sites, respectively. The daily levels obtained at the 2 sulfur and 3 nitrogen dioxide sites were then av- eraged. Allergen exposure was deter- mined by total daytime mold counts (the predominant summertime aller- gen in Atlanta) collected weekdays dur- ing the study period at the Atlanta Al- lergy and Asthma Clinic.
Meteorological Data We obtained hourly data for 5 weather- related variables (temperature, wind speed, relative humidity, barometric pressure, and solar radiation) from a state-run weather monitoring site lo- cated east of downtown Atlanta. These 5 variables have a direct or indirect im- pact on the rate of ozone formation and clearance in the lower atmosphere and, to a lesser extent, can affect levels of the primary pollutants.12 For each day and for each variable, we calculated the mean of the 12 readings from 6 AM to 6 PM.
Traffic Counts We obtained hourly traffic data col- lected by the Georgia Department of Transportation from the 4 function- ing sites (2 highways and 2 local roads) located within Atlanta’s perimeter in- terstate highway. Total 24-hour and 1-hour morning peak bidirectional traf- fic counts were available for analysis for 92% of the study weekdays and 85% of the study weekend days.
Public Transportation Data We examined the total number of pas- senger trips per day on Atlanta’s pub- lic buses and rail lines during the study period. The Metro Atlanta Rapid Tran- sit Authority provided data for aver- age weekday and weekend daily rider-
ship totals for the pre- and post- Olympic Games periods. For the Olympic period, totals for each of the 17 days were available for analysis.33
Statewide Gasoline Sales We collected and compared total gal- lons of gasoline purchased in the state of Georgia in June, July, and August 1991-1997. The Georgia State Depart- ment of Revenue routinely calculates these gallon totals based on receipts of the state fuel tax. The month of pur- chase was determined by the month in which the fuel distributors delivered the gasoline to individual filling stations.
Statistical Analysis We analyzed all collected data to de- termine the percentage change in mean values during the Olympic period com- pared with the 1996 summertime base- line period. One-way analysis of vari- ance t testing was used to determine if the daily air pollutant, meteorologi- cal, and traffic count values differed sig- nificantly between the 2 study peri- ods. Significance was defined as P#.05.
Further analysis of the asthma event data was performed using a time-series Poissonregressionmodel.Univariateand adjusted relative risk (RR) (with 95% confidence intervals [CIs]]) of asthma acute care events during the Olympic period compared with the baseline period was calculated for each of the 4 sources of data. The univariate analysis was based on the fraction of total acute care events with a primary diagnosis of asthma. The multivariate time-series Poisson model was fitted using gener- alized estimating equations to address possible serial (auto) correlation in the number of asthma events.36 Models were implemented using the GENMOD pro- cedure in SAS with AR(1) (SAS Insti- tute Inc, Cary, NC) to account for cor- relation in asthma events on a given day with the previous day’s events. The Durbin-Watson statistic was 1.96- 2.01, which indicates minimal residual serial correlation. This model was adjusted for day of the week (weekday vs weekend) and minimum tempera- ture (lagged 1 day to improve the fit of themodel).Totalmoldcountswereonly
Figure 1. Functioning Traffic Count and Air Quality Monitoring Sites in Atlanta, Ga, During the 1996 Summer Olympic Games
A
Ozone Carbon Monoxide Particulate Matter Nitrogen Dioxide Sulfur Dioxide
T Traffic Count Sites
A Allergen W Weather
Major Roadways
Monitoring Sites
Cobb Gwinnett
DeKalb
Fulton
Clayton
W
T T
T
T
Borders indicate the 5 central counties of metropolitan Atlanta included in the study.
AIR QUALITY AND CHILDHOOD ASTHMA
©2001 American Medical Association. All rights reserved. (Reprinted) JAMA, February 21, 2001—Vol 285, No. 7 899
collected on weekdays, and therefore, were not included in the multivariate model. Seasonality and time trends were not includedinthismodelduetothevery short study period covering only the summerseason.TheOlympicperiodwas modeled here instead of ozone or PM10
levels due to the correlation between these variables and because we wanted to emphasize the changes directly asso- ciated with the Olympic Games.
A second autoregressive time-series model analyzed the change in ozone con- centrations in Atlanta and the other 3 Georgia sites during the Olympic Games. The model was used to adjust for day of the week (weekday vs weekend) and trends in weather conditions (mean day- time wind speed, temperature, solar ra- diation, and barometric pressure). Rela- tive humidity was not included because it was strongly correlated with solar ra- diation (inverse relationship). The model (using AR[1]) found minimal serial cor- relation in daily ozone levels (Durbin- Watson statistic, 1.94).
In a separate analysis, Pearson coef- ficients were used to measure the cor- relation between morning peak–total traffic counts and that day’s peak ozone concentration. Significance was de- fined as P#.05.
In addition to our analyses compar- ing the Olympic period with the base- line period, data for all 73 study days wereused toexamine the relationshipbe- tween ozone accumulation and daily asthma acute care events. The mean
number of asthma events was deter- mined for days in which the peak ozone concentrations for that day plus the pre- vious 2 days averaged less than 60 ppb (low cumulative exposure), 60 to 89 ppb (moderate cumulative exposure), or 90 ppbormore (highcumulativeexposure). Controlling for day of the week and mini- mum temperature (lagged 1 day) in a Poisson regression model, the RR of asthma events was determined at cumu- lative ozone concentrations of 60 to 89 ppb and 90 ppb or more compared with ozone concentrations less than 60 ppb.
To test our hypothesis that cumula- tive exposures to air pollutants for 2 to 3 days are more strongly associated with asthma acute care events than tradi- tional single-day exposures, we com- pared the RR of asthma events per 50 ppb incremental change in ozone and 10-µg/m3 incremental change in PM10
levels based on a same-day, 2-day, and 3-day cumulative exposure measure. We used time-series regression mod- els similar to our primary asthma event model except that the dependent vari- able (Olympic vs baseline period) was replaced by a single pollutant expo- sure variable.
RESULTS Among children aged 1 through 16 years in the Medicaid claims file data- base, the number of asthma emer- gency care visits and hospitalizations decreased from 4.23 events per day dur- ing the baseline period to 2.47 events
per day during the Olympic period, a 41.6% overall decrease (TABLE 1). The number of asthma-related emergency department, urgent care visits, and hos- pitalizations among HMO enrollees de- creased 44.1% during the same time pe- riod. Asthma-related visits to 2 large pediatric emergency departments in At- lanta decreased 11.1%, and citywide hospitalizations for asthma were re- duced 19.1%. Using our 4 data sources, this equates to respectively 30, 10, 9, and 7 fewer emergency asthma events during the Olympic Games than would have been expected based on the pre- and post-Olympic period averages.
The total number of nonasthma- related acute care events per day de- creased only 3.1% in the Medicaid da- tabase and 2.1% at the pediatric emergency departments; this number increased 1.3% among HMO enroll- ees. Nonelective nonasthma hospital- izations increased 1.0% during the Olympics.
As illustrated in TABLE 2, the uni- variate RR of asthma acute care events during the Olympics compared with the baseline period was significantly re- duced in the Medicaid database (RR, 0.61; 95% CI, 0.44-0.85) and ap- proached statistical significance in the HMO database (RR, 0.56; 95% CI, 0.31-1.02). Although less than 1, the RR at the 2 pediatric emergency depart- ments and among those hospitalized was not significant (RR, 0.91 and 0.81, respectively). Adjusting the RR to con-
Table 1. Acute Asthma Events and Acute Nonasthma Events Among Children and Youth During the 1996 Summer Olympic Games Compared With the 1996 Summertime Baseline Period
Data Source Type of
Asthma Event
Acute Asthma Events Acute Nonasthma Events
Mean (SD) No. of Events Per Day
% Change
Mean (SD) No. of Events Per Day
% ChangeBaseline Period* Olympic Period† Baseline Period* Olympic Period†
Georgia Medicaid claims file
Emergency care and hospitalizations
4.23 (2.81) 2.47 (1.46) −41.6 100.5 (18.6) 97.4 (16.4) −3.1
Health maintenance organization
Emergency care, urgent care, and hospitalizations
1.36 (1.63) 0.76 (0.83) −44.1 37.6 (19.6) 38.1 (18.4) +1.3
Pediatric emergency departments
Emergency care and hospitalizations
4.77 (2.52) 4.24 (2.49) −11.1 118.4 (20.5) 115.9 (15.9) −2.1
Georgia Hospital Discharge Database
Hospitalizations 2.04 (1.53) 1.65 (1.50) −19.1 19.7 (5.1) 19.9 (3.5) +1.0
*Defined as June 21–July 18 and August 5–September 1, 1996. †Defined as July 19–August 4, 1996.
AIR QUALITY AND CHILDHOOD ASTHMA
900 JAMA, February 21, 2001—Vol 285, No. 7 (Reprinted) ©2001 American Medical Association. All rights reserved.
trol for minimum temperature and day of the week did not alter these find- ings. The data from Atlanta’s Medic- aid database remained statistically sig- nificant (adjusted RR, 0.48; 95% CI, 0.44-0.86).
FIGURE 2 shows the daily time se- ries of each of the 5 measured pollut- ants and mold counts during the Olym- pic and baseline periods. The 1-hour peak ozone concentration in Atlanta de- creased 27.9% from an average daily peak of 81.3 ppb during the baseline pe- riod to 58.6 ppb during the Olympic Games (P,.001). The daily ozone con- centrations at the 3 monitoring sites in Atlanta were highly correlated (r=0.91-0.97). Combined ozone data for the 1997 and 1998 summer sea- sons did not show a similar decrease— the average peak ozone concentration was 77.2 ppb during July 19–August 4 and 78.8 ppb during the remainder of the study dates.
Ozone concentrations at Georgia sites outside of Atlanta decreased 11.1% in Augusta (58.8 ppb vs 66.2 ppb; P=.11), 17.5% in Fannin County (50.5 ppb vs 61.2 ppb; P=.003), and 18.5% in Co-
lumbus (52.2 ppb vs 64.1 ppb; P=.01) during the Olympic Games. These ozone reductions were, respectively, 60%, 37%, and 34% less than the ozone reductions experienced in Atlanta during the same period with similar weather conditions. After controlling for the 4 weather vari- ables, serial correlation, and day of the week in a time-series regression model, the reduction in Atlanta’s ozone concen- trations during the Olympic Games was 13% (P=.06). Comparatively, the reduc- tion in ozone was calculated at 2% in Au- gusta (P=.71), 7% in Fannin County (P=.12), and 6% in Columbus (P=.30).
During the Olympic period, Atlanta additionally experienced significant re- ductions in daily carbon monoxide lev- els (1.26 vs 1.54 ppm, 18.5% decrease; P=.02) and PM10 concentrations (30.8 vs 36.7 µg/m3, 16.1% decrease; P=.01). Nitrogen dioxide levels decreased 6.8% (36.5 vs 39.2 ppb; P=.49), whereas sul- fur dioxide levels increased 22.1% (4.29 vs. 3.52 ppb; P=.65). FIGURE 3 summa- rizes these findings relative to the EPA National Ambient Air Quality Stan- dards for each of these pollutants.35 Data for the baseline period are divided into pre- and post-Olympic time periods
Figure 2. Daily Time Series of Individual Air Pollutant Levels and Mold Counts During the 1996 Summer Olympic Games and Baseline Period
150
100
50
0
P ea
k 1-
ho ur
L ev
el , p
pb
Ozone 80
60
40
20
0
24 -h
ou r
M ea
n Le
ve l,
µg /m
3
Particulate Matter ≤10 µm
80
60
40
20
0 Pre-Olympics Olympics Post-Olympics
P ea
k 1-
ho ur
L ev
el , p
pb
Nitrogen Dioxide
2.5
2.0
1.5
0.5
1.0
0 Pre-Olympics Olympics Post-Olympics Pre-Olympics Olympics Post-Olympics Pre-Olympics Olympics Post-Olympics
M ea
n 8-
ho ur
R un
ni ng
P ea
k Le
ve l,
pp m Carbon Monoxide
20
15
10
5
0 Pre-Olympics Olympics Post-Olympics
24 -h
ou r
M ea
n Le
ve l,
pp b
Sulfur Dioxide
1200
800
400
1000
600
200
0 Pre-Olympics Olympics Post-Olympics
M ol
ds /m
3
Total Daytime Mold Count
Broken line indicates incomplete data (eg, mold counts were available weekdays only).
Table 2. Univariate and Adjusted Relative Risk of Acute Asthma Events During the 1996 Summer Olympic Games Compared With the 1996 Summertime Baseline Period*
Data Source Univariate RR†
(95% CI) P Value Adjusted RR‡
(95% Confidence Interval) P Value
Georgia Medicaid claims file 0.61 (0.44-0.85) .003 0.48 (0.44-0.86) .005
Health maintenance organization
0.56 (0.31-1.02) .06 0.58 (0.32-1.06) .10
Pediatric emergency departments
0.91 (0.85-1.42) .48 0.93 (0.71-1.22) .69
Georgia Hospital Discharge Database
0.81 (0.54-1.23) .34 0.71 (0.46-1.11) .22
*RR indicates relative risk; CI, confidence interval. For definition of baseline period, see “Study Design” subsection of “Methods” section.
†Relative risk based on Poisson model (fraction of total acute care events with a primary diagnosis of asthma). ‡Time-series regression analysis was adjusted for day of week (weekday vs weekend) and minimum daily temperature
(lagged 1 day to minimize serial correlation).
AIR QUALITY AND CHILDHOOD ASTHMA
©2001 American Medical Association. All rights reserved. (Reprinted) JAMA, February 21, 2001—Vol 285, No. 7 901
demonstrating the uncharacteristic de- crease in air pollution levels during the Olympic Games.
Mean daytime weather conditions in Atlanta were determined for both the Olympic and baseline periods. Tem- perature decreased 0.67°C, wind speed increased 0.19 m/sec, and solar radia- tion decreased 29.6 W/m2 during the Olympic Games. These changes were not statistically significant. Baromet- ric pressure did not change. Total mold counts did not differ significantly dur- ing the Olympic vs the baseline period (daily mean, 597 vs 551 molds/m3; P = .58; Figure 2). Moreover, mold
counts were not correlated with same- day asthma events (average r=−0.15).
Weekday 1-hour morning peak traf- fic counts decreased 22.5% overall dur- ing the Olympic Games (range, 17.5%- 23.6%; P,.001 for all 4 sites). This amounted to a reduction of 4260 ve- hicle trips during the peak morning traf- fic hour on these 4 roads. Weekend morning peak traffic counts decreased 9.7% overall (range, 3.6%-12.3%), al- though only the change in traffic counts at the site closest to downtown was sig- nificant. Weekday total 24-hour traffic counts decreased 2.8% overall (range, 1.3%-3.6%), with the significant changes occurring at the 2 sites closest to down- town. Public transportation ridership in- creased 217% (190% on weekdays; 334% on weekends) during the Olym- pic Games. A total of 17.5 million more trips occurred on public transporta- tion throughout the Olympic Games than would be expected based on the baseline period ridership.
Weekly total gallons of gasoline pur- chased in Atlanta were not available for analysis. The number of gallons of gaso- line purchased statewide in July 1996 was 3.9% lower than June and August 1996. In contrast, July sales for 1995 and 1997 were 1.2% higher than the June and August sales for those 2 years.
To explore whether automobile traf- fic is a critical factor in urban ozone ac- cumulation, we analyzed the relation-
ship between weekday traffic counts and peak ozone concentrations on that day. We found a significant correlation be- tween 1-hour morning peak traffic counts and peak ozone concentrations at all 4 traffic-count sites (Pearson r=0.29, 0.42, 0.34, and 0.39, respec- tively; average r=0.36). No difference in this correlation was seen between the Olympic and baseline periods. An equally strong and significant correla- tion was found between total 24-hour traffic counts and ozone concentra- tions (average r=0.38; range, 0.33-0.48).
We used data from the entire 73- day study period to analyze the rela- tionship between the number of asthma acute care events on a given day and the average daily peak ozone concentra- tion during the preceding 48 to 72 hours. For data from the Medicaid, HMO, and emergency department da- tabases, the RR of asthma events in- creased stepwise at cumulative ozone concentrations 60 to 89 ppb and 90 ppb or more compared with ozone concen- trations of less than 60 ppb (TABLE 3). This trend was significant for the Med- icaid and emergency department data.
A 3-day cumulative exposure mea- sure was selected for the analysis shown in Table 3 instead of a single-day or 2-day exposure measure because it was found to be more consistently corre- lated with asthma events (TABLE 4). For data from the Medicaid and pediatric emergency department databases, the RR of asthma events per incremental changes in ozone and PM10 levels in- creased as the number of cumulative ex- posure days included increased from 1 to 2. The RR per incremental change in PM10 further increased when the number of cumulative exposure days increased from 2 to 3. However, in- creasing the cumulative exposure from 2 to 3 days did not further increase the RR per incremental change in ozone.
COMMENT Evidence linking air quality to respira- tory health has been accumulating in recent years. Many studies, including 2 conducted in Atlanta,16,37 have dem- onstrated significant associations be-
Figure 3. Mean Levels of Major Pollutants Before, During, and After the 1996 Summer Olympic Games as a Percentage of the National Ambient Air Quality Standard (NAAQS)
80
70
60
50
40
30
20
10
0 Ozone PM10 CO NO2 SO2
% o
f N A
A Q
S
Pre-Olympics (Baseline Period) Olympic Period Post-Olympics (Baseline Period)
National Ambient Air Quality Standard at time of study: ozone 1-hour peak average, 120 ppb; particulate mat- ter of 10 µm or smaller (PM10) 24-hour average, 150 µg/m3; carbon monoxide (CO) 8-hour average, 9 ppm; nitrogen dioxide (NO2) 1-hour peak average, 600 ppb; sulfur dioxide (SO2) 24-hour average, 140 ppb.35
Table 3. Relationship Between Acute Asthma Events and the 3-Day Cumulative Ozone Levels During the 73-day Summer Study Period*
Data Source
Ozone ,60 ppb† Ozone 60-89 ppb† Ozone $90 ppb†
Mean Daily
Asthma Events RR
Mean Daily
Asthma Events
RR (95% CI)
Mean Daily
Asthma Events RR (95% CI)
Georgia Medicaid claims file
2.20 1.00 3.85 1.61 (1.13-2.30)‡ 5.11 1.88 (1.24-2.83)‡
Health maintenance organization
1.07 1.00 1.15 1.11 (0.63-1.96) 1.50 1.33 (0.68-2.61)
Pediatric emergency departments
3.00 1.00 4.65 1.33 (0.98-1.81) 6.00 1.46 (1.02-2.09)§
Georgia Hospital Discharge Database
1.67 1.00 2.15 1.19 (0.77-1.84) 1.72 1.03 (0.58-2.11)
*CI indicates confidence interval. Relative risk (RR) based on Poisson model (fraction of total acute care events with a primary diagnosis of asthma); time-series regression analysis was adjusted for day of the week (weekday vs week- end) and minimum daily temperature (lagged 1 day to minimize serial correlation).
†Average of the peak ozone concentrations for 3 days (day of event plus the preceding 2 days). ‡P#.01. §P = .04.
AIR QUALITY AND CHILDHOOD ASTHMA
902 JAMA, February 21, 2001—Vol 285, No. 7 (Reprinted) ©2001 American Medical Association. All rights reserved.
tween days with high ozone levels and increased rates of asthma exacerba- tions.12-15,20-23 Our results support these previous findings and also indicate that extended reductions in ozone and PM10
concentrations at levels considerably below the EPA’s National Ambient Air Quality Standards can reduce asthma morbidity in children. Furthermore, our findings suggest that by decreasing au- tomobile emissions through citywide changes in transportation and com- muting practices, a substantial num- ber of asthma exacerbations requiring medical attention can be prevented.
We found variation in the relative change in asthma acute care events dur- ing the Olympic Games among the 4 sources of medical data. All showed a de- crease in asthma events, ranging from 11% to 44%. Only the data from the Georgia Medicaid claims file reached sta- tistical significance, which may be re- lated to the low power of this short- term intervention. Based on a power of 80% and the number of events in each database, the percent reduction in asthma events needed to detect a sig- nificant difference between the base- line and Olympic periods was 37% for the Georgia Medicaid claims file, 58% for the health maintenance organiza- tion database, 33% for the pediatric emergency room database, and 51% for the Georgia Hospital Discharge data- base (without adjustment for serial cor- relation). Other possible explanations in- clude differences between the study
population in the 4 data sources in asthma severity, medication use,38 ex- posure to other allergens, particularly those that could synergistically worsen the untoward effects of ozone and other air pollutants,39,40 extent of indoor ozone exposure,41 and of outdoor exposure.
The observed reductions in asthma events require us to address the follow- ing potentially confounding situation: did enough Atlanta children leave the city during the Olympics to signifi- cantly reduce the number of children with asthma who would seek medical attention for an acute asthma exacer- bation? If true, this could account, in whole or in part, for the reduction in asthma events observed. However, for all 4 data sources, use of nonasthma- related urgent and emergency medical services by Atlanta children changed minimally. This suggests that neither the size of the study population nor use of emergency medical services by this population changed significantly dur- ing the Olympic games.
Of the factors potentially affecting asthma morbidity that we could readily assess, air quality remains the likely cause for the decline in asthma acute care events. Our analysis demonstrated a large and significant decrease in ozone concentrations, and to a lesser extent, PM10, and carbon monoxide concentra- tions. Of all the pollutants, Atlanta’s ozone concentrations in the summer most frequently violate the National Am- bient Air Quality Standards.42 These
standards attempt to set a maximum pol- lution level, which, if exceeded, may be hazardous to the general public’s health. The standards are based, in part, on the available information regarding the health effects of the 5 major air pollut- ants. However, pollution levels below these standards may be harmful to cer- tain, high-risk populations (such as in- dividuals with asthma and the elderly). Therefore, the 28% drop in ozone con- centrations during the Olympic Games represented a substantial decrease in a potential health hazard.
Our study design and findings make it difficult to determine to what de- gree the observed reductions in ozone, PM10, carbon monoxide, and nitrogen dioxide pollution individually contrib- uted to the observed changes in health. As controlled studies of human expo- sure to multiple pollutants have dem- onstrated,12,13,19 the effects of reduced levels of ozone and these primary pol- lutants likely were additive. This is sup- ported by the finding that both ozone and PM10 levels were similarly corre- lated with asthma events. The fact that ozone and PM10 levels were highly cor- related with each other (r=0.58-0.69) additionally limits our ability to deter- mine which pollutant(s) accounted for the reduction in asthma events. This correlation between ozone and PM10
levels should have been expected, given that the environmental changes occur- ring during the study period (ie, de- creased automobile emissions and
Table 4. Comparison of the Relative Risk of Asthma Events per 50-ppb Incremental Change in Ozone and 10-µg/m3 Incremental Change in PM10 Levels Based on a Same-Day, 2-Day, and 3-Day Cumulative Exposure Measure*
Data Source
Relative Risk (95% Confidence Interval)
Ozone PM10
Same Day 2 Day 3 Day Same Day 2 Day 3 Day
Georgia Medicaid claims file 1.1 (0.88-1.47) 1.4 (1.05-1.93)† 1.4 (1.01-1.94)† 0.9 (0.57-1.49) 1.1 (0.62-1.78) 1.4 (0.80-2.48)
Health maintenance organization
1.4 (0.88-2.14) 1.2 (0.68-1.96) 1.2 (0.69-2.11) 2.1 (0.92-4.61) 1.8 (0.72-4.40) 1.8 (0.68-4.81)
Pediatric emergency departments
1.2 (0.99-1.56) 1.4 (1.04-1.79)† 1.4 (1.03-1.86)† 1.2 (0.81-1.89) 1.3 (0.81-2.08) 1.5 (0.91-2.47)
Georgia Hospital Discharge Database
0.9 (0.64-1.34) 1.0 (0.62-1.51) 1.0 (0.61-1.58) 0.7 (0.34-1.45) 0.6 (0.29-1.45) 0.8 (0.33-1.88)
*PM10 indicates particulate matter of 10 µm or smaller; same day, single-day measure of exposure (day of asthma event only); 2 day, 2-day measure of exposure (day of asthma event plus the preceding day); and 3 day, 3-day measure of exposure (day of asthma event plus the preceding 2 days). Relative risk based on Poisson time-series regression analysis was adjusted for day of the week (weekday vs weekend) and minimum daily temperature (lagged 1 day to minimize serial correlation).
†P#.05.
AIR QUALITY AND CHILDHOOD ASTHMA
©2001 American Medical Association. All rights reserved. (Reprinted) JAMA, February 21, 2001—Vol 285, No. 7 903
weather variability) would have had a strong influence on the daily levels of both these pollutants.
The more immediate question is what accounted for this change in air quality. We suggest that it was caused by changes in both meteorological conditions and automobile emissions, with decreases in peak morning rush hour traffic playing a major role. Weather conditions dur- ing the Olympic Games (increased wind speed and decreased temperature and so- lar radiation) favored less accumula- tion of ozone, but the degree of weather improvements was measurably small and not statistically significant. Even when controlling for these weather variables in a multivariate regression model, ozone levels in Atlanta during the Olympic Games were reduced 13% whereas the changes in ozone levels at the other 3 Georgia sites with the same prevailing weather patterns were reduced be- tween 2% and 7%.
Other indirect evidence supports our conclusion. The concentration of car- bon monoxide, which is primarily emit- ted directly from automobiles and is much less dependent on weather con- ditions for its accumulation in the lower atmosphere, decreased significantly during the Olympic Games. The small increase in sulfur dioxide levels (far below health hazard levels) during the Olympic Games is consistent with the increased use of diesel-powered buses,31,32 and should not have in- creased if the prevailing weather con- ditions had indeed prevented the nor- mal accumulation of air pollutants in Atlanta. The amount of emissions from stationary sources (eg, power plants and industry) did not change during the Olympic Games.31,32 The additional electrical needs required during the Olympic Games came from power stations outside the immediate At- lanta area and, therefore, would not have caused the increase in sulfur dioxide observed.
Evidence of changes in automobile traffic and emissions include the marked decreases in weekday and weekend morningpeaktrafficcountsatall4 traffic- count sites, the statistically significant
decreases in weekday total traffic counts at the2traffic-countsitesclosest todown- town Atlanta, the statistically signifi- cantcorrelationbetweenweekdaymorn- ing peak and 24-hour total traffic counts and that day’s peak ozone concentra- tion, the3.9%decrease instatewidegaso- line sales in Julycomparedwith Juneand August, and the 217% increase in over- all public transportation use. These traf- fic data probably underestimate the impact of the alternative transportation strategies on local residents of Atlanta because they include automobile use by the estimated 1 million visitors during the Olympic intervention period. Using this same logic, however, the increase in public transportation use is probably an overestimation of the behaviors of local residentssince italso includesusebyvisi- tors to Atlanta.
The science of ozone formation helps explain our findings. The moderate al- terations in morning traffic levels (and probably traffic flow) experienced dur- ing the Olympic Games would have de- creased the buildup of ozone precur- sors emitted into the atmosphere from 7 AM through 2 PM. Without sufficient atmospheric concentrations of these precursors being present during this time of maximum sunlight and heat, rapid ozone production and accumu- lation could not occur, thus leading to lower than anticipated peak ozone lev- els. During a period of 17 days, this ap- peared to have contributed to the im- proved respiratory health of children with asthma residing in Atlanta. What motivated businesses and individuals to change their transportation and com- muting behaviors temporarily is a cru- cial question, which has not been prop- erly addressed. Fear of traffic and lack of parking, and social pressures to con- form certainly played a role. How this can be adapted to more routine condi- tions remains a major public health challenge. For example, Atlanta’s Clean Air Campaign43 (largely initiated after the Olympic Games) has been shown to increase use of alternative commut- ing methods within 3 companies that promoted this.44 But the effects of this citywide campaign on air pollution to
date appear to be small compared with what was observed during the Olym- pic Games.
The weight of evidence linking air quality to respiratory health contin- ues to grow. Our findings suggest that efforts to decrease ozone and PM10 con- centrations from moderate to low lev- els can decrease the burden of asthma. Consistent with our findings for the Olympic period, we found an increas- ing correlation between daily asthma events and ozone and PM10 levels as the period of exposure was increased from less than 24 hours preceding the asthma event to the 48 to 72 hours preceding the event. When using this 3-day cu- mulative exposure measure, the risk of asthma exacerbations increased sub- stantially after only moderate levels of ozone exposure (60-89 ppb). These data suggest that the cumulative effect of moderate levels of ozone and other pol- lutants during a more extended pe- riod is as or more important to respi- ratory health than single-day levels exceeding the national standards (ie, 1-hour peak ozone .120 ppb or 8-hour peak .80 ppb). Recent research with PM10 supports the important effect of extended exposure to pollutants on res- piratory health.45,46
Our study methods and findings have several limitations. The special nature of the Olympic Games, the relatively short intervention period and limited statistical power, the lack of an up- dated traffic counting system, and the limited number of air pollution moni- toring sites (for PM10 and carbon mon- oxide) make firm conclusions diffi- cult. Unmeasured medical or social factors may have influenced our find- ings. A citywide alternative transpor- tation and commuting intervention, not associated with a special event, would provide a better study situation. How- ever, the power of the Olympic Games to transform behavioral norms should not be underestimated and deserves close scrutiny for its lessons.
We conclude that the alternative transportation plan in Atlanta during the Olympic Games reduced ozone and other air pollutants and was associated
AIR QUALITY AND CHILDHOOD ASTHMA
904 JAMA, February 21, 2001—Vol 285, No. 7 (Reprinted) ©2001 American Medical Association. All rights reserved.
with a significant, albeit temporary, de- crease in the burden of asthma among Atlanta’s children.
Author Affiliations: Epidemic Intelligence Service (Dr Friedman) and Chronic Disease, Injury, and Environ- mental Epidemiology Section (Dr Powell), Epidemiol- ogy and Prevention Branch, Georgia Division of Public Health, and Epidemiology Program Office, Centers for Disease Control and Prevention (Dr Friedman and Ms Hutwagner), Department of Pediatrics, Morehouse School of Medicine, and Georgia Pediatric Pulmonary Associates (Dr Graham), Division of Pediatric Pulmo- nary and Critical Care Medicine, Egleston Children’s Hos- pital and Emory University (Dr Teague), Atlanta, Ga. Author Contributions: Dr Friedman was the study co-
ordinator and principal investigator, and participated in the study concept and design, acquisition of data, analysis and interpretation of data, drafting and criti- cal revision of the manuscript, and provided admin- istrative, technical, or material support. Dr Powell participated in the study concept and de- sign, acquisition of data, analysis and interpretation of data, drafting and critical revision of the manu- script, and provided supervision. Ms Hutwagner participated in the analysis and inter- pretation of data, drafting and critical revision of the manuscript, and provided statistical expertise. Dr Graham participated in the study concept and de- sign, acquisition of data, critical revision of the manu- script, and provided administrative, technical, or ma- terial support. Dr Teague participated in the study concept and de- sign, acquisition of data, critical revision of the manu-
script, and provided administrative, technical, or ma- terial support. Acknowledgment: We thank the many persons and their respective institutions who provided us with the data and/or technical expertise without which this study could not have been completed. This includes Rafael Balagas and Bill Murphy at the Georgia Envi- ronmental Protection Division, Deborah Griffin and Dennis Tolsma at Kaiser-Permanente, Gene McDow- ell at Scottish-Rite Medical Center, Egleston Chil- dren’s Hospital, Phil Harris at the Georgia Depart- ment of Medical Assistance, Dale Shuirman at the Georgia Department of Revenue, Chris Porter at Cambridge Systematics, the Georgia Department of Transportation, the Atlanta Allergy and Asthma Clinic, Metro Atlanta Rapid Transit Authority, and Darren Palmer at the US Environmental Protection Agency.
REFERENCES
1. Centers for Disease Control and Prevention. Mea- suring childhood asthma prevalence before and after the 1997 redesign of the National Health Interview Survey—United States. MMWR Morb Mortal Wkly Rep. 2000;49:908-911. 2. Centers for Disease Control and Prevention. Sur- veillance for asthma—United States, 1960-1995: CDC surveillance summaries. MMWR Morb Mortal Wkly Rep. 1998;47:SS1-SS27. 3. Centers for Disease Control and Prevention. Asthma—United States, 1982-1992. MMWR Morb Mortal Wkly Rep. 1995;43:952-955. 4. Centers for Disease Control and Prevention. Asthma mortality and hospitalization among children and young adults—United States, 1980-1993. MMWR Morb Mortal Wkly Rep. 1996;45:350-353. 5. Abramson MJ, Kutin J, Czarny D, Walters EH. The prevalence of asthma and respiratory symptoms among young adults: is it increasing in Australia? J Asthma. 1996;33:189-196. 6. Anderson HR. Increase in hospital admissions for childhood asthma: trends in referral, severity and re- admissions from 1970 to 1985 in a health region of the United Kingdom. Thorax. 1989;44:614-619. 7. Goren AI, Hellmann S. Has the prevalence of asthma increased in children? evidence from a long term study in Israel. J Epidemiol Community Health. 1997;51: 227-232. 8. Mullaly DI, Howard WA, Hubbard TJ, et al. In- creased hospitalizations for asthma among children in the Washington, DC area during 1961-1981. Ann Al- lergy. 1984;53:15-19. 9. Becklake MR, Ernst P. Environmental factors. Lan- cet. 1997;350(suppl 2):10-13. 10. Schenker M. Air pollution and mortality. N Engl J Med. 1993;329:1807-1808. 11. Abramson MJ, Marks GB, Pattemore PK. Are non-allergenic environmental factors important in asthma ? Med J Aust. 1995;163:542-545. 12. Holgate ST, Samet JM, Koren HS, Maynard RL. Air Pollution and Health. London, England: Aca- demic Press; 1999. 13. Committee of the Environmental and Occupa- tional Health Assembly, American Thoracic Society. Health effects of outdoor air pollution. Am J Respir Crit Care Med. 1996;153:3-50. 14. Brunekreef B, Dockery DW, Krzyzanowski M. Epi- demiologic studies on short-term effects of low lev- els of major ambient air pollution components. Envi- ron Health Perspect. 1995;103(suppl 2):3-13. 15. Tatterfield AE. Air pollution: brown skies re- search. Thorax. 1996;51:13-22. 16. White MC, Etzel RA, Wilcox WD, Lloyd C. Exac- erbations of childhood asthma and ozone pollution in Atlanta. Environ Res. 1994;65:56-68.
17. Anderson HR, Ponce de Leon A, Bland JM, et al. Air pollution and daily mortality in London: 1987-92. BMJ. 1996;312:665-669. 18. Buchdahl R, Parker A, Stebbings T, Babiker A. As- sociation between air pollution and acute childhood wheezy episodes: prospective observational study. BMJ. 1996;312:661-665. 19. Avol EL, Linn WS, Shamoo DA, et al. Respiratory effects of photochemical oxidant air pollution in ex- ercising adolescents. Am Rev Respir Dis. 1985;132: 619-622. 20. CodyRP,WeiselCP,BirnbaumG,LioyPJ.Theeffect of ozone associated with summertime photochemical smog on the frequency of asthma visits to hospital emer- gency departments. Environ Res. 1992;58:184-194. 21. Romieu I, Meneses F, Sienra-Monge JJ, et al. Ef- fects of urban air pollutants on emergency visits for childhood asthma in Mexico City. Am J Epidemiol. 1995;141:546-553. 22. Romieu I, Meneses F, Ruiz S, et al. Effects of air pollution on the respiratory health of asthmatic chil- dren living in Mexico City. Am J Respir Crit Care Med. 1996;154:300-307. 23. Thurston GD, Lippmann M, Scott MB, Fine JM. Summertime haze air pollution and children with asthma. Am J Respir Crit Care Med. 1997;155:654-660. 24. Pope CA. Respiratory disease associated with com- munity air pollution and a steel mill, Utah Valley. Am J Public Health. 1989;79:623-628. 25. Livingstone AE, Shaddick G, Grundy C, Elliott P. Do people living near inner city main roads have more asthma needing treatment? case-control study. BMJ. 1996;312:676-677. 26. Edwards J, Walters S, Griffiths RC. Hospital ad- missions for asthma in pre-school children: relation- ship to major roads in Birmingham UK. Arch Environ Health. 1994;49:223-237. 27. Wjst M, Reitmeir P, Dold S, et al. Road traffic and adverse effects on respiratory health in children. BMJ. 1993;307:596-600. 28. Weiland SK, Mundt KA, Ruckmann A, Keil U. Self- reported wheezing and allergic rhinitis in children and traffic density on street of residence. Ann Epidemiol. 1994;4:243-247. 29. Murakami M, Ono M, Tamura K. Health prob- lems of residents along heavy traffic roads. J Hum Er- gol (Tokyo). 1992;19:101-106. 30. What You Can Do to Reduce Air Pollution. Wash- ington, DC: United States Environmental Protection Agency Office of Air and Radiation; 1992;12. EPA 450- K-92-002. 31. Balagas R. Improved Air Quality in Atlanta Dur- ing the 1996 Olympic Summer Games. Atlanta: Geor- gia Environmental Protection Division, Georgia De- partment of Natural Resources; 1996.
32. Porter C. Changes in Air Quality and Transpor- tation Associated With the 1996 Atlanta Summer Olympics. Cambridge, Mass: Cambridge Systemat- ics Inc; 1997. Prepared for: National Cooperative High- way Research Program (NCHRP 8-33). 33. Metropolitan Atlanta Rapid Transit Authority. The Way to the Games: A Report on Mass Transit During the 1996 Summer Olympic Games. Atlanta, Ga: MARTA; 1996. 34. Busse WW, Holgate ST. Asthma and Rhinitis. Cambridge, Mass: Blackwell Science; 1995. 35. Measuring Air Quality: The Pollutant Standards Index. Washington, DC: Environmental Protection Agency Office of Air Quality Planning and Stan- dards; 1994. EPA 451-K-94-001. 36. Zeger SL, Liang KY. Longitudinal data analysis for discrete and continuous outcomes. Biometrics. 1996; 42:121-130. 37. Tolbert P, Mulholland JA, MacIntosh DL, et al. Air quality and pediatric emergency room visits for asthma in Atlanta, Georgia. Am J Epidemiol. 2000;151:798- 810. 38. Hartert TV, Windom HH, Peebles RS, Friedhoff LR, Togias A. Inadequate outpatient medical therapy for patients with asthma admitted to two urban hos- pitals. Am J Med. 1996;100:386-394. 39. Jorres R, Nowak D, Magnussen H. The effect of ozone exposure on allergen responsiveness in sub- jects with asthma or rhinitis. Am J Respir Crit Care Med. 1996;153:56-64. 40. Molfino NA, Wright SC, Katz I, et al. Effect of low concentrations of ozone on inhaled allergen re- sponses in asthmatic subjects. Lancet. 1991;338:199- 203. 41. Hayes SR. Use of an indoor air quality model (IAQM) to estimate indoor ozone levels. J Air Waste Manage Assoc. 1991;41:171-181. 42. Environmental Protection Agency Green Book. Available at: http://www.epa.gov/oar/oaqps/greenbk /index.html. Accessed February 2, 2001. 43. Clean Air Campaign. Available at: http://www .cleanaircampaign.com. Accessed February 2. 2001. 44. Centers for Disease Control and Prevention. Cor- porate action to reduce air pollution—Atlanta, Geor- gia, 1998-1999. MMWR Morb Mortal Wkly Rep. 2000;49:153-156. 45. Delfino RJ, Zeiger RS, Seltzer JM, Street DH. Symptoms in pediatric asthmatics and air pollution: differences in effects by symptom severity, anti- inflammatory medication use and particulate averag- ing time. Environ Health Perspect. 1998;106:751- 761. 46. Schwartz J. The distributed lag between air pol- lution and daily deaths. Epidemiology. 2000;11:320- 326.
AIR QUALITY AND CHILDHOOD ASTHMA
©2001 American Medical Association. All rights reserved. (Reprinted) JAMA, February 21, 2001—Vol 285, No. 7 905
memo3/HealthyEquitableTransportationPolicy.pdf
PolicyLink Prevention Institute Convergence Partnership
Healthy, Equitable Transportation Policy RECOMMENDATIONS AND RESEARCH
Design by Chen Design Associates
Leslie Yang for PolicyLink
PolicyLink PolicyLink is a national research and action institute advancing economic and social equity by Lifting Up What Works.®
Prevention Institute Putting prevention and equitable health outcomes at the center of community well-being.
This report was commissioned by the Convergence Partnership which includes the following institutions: The California Endowment Kaiser Permanente The Kresge Foundation Nemours Robert Wood Johnson Foundation W.K. Kellogg Foundation Centers for Disease Control and Prevention as technical advisors
Healthy, Equitable Transportation Policy Recommendations and Research
EDITED BY
SHirEEn MalEkafzali Senior Associate, PolicyLink
p
g . 2
>
>
5 Foreword Congressman James Oberstar, Chairman of the House Transportation and Infrastructure Committee
6 Preface Angela Glover Blackwell, Founder and CEO, PolicyLink
9 The Transportation Prescription: A Summary of Findings and a Framework for Action Judith Bell, M.P.A., President, PolicyLink Larry Cohen, M.S.W., Founder and Executive Director, Prevention Institute
21 Chapter 1. Health Effects of Transportation Policy Judith Bell, M.P.A., President, PolicyLink Larry Cohen, M.S.W., Founder and Executive Director, Prevention Institute
27 Chapter 2. Transportation Authorization 101: A Backgrounder Susan Polan, Ph.D., Associate Executive Director, American Public Health Association Tracy Kolian, M.P.H., Senior Health Policy Analyst, American Public Health Association Shireen Malekafzali, M.P.H., Senior Associate, PolicyLink
35 Transportation Choices
37 Chapter 3. Public Transportation and Health Todd Litman, M.E.S., Founder and Executive Director, Victoria Transport Policy Institute
63 Chapter 4. Walking, Bicycling, and Health Susan Handy, Ph.D., Professor of Environmental Science and Policy, Director of the Sustainable Transportation Center, University of California, Davis
79 Chapter 5. Roadways and Health: Making the Case for Collaboration Catherine L. Ross, Ph.D., Director, Center for Quality Growth and Regional Development, Harry West Chair, Georgia Tech
Contents
97 Key Issues
99 Chapter 6. Breaking Down Silos: Transportation, Economic Development, and Health Todd Swanstrom, Ph.D., E. Desmond Lee Professor of Community Collaboration and Public Policy Administration, University of Missouri, St. Louis
113 Chapter 7. Sustainable Food Systems: Perspectives on Transportation Policy Kami Pothukuchi, Ph.D., Associate Professor of Urban Planning, Wayne State University Richard Wallace, Senior Project Manager, Center for Automotive Research
131 Chapter 8. Traffic Injury Prevention: A 21st-Century Approach Larry Cohen, M.S.W., Founder and Executive Director, Prevention Institute Leslie Mikkelsen, R.D., M.P.H., Managing Director, Prevention Institute Janani Srikantharajah, B.A., Program Coordinator, Prevention Institute
146 Author Biographies
150 Acknowledgments
151 Notes
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 3
<
<
C o
n te
n ts
chapter name
p g
. 4
>
>
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 5
<
<
F o
re w
o rd
Discussions of public health and wellness often are limited to the health and medical fields. It is my hope that soon, the transportation sector will be part of the discussion and play a role in providing solutions to improving the nation’s overall health, well-being, and quality of life.
One of my goals as Chairman of the Committee on Transportation and Infrastructure is to create a new model for surface transportation, one that invests in alternative modes and promotes active, healthy lifestyles. Public health and transportation policy choices are inextricably linked. The transportation sector is responsible for one-third of the greenhouse gas emissions in the United States. Our infrastructure and land use choices often dictate our daily travel, and whether or not we have access to clean, healthy transportation options. And in any given year, approximately 40,000 Americans are killed on our roadways. The policy decisions we make regarding transportation have repercussions on public health throughout our society.
For too long now, our transportation decision making has failed to address the impacts that our infrastructure network has on public health and equity. The asphalt poured and lane miles constructed enhanced our mobility and strengthened our economic growth; but too often, this auto-centric mindset took hold and crowded out opportunities to invest in a truly sustainable intermodal transportation system, in particular a system that meets the needs of underserved communities.
The failure to link transportation and land use decision making, and to consider the public health effects of these choices, has led to a tilted playing field that has made driving the easiest—and often the only—option available in many parts of the country. Our transportation policies and investments must do more to provide access for all through various modes. Transit, walking, and bicycling all have a significant role to play in lowering our dependence on foreign oil, reducing our greenhouse gas emissions and air pollutants,
and helping Americans incorporate exercise and fresh air into their daily travel routines. We must also continue our pursuit to reduce the number—and rate—of traffic fatalities and injuries that occur each year.
Our most recent surface transportation legislation, enacted in 2005, took important steps toward building a healthier infrastructure by investing billions of dollars in safety, public transit, walking, and bicycling. This legislation is helping to construct safer infrastructure, enable workforce development, build new transit lines, repair existing systems, and establish non- motorized transportation networks. We also enacted the Safe Routes to School program, which allows states to invest in safety improvements and education campaigns to get kids walking and biking to school again. This program has shown great early success and has the ability to change the habits of an entire generation.
Environmental sustainability, access, and our collective well-being must combine with mobility and safety as the cornerstones of our transportation investments. The following report represents an important contribution to our emerging understanding of the connections between transportation and public health and is an invaluable resource for policymakers and all those interested in building healthy communities. With a greater recognition of the strong linkage between public health and transportation, I believe we can build a network that supports our mobility and creates access and economic strength while promoting equity, sustaining our good health and quality of life.
Congressman James Oberstar
Chairman of the House Transportation and Infrastructure Committee
Foreword Congressman James Oberstar
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 6
>
>
P
re fa
c e
Transportation policy has enormous potential to catalyze the development of healthy communities of opportunity. The upcoming authorization of the federal surface transportation bill represents the single biggest federal opportunity to influence how our communities, cities, and regions are shaped.
Transportation impacts health directly; it affects air quality, injury risk, physical activity levels, and access to necessities such as grocery stores. Transportation is also one of the largest drivers of land use patterns; it thus determines whether communities have sidewalks and areas to play and be physically active as well as whether communities are connected to or isolated from economic and social opportunities.
Research shows that low-income communities and communities of color often do not have access to the benefits our transportation system can provide, yet they bear the burdens of that system. For example, many low-income neighborhoods have little or no efficient, reliable public transportation to get them to jobs and essential goods and services. But these communities are often situated near bus depots, highways, and truck routes, where pollution levels are high—and not coincidentally, asthma rates are high as well. In addition, many of these same communities live without safe, complete sidewalks or bike paths, making walking and biking difficult and often dangerous. As a result, these neighborhoods
have low levels of physical activity and high rates of chronic diseases. Creating a more equitable transportation system must lie at the core of any analysis of transportation or health, and it must guide all reform.
The Convergence Partnership, the collaborative of funders that commissioned this project, embraces the imperative that health and equity be central to transportation policy debates. Further, the Convergence Partnership recognizes how transportation policy is connected to the Partnership’s broader efforts to support environmental and policy changes that will create healthy people and healthy places. The Partnership’s steering committee includes: The California Endowment, Kaiser Permanente, the Kresge Foundation, Nemours, the Robert Wood Johnson Foundation, and the W. K. Kellogg Foundation. The Centers for Disease Control and Prevention serves as technical advisor.
In this project, leading academic researchers and advocates working at the intersection of transportation policy, equity, and public health identify opportunities for creating transportation systems that promote health and equity. This report synthesizes their insights and offers concrete recommendations for change.
Reform is long overdue. Climate change, shameful health disparities, growing rates of chronic diseases—transportation policy has contributed to these problems, and now it must
Preface Angela Glover Blackwell
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 7
<
<
P re
fa c
e
address them. Increasing rates of poverty and a severe economic downturn add to the urgency for reform.
This report intentionally uses the term authorization and not the more common word, reauthorization, in reference to the surface transportation bill. We want to make clear that new thinking and innovative approaches are necessary to meet the needs of a changing and diverse America.
Many advocates are already working hard to push for fundamental reform. This report was written for community leaders, policymakers, funders, practitioners, and advocates interested in an overarching strategy to promote active living and to build healthy communities of opportunity. PolicyLink, Prevention Institute, and the Convergence Partnership believe that building healthy communities requires a collaboration of stakeholders from diverse fields and sectors. Together, we can identify and support shared solutions.
The project recognizes that effective strategies to improve health, particularly in vulnerable communities, often fall outside the conventional domain of health policy, yet deserve equal attention. Federal transportation policy is a critical opportunity at our fingertips. Leveraging the strength of collaboration and networking can yield powerful results. Let’s seize the moment.
Angela Glover Blackwell
Founder and CEO PolicyLink
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 8
>
>
C h
a p
te r
2
chapter name
p g
. 8
>
>
The Transportation Prescription: A Summary of Findings and a Framework for Action JUDITH BELL, M.P.A. President, PolicyLink
LARRY COHEN, M.S.W. Founder and Executive Director, Prevention Institute
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 0
>
>
In St. Louis, MO, major cuts in bus service this spring left workers, students, disabled people, and elderly residents stranded and feeling bereft. Stuart and Dianne Falk, who are both in wheelchairs, told CNN they no longer would be able to get to the gym or the downtown theater company where they volunteer. “To be saddled, to be imprisoned, that is what it is going to feeling like,” Stuart Falk said.1
In West Oakland, CA, families have no escape from the diesel exhaust belching from trucks at the nearby port: The air inside some homes is five times more toxic than in other parts of the city. “I’m constantly doing this dance about cleaning diesel soot from my blinds and window sills,” 57-year-old Margaret Gordon told the San Francisco Chronicle.2
In Seattle, WA, Maggieh Rathbun, a 55-year- old diabetic who has no car, takes an hour-long bus ride to buy fresh fruits and vegetables. She cannot haul more than a few small bags at a time so she shops frequently—if she feels well enough. “It depends on what kind of day I’m having with my diabetes to decide whether I’m going to make do with a bowl of cereal or try to go get something better,” she told the Seattle Post-Intelligencer.3
Our transportation system has an enormous impact on our way of life, on the air we breathe, and on the vitality of our communities. Transportation choices influence personal decisions about where to live, shop, attend school, work, and enjoy leisure. They affect stress levels, family budgets, and the time we spend with our children. Although most people don’t think of it as a determinant of health, our transportation system has far-reaching implications for our risk of disease and injury. Transportation policies and accompanying land use patterns contribute to the glaring health disparities between the affluent and the poor and between white people and people of color.
This report demonstrates that transportation policy is, in effect, health policy—and
environmental policy, food policy, employment policy, and metropolitan development policy, each of which bears on health independently and in concert with the others. Longstanding transportation and land use policies are at odds with serious health, environmental, and economic needs of the country, and they have harmed low-income communities and communities of color especially. Forward- thinking transportation policies must promote healthy, green, safe, accessible, and affordable ways of getting where we need to go. They also must go hand in hand with equitable, sustainable land use planning and community economic development.
Streets and roads are the largest chunks of property owned by most cities and states. We have choices to make about how to use, and share, that real estate. Who decides? Who benefits? Who pays? Transportation policy at all levels of government can be a vehicle to promote public health, sustainability, equitable opportunity, and the economic strength of neighborhoods, cities, and regions. But that will happen only if advocates, experts, and organizers steeped in all these issues bring their knowledge and passion to critical transportation decisions. The upcoming authorization of the most important transportation legislation in the United States, the federal surface transportation bill, makes this a pivotal moment to bring a broad vision for health and equity to transportation policy.
Tra nspor tation in a merica : a new Vision
Underlying this report is a vision of transportation as more than a means to move people and goods, but also as a way to build healthy, opportunity-rich communities. Health is often viewed from an individual perspective. Yet, each resident in a region is both an individual and part of a larger community. Therefore, our vision for healthy, equitable communities is one that extends beyond
The Transportation Prescription
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 1
<
<
T h
e T
ra n
s p
o rt
a ti
o n
P re
s c
ri p
ti o
n
individual outcomes and creates conditions that allow all to reach their full potential. It does not force us to balance one individual against another. It provides the opportunity for everyone to participate in their community, be healthy, and prosper.
Transportation systems are essential to the competitiveness of the nation and the viability of regions. Building America’s Future, a bipartisan coalition of elected officials, views increased transportation investment as a key to the economic growth and job creation needed to strengthen cities and rural communities.4 The American Recovery and Reinvestment Act (ARRA), the nearly $1 trillion stimulus package passed by Congress and signed by President Obama in early 2009, emphasizes transportation investments to revive the ailing economy and rebuild regions.5 The act galvanized advocates to push government agencies to spend the money in ways that promote health, protect the environment, and benefit everyone. Now momentum is building to bring a focus on health and equity to the next version of the federal surface transportation bill.6
Over the past half-century, federal transportation policy has changed the American landscape, physically, socially, and culturally. Beginning with the Federal-Aid Highway Act of 1956 authorizing the Interstate Highway System, the leading transportation priority by far has been what planners call mobility and which became synonymous with the movement of more and more cars and goods farther and faster. Mobility advanced the nation’s growth and prosperity, and it formed our sense of identity as well as our image abroad. The car was more than a machine to get us around; it stood as a symbol of American freedom, ingenuity, and manufacturing prowess.
While some have few or no transportation choices due to limited transportation infrastructure and resources in their communities, many Americans do have the
opportunity to make choices about how to travel and where to go. For these people, the car provides the means to flee the city, buy a quarter-acre patch of suburbia, and drive to their hearts’ content without giving much thought to the disinvested neighborhoods left behind, or the farmland lost to development, or the fossil fuels and other natural resources their lifestyles consumed. Community environments, however, affect the choices individuals make, and public policy molds those environments. As the nation confronts severe economic, environmental, and health challenges as well as the widening gulf between rich and poor, it is becoming clear that we must make different choices as individuals and as a society.
A new framework for transportation policy and planning is emerging. Rather than focus almost exclusively on mobility (and its corollaries, speed and distance), this framework also emphasizes transportation accessibility. In other words, instead of designing transportation systems primarily to move cars and goods, the new approach calls for systems designed to serve people—all people—efficiently, affordably, and safely. This approach prioritizes investments in: (1) public transportation, walking, and bicycling—transportation modes that can promote health, opportunity, environmental quality, and indeed mobility for people who do not have access to cars; and (2) communities with the greatest need for affordable, safe, reliable transportation linkages linkages to jobs, and essential goods and services—chiefly, low- income communities and communities of color.
The goal is to improve transportation for everyone while delivering other important payoffs, including better respiratory and cardiovascular health; improved physical fitness; less emotional stress; cleaner air; quieter streets; fewer traffic injuries and deaths; and greater access to jobs, nutritious foods, pharmacies, clinics, and other essentials for healthy, productive living.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 2
>
>
The Transportation Prescription
This new vision is at the core of a burgeoning movement to shape transportation policy to support work in a number of critical areas, such as climate change, sustainable agriculture, the prevention of chronic diseases, workforce development, and neighborhood revitalization. Advocates and experts in public health, environmental justice, labor, community economic development, food policy, and other fields and disciplines have important roles to play in transportation debates. A broad range of interests working in partnership, can craft innovative, environmentally sound solutions that benefit everyone, rather than plans that reflect the motor vehicle orientation of road engineers and builders. Government transportation agencies and developers—the architects of our transportation systems for decades—must be held accountable for how their investments affect the economic prospects of regions, the health of communities, and the well-being of residents.
This shift in thinking about what transportation policy must achieve and who should drive it stems from a long list of factors. Among them: near-crippling congestion in many metropolitan areas; renewed interest in city living and a hunger for shorter commutes; demographic changes (including the increasing number of people over 65 and immigrants, two groups less likely to drive or own cars); the rise in obesity; the enduring poverty in inner-city and rural communities; the growing understanding of the connections among health, the built environment, and transportation plans; and the increasing frustration among residents and advocates about the limited accountability and inequitable transportation decision-making processes at the state and regional levels which over represent suburban and white male interests.
But the push to reform transportation (along with its cousin, land use planning) has gained urgency in the face of three massive challenges that are upending the status quo of every field and that go to the heart of our love affair with the car: (1) Climate change, with its threat of global ecological upheaval. (2) U.S. dependence
on foreign oil, which carries grave risks for our economy and security. (3) A healthcare system crumbling under the demands of skyrocketing rates of diabetes and other chronic diseases associated with sedentary lifestyles, and astronomical costs. Transporting goods, services, and people accounts for about one-third of greenhouse gas emissions and two-thirds of petroleum consumption in the United States.7 As the National Surface Transportation Policy and Revenue Study Commission noted in its landmark report, Transportation for Tomorrow, the environmental gains we achieve through incremental fixes such as higher fuel-efficiency standards, though important, will be trumped by increases in driving and traffic if we continue on our current policy course.
The good news is that change can happen, and inspiring examples abound. In the rural San Joaquin Valley in California, where public transportation has been virtually nonexistent, a new system of publicly managed vanpools is connecting farm worker families to jobs, schools, and medical services.8
In Chicago’s West Garfield Park, an alliance of residents, activists, and faith-based organizations not only successfully fought the closure of the rail line that linked the neighborhood to downtown; they also transformed a transit stop into an anchor of development of shops, community services, and moderately priced housing.9
In port cities around the country, many groups are working to reduce pollution from ships, locomotives, and trucks, some of the worst emitters of soot and greenhouse gases. In the Los Angeles region—one of a number of regions where the movement of goods represents a significant part of transportation investment and economic activity, and where ports and freeways abut low-income neighborhoods—the Coalition for Clean and Safe Ports has formed an effective alliance of residents, truck drivers, public health experts, environmentalists, environmental justice
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 3
<
<
T h
e T
ra n
s p
o rt
a ti
o n
P re
s c
ri p
ti o
n
activists, unions, immigrant groups, and public officials to push for clean air solutions.10
The authorization of the next federal surface transportation bill presents an immense opportunity to broaden such engagement and to forge an equitable policy response to the unprecedented challenges facing the country. The bill authorizes federal funding for highways, highway safety, public transportation, and bicycling and pedestrian infrastructure for approximately six years.11 It transfers hundreds of billions of dollars from the federal government to states and localities. It also triggers hundreds of billions more in matching state and local spending. The bill marks the largest transportation expenditure in the United States.
But the legislation does more than provide money. It also communicates national policy priorities. Will we build roads on the farthest edges of regions or fix aging roads and bridges in cities and inner-ring suburbs? Will we invest in healthy, green transportation—bicycle lanes, safe sidewalks for walking, clean buses, ridesharing, light rails? Will we ensure that all voices are equitably represented in transportation decision-making processes? And will we include incentives and requirements for affordable housing near public transportation to ensure broad access to the job opportunities and services that transit oriented development stimulates? Or will we spend most of the money as we have for decades: on new and bigger highways with little public accountability? The bill establishes funding categories and requirements and in some cases gives communities and metropolitan regions flexibility to shape strategies to local needs. The new law is a chance to design communities for health, sustainability, and opportunity—and to give all Americans physically active, clean, affordable, convenient, reliable, and safe options to get where they need to go.
W hat Does Hea lthy, Equitable Tra nspor tation Policy look like?
Our current transportation system has many direct health consequences: pollution-related asthma, steep declines in physical activity, and the associated rise in obesity and chronic illnesses are just a few examples. Transportation affects health indirectly by connecting people— or by failing to provide connections—to jobs, medical care, healthy food outlets, and other necessities. For more details on the connections between transportation and health see Chapter 2, Health Effects of Transportation Policy.
The National Surface Transportation Policy and Revenue Study Commission—created by Congress in 2005 to examine the condition and future needs of our network of highways, ports, freight and passenger railroads, and public transportation systems—reached a sobering conclusion: “The nation’s surface transportation network regrettably exacts a terrible toll in lost lives and damaged health.”12 Nowhere is the toll higher than among low-income people and people of color.
Research shows that when properly designed, transportation systems can provide exercise opportunities, improve safety, lower emotional stress, link poor people to opportunity, connect isolated older adults and people with disabilities to crucial services and social supports, and stimulate economic development. Healthy, equitable transportation policy draws on that research to create transportation systems that benefit everyone.
Specifically, healthy, equitable transportation policy:
• Supports the development of accessible, efficient, affordable, and safe alternatives to car travel, and especially to driving solo. These alternatives enable everyone to walk more, travel by bicycle, and use public
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 4
>
>
transportation more—in other words, to get around in ways that improve health, expand access to opportunity, and reduce toxic pollutants and greenhouse gas emissions.
• Works hand in hand with sustainable land use planning. Together, they encourage and support high-density, mixed-use, mixed- income metropolitan development and affordable housing with good access to transportation options. Together, they focus, particularly, on underserved and economically isolated communities.
• Recognizes that income is important to health, and that good transportation has an impact on family income. Healthy, equitable transportation policy support systems that connect all people, especially low-income and underserved communities, to employment and other opportunities. It also encourages hiring low-income residents of color for well- paying jobs in transportation construction, maintenance, and service.
• Understands the importance of ensuring equal representation. All community members, regardless of race, gender or geographical location should be equitably represented and involved in making decisions which impact their communities, their infrastructure and their options for travel.
• Recognizes that access to healthy foods is integral to good health and that transportation systems are integral to food production and distribution. Healthy, equitable transportation policy explicitly addresses food access issues, including transportation to grocery stores and food transport practices.
This summary draws on the six thematic chapters in this book authored by academics and advocates working at the intersection of transportation, health, and equity. Each chapter describes innovative transportation and land use policies, strategies, and programs built on
a foundation of equity and sustainability. Three chapters in this collection address transportation options:
• Todd Litman, M.E.S., founder and executive director of the Victoria Transport Policy Institute in British Columbia, identifies numerous economic, social, and environmental benefits that can result from public transportation improvements. Among them: reduced traffic crashes, improved physical fitness and health, energy conservation, reduced pollution emissions, increased community livability, increased affordability, consumer savings, economic development, and expanded opportunity. Litman contends that improving public transportation is one of the most cost- effective ways to improve public health, and better health is one of the most significant potential benefits of public transportation improvements. He identifies policy and planning reforms to create a more diverse and efficient transportation system. He recommends developing a strategic vision of high-quality public transportation services, with supportive land use policies to provide basic mobility to people who are socially isolated, economically disadvantaged, or physically disabled, as well as to attract “discretionary” travelers, or people who would otherwise drive for a particular trip.
• Susan Handy, Ph.D., director of the Sustainable Development Center at the University of California at Davis, argues that increasing walking and bicycling while assuring safety, particularly for low-income families, children, and older adults, is an important goal for federal transportation policy. Walking and bicycling, or “active travel,” are low-cost, physically active, and environmentally clean alternatives to driving, yet they represent fewer than 10 percent of all trips in the United States. In addition to expanding specialized programs for active travel, the federal government should assist, enable, encourage, and, in some instances,
The Transportation Prescription
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 5
<
<
T h
e T
ra n
s p
o rt
a ti
o n
P re
s c
ri p
ti o
n
require state, regional, and local governments to address pedestrian and bicycling needs.
• Catherine L. Ross, Ph.D., the Harry West Chair and director of the Center for Quality Growth and Regional Development at Georgia Institute of Technology, argues that roadways are more than transport routes; they are also our primary spaces for civic, social, and commercial enterprise. Roadways—highways in particular—receive the largest share of federal transportation dollars by far. Federal policy has historically emphasized highways designed to move large numbers of cars and freight vehicles at high speeds. Ross argues for greater investments in roadways that integrate physical activity, enrich social interaction, increase safety, and provide transportation linkages in underserved communities. She urges policymakers and others to consider expanded assessments of the effects of roadways on health, through the use of methodologies similar to health impact assessment (HIA).13
The remaining papers offer transportation policy perspectives in key areas that have a significant impact on public health and equity:
• Todd Swanstrom, Ph.D., the E. Desmond Lee Professor of Community Collaboration and Public Policy Administration at the University of Missouri, St. Louis, makes the case that federal transportation policy can and should address economic development, particularly in communities left behind by decades of transportation planning that favored car travel and encouraged sprawl. Targeted transportation investment can promote economic opportunity and reduce health disparities by (1) improving transportation linkages between housing and employment hubs and between residential neighborhoods and clinics, pharmacies, and grocery stores; and (2) encouraging affordable, high-density, mixed-use transit
oriented development14; and (3) creating workforce strategies to ensure that jobs in the large, growing transportation sector are open to all, including minority and women workers and contractors. Swanstrom also asserts that while the goals of equity and environmental sustainability are not mutually exclusive, policymakers and advocates must address the short-term needs of low-income families who live in places where driving is essential.
• Kami Pothukuchi, Ph.D., associate professor of urban planning at Wayne State University, and Richard Wallace, M.S., senior project manager at the Center for Automotive Research, argue that federal transportation policy should seek to increase access to healthy foods. Today’s transportation networks make large quantities of foods from around the nation and the globe readily available for many Americans, but industrialized agriculture and the concentrated structure of food retail have negative health and environmental consequences for low-income communities, especially people of color, inner-city and rural residents, and immigrant farm workers. For example, urban and rural communities often have fewer and smaller supermarkets than suburban communities (if they have any at all) as well as more limited selections of healthy foods. As a result, residents eat fewer fruits and vegetables and have higher rates of diet-related illnesses. In addition, long- distance food hauling has a disproportionate impact on the air quality and noise levels in poor and minority communities along freight routes. Although food access falls outside the traditional realm of transportation policy, improved public transportation, transit oriented development, and cleaner methods to move freight can increase access to healthy foods in underserved communities, reduce air and noise pollution, and foster local, sustainable agri-food systems.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 6
>
>
• Larry Cohen, M.S.W., Leslie Mikkelsen, R.D., M.P.H, and Janani Srikantharajah, B.A., of Prevention Institute argue that traffic crashes are preventable and that federal transportation policy must make safety for all travelers a priority. Traffic crashes rank as the leading cause of death for people ages one to 34 and contribute to unnecessary human, social, and economic costs. Resources should be directed to communities with the least infrastructure to support safe walking, bicycling, and public transportation use and continue to support effective vehicle safety and occupant protection strategies. Traffic safety is an important strategy not only to reduce injuries and death but also to encourage physical activity, improve air quality, and increase transportation accessibility.
The federa l Tra nspor tation legacy a nd Cha llenges a hea d
Transportation in America is a federal system, not a centralized, national system. Federal policy plays a critical role, not by dictating practices but by enabling and encouraging innovation by states, regional transportation organizations, transit operators, and other agencies. This happens in several ways.
First, the federal government sends billions of dollars for transportation to states and localities. For example, the American Recovery and Reinvestment Act provides nearly $50 billion to build and repair roads, bridges, railways, and ports. The current surface transportation bill, SAFETEA-LU (Safe, Accountable, Flexible, Efficient Transportation Equity Act: A Legacy for Users), set to expire in September 2009, guaranteed $244.1 billion over six years. These dollars, in turn, leverage direct infrastructure investments by state governments, local governments, and private investors.
Second, the policies and requirements embedded in federal transportation programs influence state and local land use decisions and transportation priorities. Many observers contend that transportation stands as one of the biggest policy successes in United States history. The Federal-Aid Highway Act of 1956 and its progeny promoted mobility, which contributed mightily to American growth and prosperity. However, many advocates take a more nuanced view of the federal legacy. They point to the health, equity, and environmental consequences of an ethic that held the faster, the farther, the better, as well as the consequences of policies focused almost wholly on car and truck travel, with little accountability to goals beyond mobility.
Either way, the current transport system is no longer sustainable or fixable by incremental changes such as pilot projects, encouragements, and small incentives. As the National Surface Transportation Policy and Revenue Study Commission, created by SAFETEA-LU, wrote in its final report to Congress: “The strong and dynamic American surface transportation system is becoming a thing of the past.”
At 300 million people, the nation’s population has doubled since the creation of the Interstate Highway System. We will number 420 million by 2050. “Congestion was once just a nuisance. Today gridlock is a way of life,” the commission’s report said. Growing transportation demand threatens to dwarf regulatory and legislative efforts to mitigate its health and environmental consequences. Increases in total vehicular mileage have all but wiped out the gains achieved through hard-won regulations on fuel efficiency and emissions control. Expansion of freeways cannot get us out of these problems; it will only make them worse. The more we have expanded highways, the more traffic we have created. The United States needs multi-modal systems with public
The Transportation Prescription
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 7
<
<
T h
e T
ra n
s p
o rt
a ti
o n
P re
s c
ri p
ti o
n
transportation that efficiently serves a large segment of the population, using existing streets and highways.
The Intermodal Transportation Efficiency Act (ISTEA), the 1991 version of the federal surface transportation bill, was supposed to lead us there. The act incorporated significant policy change. Since then, the stated goal of federal transportation policy has been to expand access and improve efficiency through an interconnected multi-modal system that supports highways, public transportation, walking, and biking. This goal has yet to be achieved. Funding mechanisms and formulas have continued to favor highway construction and car travel. For example, the allocation formula for the Surface Transportation Program (STP), the largest program within the federal bill, rewards states that consume more gas, have more miles of highway, and have residents who drive a lot.15 Alternatives to driving remain underinvested. Approximately 80 percent of the surface transportation bill is allocated for distribution through the Federal Highway Administration for mostly highway programs, while less than 20 percent goes to the Federal Transit Agency for public transportation. Other modes of travel constitute a minute amount of spending in comparison to highways and public transportation.
Case in point: walking is the only travel mode that has not had significant declines in casualties in 40 years. Yet only a tiny share of transportation funding goes to infrastructure initiatives that would make walking and biking safer. Walking and bicycling accounted for 8.6 percent of all trips in 2001 but 12 percent of traffic deaths.16
Another case in point: operating costs for public transportation systems present a huge challenge for many communities. Yet federal transportation investment is focused on capital projects. For example, cities with 200,000 people or more may not use grants from the
U.S. Department of Transportation’s main public transportation programs for transit operating costs.17 In the face of budget shortfalls, local and regional transportation agencies throughout the country have cut service, hiked fares, and deferred maintenance—arguably at a time when people need affordable, reliable links to jobs more than ever.
While federal policy plays a significant role in shaping transportation systems, states and metropolitan regions are also critical agents of change. The new surface transportation bill offers an opportunity to increase support, encouragement, and pressure for integrating land use and transportation planning to promote balanced regional growth, equitable economic opportunity, and healthy communities for all.
a foundation for 21st- Centur y Tra nspor tation Policy
Healthy, equitable transportation policy is grounded in four principles. These may also serve as benchmarks to assess the impacts of transportation plans on public health, equity, and environmental quality:
1. Develop transportation policies and plans that support health, equity, and environmental quality. Federal, state, and local transportation policies should be aligned with the top health and environmental goals of federal departments and agencies. For example, transportation policies should be aligned with the Department of Health and Human Services’ strategic goals to promote health equity and foster the economic and social well-being of individuals, families, and communities. Transportation policies should also support the CDC’s commitment to eliminate health disparities and to promote its “healthy people in healthy places” goals.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 8
>
>
2. Prioritize transportation investments in distressed regions, low-income neighborhoods, and communities of color. Federal, state, and local transportation agencies should emphasize projects that will revitalize the economy of struggling communities, lower health disparities, and will connect vulnerable populations to jobs, business opportunities, healthy food outlets, medical services, and other necessities. Government agencies must ensure that these projects are financially sustainable by providing adequate funding for maintenance and operations. The jobs associated with transportation construction, maintenance, and service should be available to low- income people and communities of color.
3. Emphasize accessibility, instead of simply mobility, in transportation policies and programs at all levels of government as well as across sectors and policy silos. Transportation systems should give communities wider access to all the things that are necessary for a good life, not to move people faster and farther. The definition of access must also include affordability. If transportation is physically accessible, yet unaffordable, it is not truly accessible.
4. Ensure transparency, accountability, and meaningful participation by residents, advocates with diverse interests, and experts from different fields. State and regional transportation officials and private developers must engage new partners in decision making and provide the data, training, and resources to allow full, informed participation by the people affected most by decisions and investments. Voices and expertise from local communities, public health, environmental justice, community development, and other arenas can help ensure that transportation plans respond to local needs and deliver health, environmental, and economic benefits broadly.
Policy a nd Prog ra m Priorities to improve Hea lth a nd Equity
Government at all levels must consider the health and equity impacts of transportation investments at the beginning of decision-making processes. Public and private transportation investments must be designed to promote health rather than to erode it. The following recommendations can help policymakers and planners achieve these ends:
1. Prioritize investments in public transportation, including regional systems that connect housing and jobs as well as local services that improve access to healthy foods, medical care, and other basic services. Investments should include capital costs as well as costs for maintenance and operations. Because older diesel buses have high emission rates and since bus depots and other facilities are often concentrated in low-income and minority neighborhoods, policies must be in place to ensure that expanded public transportation does not lead to increased exposure to pollutants in these same communities.
2. Prioritize investments in bicycle and pedestrian infrastructure to make walking and biking safer and more convenient. Strategies include complete streets designed with all users in mind, not just drivers; traffic-calming measures; and safe routes to transit and Safe Routes to Schools programs, which create infrastructure and programming to support safe walking and bicycling to bus stops, rail stations, and schools. Targeted infrastructure investments should also support walking and bicycling in rural communities by, for example, improving road shoulders and building trails to town centers.
The Transportation Prescription
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 9
<
<
T h
e T
ra n
s p
o rt
a ti
o n
P re
s c
ri p
ti o
n
3. Encourage equitable transit oriented development by creating incentives for integrated land use and transportation planning. Transit oriented development must emphasize affordability and accessibility. It also must incorporate affordable housing and commercial properties that provide jobs, services, and essential goods near people’s homes. Because people of all income levels desire walkable neighborhoods and shorter commutes, displacement of longtime neighborhood residents can be an unintended consequence of transit oriented development. Policymakers must ensure that the local residents guide planning and development and that equity is a goal from day one.
4. Create incentives and accountability measures to ensure that transportation plans account for their impacts on health, safety, and equity. New projects must be held accountable for better results. Government investment should support the creation of tools that more sensitively and accurately measure walking and bicycling practices and improved outcomes. Health impact assessment is an emerging methodology to evaluate the effects of policies, programs, and plans on the health of a population and should be considered an important tool. People should also have the right to sue under Title VI of the Civil Rights Act of 1964 if they suffer disparate impacts from federal transportation investments, and the U.S. Department of Transportation should have the power to withhold dollars if investments are not made equitably.18
5. Give state, regional, and local government agencies and organizations more flexibility to move dollars among funding categories and to target spending to meet local needs. Greater flexibility would give communities more leeway to fund walking, bicycling, and public transportation programs. It would
also enable communities to invest in fixing, maintaining, and operating local bus and rail systems. Flexibility should be strongly tied to new standards for accountability, transparency, and inclusion which ensure all people impacted by transportation decisions are equitably represented in the decision- making process.
6. Prioritize transportation investments in communities with high unemployment and poverty rates to stimulate economic growth and provide access to jobs. The American Recovery and Reinvestment Act (ARRA) has language to direct resources to struggling and disinvested communities. The new version of the surface transportation bill should include similar language and expand on this commitment by creating strong accountability and enforcement measures tied to achieving equitable economic benefits.
7. Make sure that jobs and contracts created by federal transportation investments reach low-income people and communities of color. A Sense of Congress amendment to SAFETEAU-LU, passed in 2005, encourages local hiring provisions for highway construction projects. Some projects aim for 30 percent of workforce hours to be filled by employees who live in the community. Local hiring should be made a requirement, not just encouraged. It should also be expanded beyond highway projects to include public and mass transit development. Capital investments should also fund workforce development programs to train local residents for jobs in the transportation sector.19
8. Support the development of cleaner bus and truck fleets and invest in freight rail infrastructure to reduce greenhouse gas emissions, improve local air quality, promote health, and foster energy independence.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 2 0
>
>
9. Advance safety for all travelers, with particular emphasis on those at the highest risk of car injuries and death. Investments should continue advancing known vehicle safety and occupant-protection strategies as well as roadway and community design modifications to protect the safety of pedestrians, bicyclists, drivers, and passengers.
10. Support policies and programs that increase access to healthy foods. Promote public-private van and bus systems to shuttle customers to grocery stores. Expand weekend bus service to connect low-income neighborhoods to supermarkets and other food outlets. Invest in safe and affordable transportation for farm and food production workers. Promote sustainable modes of transporting foods from farms to stores as well as policies to increase the viability of local and regional farming.
11. Give low-income rural communities greater access to public transportation funds from the surface transportation bill providing the opportunity to access employment and education opportunities. Low-density and long travel distances make developing and operating conventional bus and rail systems financially challenging. Federal public transportation dollars should support economically efficient innovations, such as vanpools and voucher programs.
Conclusion
The authorization of the next federal surface transportation bill can be a starting point for creating many changes Americans say they want: better health, cleaner air, more time with our families, opportunities to connect with our neighbors. The new legislation can also mark an important step toward building a society in which everyone can participate and prosper, and no community is left behind.
Change will not come easily. The car culture has deep roots in America. The interest groups supporting highway investment are powerful and well-funded. But advocates and grass-roots activists around the country have demonstrated that change can happen. They have successfully fought for cleaner buses and for public transportation in communities that never had it. They have transformed train stations into centers of vibrant community development in disinvested neighborhoods. They have pressured local officials and supermarket operators to provide free bus rides so families can shop for food.
Now is the time to tap into that kind of energy and lift successes like these to the level of federal policy. Leaders, experts, and advocates from many spheres—public health, environmental justice, food policy, agriculture, labor, equity, community economic development, business, and government—must join in partnership to push for broad reform. Collectively, we can gain power and build political support for creating transportation systems that address the big challenges we face and that nourish healthy communities throughout our nation.
The Transportation Prescription
Health Effects of ch. 1 Transportation Policy JUDITH BELL, M.P.A. President, PolicyLink
LARRY COHEN, M.S.W. Founder and Executive Director, Prevention Institute
ABSTRACT >> There is a deep and evolving knowledge base about the links between transportation and health. Research shows that when properly designed, transportation systems can provide exercise opportunities, improve safety, lower emotional stress, link poor people to opportunity, connect isolated older adults and people with disabilities to crucial services and social supports, and stimulate economic development. Conventional auto mobility-focused planning by local, regional, and state transportation agencies generally overlooks or undervalues the impacts of transportation investments on health and equity.
This chapter provides an overview of the impacts of transportation on health. Subsequent chapters on transportation options and key issues provide further detail.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 2 2
>
>
C h
a p
te r
1
introduction
Our current transportation system has many direct health consequences: pollution-related asthma, steep declines in physical activity, and the associated rise in obesity and chronic illnesses are just a few examples. Transportation affects health indirectly by connecting people— or by failing to provide connections—to jobs, medical care, healthy food outlets, and other necessities. The National Surface Transportation Policy and Revenue Study Commission—created by Congress in 2005 to examine the condition and future needs of our network of highways, ports, freight and passenger railroads, and public transportation systems—reached a sobering conclusion: “The nation’s surface transportation network regrettably exacts a terrible toll in lost lives and damaged health.”1 Nowhere is the toll higher than among low- income people and people of color.
Direct Hea lth Ef fects
Pollution
Pollutants from cars, buses, and trucks are associated with impaired lung development and function in infants2 and children,3 and with lung cancer,4 heart disease, respiratory illness,5 and premature death.6 Long-term exposure to pollution from traffic may be as significant a threat for premature death as traffic crashes and obesity.7 In California alone, pollution is a factor in an estimated 8,800 premature deaths a year.8
The main culprits are fine particulate matter, including diesel exhaust particles; ground-level ozone, a toxic component of smog formed when tailpipe emissions from cars and trucks react with sunlight and oxygen; and nitrogen oxide (NOx), which contributes to the formation of ozone and smog. The health risks are exacerbated by transportation patterns that often embed heavy traffic and diesel-spewing facilities in poor and predominantly minority neighborhoods. The American Lung Association has found that 61.3 percent of African American children, 67.7 percent of Asian American children, and 69.2 percent of Latino children live in areas that exceed air-quality standards for ozone, compared with 50.8 percent of white children.9 Ground-level ozone, a gas, can chemically burn the lining of the respiratory tract.
Air pollution is also “one of the most underappreciated” triggers of asthma attacks, according to the Centers for Disease Control and Prevention (CDC).10 More than 20 million Americans—roughly seven percent of adults and nearly nine percent of all children—have asthma. In poor and minority communities, the rates are considerably higher. For example, in Harlem and Washington Heights in northern Manhattan, home to mostly low-income African American and Latino residents, one in four children suffers from the disease.11 Research shows that air pollution can trigger the wheezing, coughing, and gasping for breath
Health Effects of Transportation Policy
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 2 3
<
<
H e
a lt
h E
ff e
c ts
o f
Tr a
n s
p o
rt a
ti o
n P
o li
c y
that signal an attack in people with asthma. But a study in 10 Southern California cities raises the troubling possibility that pollution can also lead to the onset of the disease. The study found that the closer children live to a freeway, the more likely they are to develop asthma.12
Environmental justice activists have called attention for years to the connections among pollution, illness, and transportation policy— and the burden on communities of color. For instance, in the mid-1990s, West Harlem Environmental Action (WE ACT) used mapping, air monitoring, and resident surveys to show that the neighborhood’s asthma rates were linked to its dubious status as the diesel capital of New York City. When WE ACT began work on the issue, Harlem housed six of the city’s eight bus depots and 650 Port Authority buses. The group played an important role in getting the city to convert buses to clean fuel.13
Pollution from freight transport is another big concern around the country. To meet America’s insatiable demand for goods, ports and highways are continually expanding to accommodate more ships, locomotives, and trucks. Ports frequently border low-income and minority neighborhoods, and highways often run through them. The upshot: some of the worst emitters of fine particles, soot, and greenhouse gases (GHGs) are a growing presence in already vulnerable communities.
Climate Change
GHGs are not pollutants in the classical sense. They cause the atmospheric changes and resulting climate disruptions that are projected to alter the natural and built environments on which society relies.14 The health risks come largely from those environmental alterations. In a major shift in federal policy, the Environmental Protection Agency in April 2009 adopted the position that greenhouse gases pose a danger to human health and welfare. A few weeks later, the Climate Change and Health Protection and Promotion Act, H.R. 2323, was introduced
in the House of Representatives.15 The bill would direct the Department of Health and Human Services to develop a national strategic action plan to prepare for and respond to the health effects of climate change.
Researchers are just beginning to assess the specific health dangers in the United States; most of the published data to date come from abroad. A recent report predicts that kidney stones, linked to dehydration, may increase by as much as 30 percent in the driest regions of the United States.16 So far, however, there are more questions than answers. How will less rainfall affect the potential for waterborne diseases? Food supplies? Food prices? How will extreme weather conditions such as heat waves or hurricanes affect mental health? Physical activity? Population displacement?
Scientists believe that climate change could exacerbate a number of current health problems, including heat-related deaths, diarrheal diseases, allergies, and asthma.17 Those already at highest risk—the poor, minorities, children, and older adults—will be even more vulnerable. Policy neglect would compound the problems. Hurricane Katrina revealed, to a horrified public, the disastrous results that can occur when nature (the sort of extreme storm that experts expect to occur more frequently as the earth’s temperature changes) combines with government disregard (in this case, the poorly maintained levees that failed to protect New Orleans from catastrophic flooding) as well as resource inequities (the lack of transportation, which made evacuation impossible for thousands of people).
The urgent need to reduce GHGs has catapulted transportation policy into the limelight. The United States has only about five percent of the world’s population but contributes nearly 25 percent of GHGs, mainly because of fossil fuel consumption, motor vehicle emissions, and industrial agricultural practices (which themselves are promoted by our transportation system).
ch. 1
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 2 4
>
>
C h
a p
te r
1
Improving vehicle technology, while important, is not enough. Americans need to drive less. That will happen only if walking, bicycling, and public transportation become feasible, efficient alternatives to driving in many more communities, and if land use patterns are changed so people no longer have to jump in the car for every trip.
Physical Activity
Sixty percent of adults in the United States do not meet recommended levels of physical activity, and 25 percent are completely sedentary.18 African Americans and Latinos are less likely than whites to get enough daily physical activity.19 The links between physical activity and health are well established. Sedentary lifestyles are estimated to contribute to as many as 255,000 deaths each year.20 Many children and teens are already at risk for heart disease and type 2 diabetes, once considered “adult” ailments. Today’s youth may turn out to be the first generation in modern history to live shorter lives than their parents.21
Physical inactivity is an important factor in the rising rates of obesity and chronic disease—and transportation practices strongly influence physical activity habits. The more time a person spends in a car, the more likely he or she is to be overweight. Conversely, higher rates of walking and bicycling are associated with lower rates of obesity. A 2004 study found that every additional hour spent in a car is associated with a six percent increase in the likelihood of obesity, and every additional kilometer walked is associated with a 4.8 percent reduction.22
There are many ways to be physically active, but quite a few require time, skill, and money. Walking and bicycling not only for recreation but also for transportation are the most practical ways to improve fitness. They are often the only viable option for low-income residents who live in neighborhoods without parks, who cannot
afford gym memberships, and who do not have the luxury of leisure time.
People who use public transportation tend to walk to and from bus stops and train stations, increasing their likelihood of meeting physical activity recommendations.23 Residents of compact neighborhoods walk, bike, and use public transportation more than residents of spread-out communities, and they have lower rates of obesity.
Mental Health
Rush-hour gridlock, long waits for the bus, and arduous commutes are stressful. They take time away from family, friends, and the activities that provide emotional sustenance: hobbies, religion, sports, clubs, civic engagement, and volunteer commitments. Every 10 minutes spent commuting is associated with a 10 percent drop in the time spent traveling for social purposes.24
Many people find commuting by high-quality public transportation to be less stressful than commuting by car. As we discuss below, the financial costs associated with long commutes
Health Effects of Transportation Policy
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 2 5
<
<
H e
a lt
h E
ff e
c ts
o f
Tr a
n s
p o
rt a
ti o
n P
o li
c y
ch. 1
exacerbate the stress, particularly in low-income households.
Safety
Traffic crashes are a leading cause of death and injury for Americans in the prime of life.25 In 2000, motor vehicle crashes cost $230.6 billion in medical costs, property damages, lost worker productivity, travel delays, and other expenses.26 That figure equals about half of all spending on public education from kindergarten through 12th grade.
Native Americans die in traffic crashes at more than 1.5 times the rate of other racial groups.27 African Americans drive less than whites but die at higher rates in car crashes. Walking, too, is also more dangerous in communities of color. CDC data in the mid-1990s revealed that the pedestrian death rate for Latino males in the Atlanta metropolitan area was six times greater than for whites.28 African Americans make up 12 percent of the U.S. population but account for 20 percent of pedestrian deaths.29
Inequitable transportation policies and resources contribute to these disparities. Low- income people and people of color have fewer resources to buy products that improve safety, such as late-model cars and new child safety seats. In underinvested neighborhoods, poorly designed streets, neglected road maintenance, inadequate lighting, limited sidewalks, and minimal traffic enforcement place residents at higher risk of injury.
Safety is also a huge concern for older adults—the fastest-growing segment of the population—and for rural residents. Driving skills decline with age, and frailty makes older adults especially vulnerable in a collision.30 They are more likely to be killed or injured in a crash of a given severity than any other age group.31 Older adults also walk slower and are more susceptible to pedestrian injuries.
Although less than a quarter of all driving in the United States takes place in rural settings,32 more than half of all motor vehicle crashes occur there.33
The more we drive, the more likely we are to get hurt or die in a crash; there is a strong positive relationship between per capita vehicle miles traveled and traffic casualty rates.34 Communities with high annual mileage tend to have higher traffic death rates than communities where people drive less. Passengers on buses, light rail, and commuter rail have about one- tenth the traffic death rate as people in cars.
Investments in public transportation and walking and bicycling infrastructure can reduce injuries and deaths. Contrary to popular belief that more walkers and cyclists lead to more casualties, greater numbers of walkers and bicyclists actually decrease the risks.35
indirect Hea lth Ef fects
Transportation is a lifeline. We depend on it to get to work, school, the doctor’s office, the bank, the supermarket, the gym, or a friend’s house. People without reliable, efficient, affordable ways to get around are cut off from jobs, social connections, and essential services. Access to transportation, to economic and social opportunity, and to resources for healthy living are inextricably linked. Gaps in all three areas feed on one another in complex ways. Policy reforms that put health equity objectives at the center of transportation planning and funding decisions can reduce these inequities.
Transportation, Income, and Health
As housing and jobs have moved farther apart, the distance has created employment barriers for anyone without unlimited ability to drive. Nineteen percent of African Americans and 13.7 percent of Latinos lack access to automobiles,
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 2 6
>
>
C h
a p
te r
1
compared with 4.6 percent of whites. Poverty complicates the problem: 33 percent of poor African Americans and 25 percent of poor Latinos lack automobile access, compared with 12.1 percent of poor whites.36 Cars owned by low-income people tend to be older, less reliable, and less fuel-efficient. This makes commuting to work unpredictable and more expensive, at best.
Income is an important determinant of health.37 The association between poverty and poor health is well documented. Jobs with good wages, including those in the transportation sector, are essential to sustaining health.
Transportation impacts not only family earnings but also expenses. The cost of getting around takes a significant bite out of household budgets. The general standard holds that a family should spend no more than 20 percent of income on transportation, or the costs will eat into other necessities, such as nutritious foods and medical care.38 The average family in the United States spends about 18 percent of after-tax income on transportation, but this varies significantly by income and geography. For example, low-wage households (earning $20,000 to $35,000) living far from employment centers spend 37 percent of their incomes on transportation.39 In neighborhoods well served by public transportation, families spend an average of nine percent.40
Older Adults and People with Disabilities
More than one in five Americans ages 65 and older do not drive because of poor health or eyesight, limited physical or mental abilities, concerns about safety, or because they have no car. More than half of nondrivers, or 3.6 million Americans, stay home on any given day—and more than half of that group, or 1.9 million, have disabilities.41 Isolation is especially acute in rural communities, sprawling suburbs, and black and Latino communities. Compared with
older drivers, older nondrivers take 15 percent fewer trips to the doctor; 59 percent fewer trips to shops and restaurants; and 65 percent fewer trips for family, social, and religious activities.42
When affordable, high-quality public transportation and safe, walkable streets are available, older adults take advantage of them. More than half of older adults make walking a regular activity. More than half of older nondrivers in dense communities use public transportation at least occasionally, compared with one in 20 in spread-out communities.43
The Americans with Disabilities Act (ADA) of 1990 significantly expanded transportation options for people with disabilities. ADA required public bus and rail operators to provide accommodations, such as lifts and ramps, to enable people in wheelchairs to ride. But street design in most communities makes traveling to and from bus stops challenging—and often unsafe—for people with disabilities. Paratransit systems, which use vans or shared taxis to transport people door-to-door, are helpful, but many systems are stretched thin and require appointments well in advance.
Conclusion
Transportation and health: until recently, policymakers, government officials, advocates, and indeed, most Americans thought of these as distinct realms. But research shows that how we get around and how we transport goods and services have a profound impact on individual, community, and public health. Further, inequities in transportation resources contribute to the pronounced health disparities in the United States and to the growing income gap between the affluent and the poor. An overarching transportation policy that does not seriously consider public health, environmental quality, and equitable access will inevitably damage all three. Health and equity must be at the center of transportation planning and investments.
Health Effects of Transportation Policy
p
g . 2 7
<
<
Transportation Authorization 101: ch. 2 A Backgrounder SUSAN POLAN, Ph.D. Associate Executive Director, American Public Health Association
TRACY KOLIAN, M.P.H. Senior Health Policy Analyst, American Public Health Association
SHIREEN MALEKAFZALI, M.P.H. Senior Associate, PolicyLink
ABSTRACT >> For most people, federal policy seems removed from day-to-day life in their communities. But the federal surface transportation bill is a critical determinant of how our communities are formed, how they grow, and what types of transportation choices—if any— are available to us. Highways, rail systems, sidewalks, biking and walking paths, transit oriented development—all of these, and more—are shaped in large part by the federal transportation authorization. And federal transportation dollars are a major source of funding for states and metropolitan areas as they build new infrastructure and maintain existing transportation systems.
This publication discusses the connections between transportation and health; the analysis and the recommendations focus on the upcoming authorization of the federal surface transportation bill as a key opportunity for promoting health and equity. This section orients readers to the bill by briefly describing what the legislation includes, how it is authorized, and by whom—naming key committees and policymakers. This chapter also explains how federal funding is allocated to states and metropolitan regions to pay for public transportation systems, highways, bridges, sidewalks, bike paths, and other transportation projects in our communities.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 2 8
>
>
C h
a p
te r
2
Over v iew
Approximately every five years, Congress passes a new surface transportation bill and authorizes the U.S. Department of Transportation (DOT) to implement it. This bill sets federal transportation policy and designates transportation funding to states directly through formulas or through competitive grant programs for which states can apply. The programs and projects in the bill are funded through the Highway Trust Fund, which draws on a nationwide 18-cent per gallon tax on gas. The current law, passed in 2005, is called the Safe, Accountable, Flexible, Efficient, Transportation Equity: A Legacy for Users, or SAFETEA-LU. It represents a $244.1 billion federal investment in transportation infrastructure. SAFETEA-LU is set to expire September 30, 2009, and Congress must authorize a new bill. A new bill may also be postponed through extension of SAFETEA-LU until lawmakers are prepared to pass a new bill.
This report intentionally uses the term authorization and not reauthorization when referring to the process of developing a new surface transportation bill. “Authorization” symbolizes the significant reform necessary in the existing bill to meet current and future needs of a changing and diverse U.S. population. Reform is long overdue. With imperatives such as climate change, growing rates of chronic diseases and health disparities, increasing poverty rates, and an economic downturn, transportation policy must connect with national priorities, consider its impacts on these critical issues, and help to significantly change them. A reauthorization of the current bill will not address these challenges. A new federal transportation policy is needed to align its goals and actions to national priorities, address critical issues facing Americans, and ensure accountability and equity.
SAFETEA-LU includes a whopping 108 programs, each with distinct funding allocations and eligible activities for which funding may be used. For example, the eligible activities for one
program, the Safe Routes to School Program, includes activities related to the planning, design, and construction of infrastructure projects that improve the ability of students to walk and bike to school; states can use a portion of the funds for noninfrastructure-related activities to encourage walking and bicycling to school. The overall goal of the program is to enable and encourage walking and bicycling to school in a safe and appealing manner.1
An authorization establishes programs and sets ground rules under which the programs operate including the amount of funding available, how the funds are distributed, the length of time the funds can be used, and a list of eligible activities. Subsequent authorizations can change programs, eliminate programs, and create programs.
In the past several months, Congress and the DOT have been preparing to introduce a new federal surface transportation bill. Advocates have been gearing up to make sure this immense investment reflects the needs of all Americans. Right now is a crucial time to engage in transportation policy and to work to ensure that the policies and funding levels set for the next several years are aligned with important goals and ideals—health, safety, sustainability, economic opportunity, and equity.
The new bill could have enormous impacts on the funding available for various modes of travel as well as specific projects, thus influencing the decisions transportation planners and engineers make at the local level. For example, a region could expand a roadway instead of creating a subway system because there is more federal funding readily available for the highway project and the project evaluation and approval process for major transit investments is substantially more burdensome than the highway process. The federal pot of money for highway projects is far bigger than the pot available for public transportation. Currently, approximately 80 percent of federal transportation dollars go to the Federal Highway Administration (FHWA) as part of highway programs, while merely
Transportation Authorization 101
ch. 2
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 2 9
<
<
Tr a
n s
p o
rt a
ti o
n A
u th
o ri
za ti
o n
1 0
1
one-fifth, or 20 percent, goes to the Federal Transit Agency (FTA) to be used for public transportation infrastructure. Only a very small portion of overall transportation funds are used for walking and biking infrastructure or other programs and most are administered through FHWA and FTA.
The first federal surface transportation bill, the Federal Aid Highway Act (popularly known as the National Interstate Defense Highways Act), was passed in 1956 as a means to fund a massive interstate highway system from coast to coast. Since the inception of the federal surface transportation bill, it has focused on highways as the key mode of travel. The 1991 surface transportation bill, the Intermodal Surface Transportation Efficiency Act (ISTEA), critically shifted the focus of federal transportation policy. In addition to funding traditional highway and transit programs, ISTEA included money for projects aimed at improving air quality, reducing congestion, and providing pedestrian and biking infrastructure. It launched the beginning of a more environmentally sensitive and multi-modal approach to transportation planning.2 While these laws made great strides at the time, we are far from implementing a truly multi-modal system where public transportation, walking, and biking are on equal footing with highways.
The next surface transportation bill must set about the urgent task of repairing and maintaining our transportation assets, building new transportation connections, and making our current system work more efficiently and safely to create complete and healthy communities that address the transportation needs of all communities. Modern and affordable public transportation, safe places to walk and bicycle, smarter highways that use technology to better manage congestion, land use policies that reduce travel demand by locating more affordable housing near jobs and services, and long-distance rail networks all have the potential to help us reduce our dependency on foreign oil, slow climate change, improve social equity, enhance public health, and fashion a vibrant new economy.
The authorization Process
The U.S. Senate and the U.S. House each develops a transportation bill and then reconciles their differences before presenting a final bill to the president. In the House, the Transportation and Infrastructure Committee (T&I Committee), chaired by Rep. James Oberstar (D-MN), has primary jurisdiction over the bill. At time of printing, Chairman Oberstar has been working hard to write and pass a new bill with limited to no extensions to the current bill, SAFETEA-LU. Since SAFETEA-LU expires on September 30, 2009, some form of extension is likely to take place though it still remains unclear whether it will be a short extension or a longer 18-month extension as suggested by the administration.
The House T&I Committee has two counterparts in the Senate, where the jurisdiction is slightly more diffused. The Senate Environment and Public Works Committee (EPW Committee), chaired by Sen. Barbara Boxer (D-CA), has primary jurisdiction over the highway portion of the transportation bill, while the Senate Banking, Housing and Urban Affairs Committee (Banking Committee), chaired by Sen. Christopher Dodd (D-CT), has primary jurisdiction over public transportation portions. Both T&I, EPW and Banking have subcommittees focused on surface transportation that must develop and pass the first draft of the bill out of the subcommittees: the Highway and Transit Subcommittee of T&I, chaired by Rep. Peter DeFazio (D-OR); EPW’s Transportation and Infrastructure Subcommittee, chaired by Sen. Max Baucus (D-MT), and the Banking Committee’s Housing, Transportation and Community Development Subcommittee, chaired by Sen. Robert Menendez (D-NJ). Because of its financing mechanisms, the bill must also go through the House Ways and Means Committee, chaired by Rep. Charles Rangel (D-NY), and the Senate Finance Committee, chaired by Sen. Baucus. Other committees are also involved on the Senate side to a lesser degree. The following diagram traces the path of the transportation bill
ch. 2
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 3 0
>
>
C h
a p
te r
2
Transportation Authorization 101
Diagram 1. Surface Transportation Bill Authorization Process through Congress
Source: Chart from Federal Highway Administration, http://www.fhwa.dot.gov/reports/financingfederalaid/ authact.htm.
Subcommittee Bill
Committee Bill
Senate Bill
HOUSE OF REPRESENTATIVES
Public Hearings
Subcommittee Bill
Committee Bill
House Bill
SENATE
Public Hearings
Any Differences?
No
President
Veto
Override Veto?
No
Start Over SURFACE TRANSPORTATION ACT
Yes
Approval
Conference Committee
Conference Bill
Floor Action
Yes
Diagram 1: Surface Transportation Bill Authorization Process through Congress
2-1
Source: Chart from Federal Highway Administration, http://www.fhwa.dot.gov/reports/financingfederalaid/authact.htm.
ch. 2
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 3 1
<
<
Tr a
n s
p o
rt a
ti o
n A
u th
o ri
za ti
o n
1 0
1
through Congress.
At each level of deliberation—whether subcommittee, committee, or floor—there is an opportunity to educate policymakers and their staff about the connections among transportation, equity, and health and to propose recommendations that will benefit the American public. While all representatives are important when the bill hits the floor of the Senate and House, key committee members are particularly influential in how the bill develops. Each subcommittee and committee has numerous representatives who can weigh in. Members of Congress are elected to serve us, the American people, and they often look to their various constituencies for advice. Advocates on Capitol Hill are making their interests known, and those outside of the nation’s capital are building coalitions, calling their elected representatives, and setting up appointments to voice their needs. The time to act is now.
federa l Oversig ht a nd administration
The U.S. Department of Transportation and its implementing agencies—including the Federal Transit Agency, the Federal Highway Administration, and the National Traffic Highway Safety Administration—administer the funds authorized by the surface transportation bill.
The Highway Trust Fund (HTF) is the primary funding source for transportation. Like other federal trust funds the HTF is a financing mechanism to account for taxes collected by the federal government which are earmarked for a specific purpose or program. Initially, the HTF funded highways only. Later, Congress established that a portion of the funds should be used for public transportation creating the Mass Transit Account as part of HTF in 1983. Currently the Mass Transit Account receives 2.86 cents out of the 18 cent per gallon gasoline tax.3 Recently the HTF has not collected
enough revenue from the gas tax to cover the expenditures it supports. Congress has supplied funds from the general treasury to stop the gap, but this is not a sustainable solution. Congress and advocates are exploring new revenue streams to close the immense funding shortfalls. These include indexing the gas tax to inflation, imposing user fees such as toll or congestion pricing, or levying a sales tax on oil. Financing is an important debate, given the regressive nature of some forms of taxation and fees and the public’s resistance to raising taxes.
At the national level, there are three broad categories of federal transportation funding— highways, public transportation, and highway and motor vehicle safety. Each of these categories represents funding from numerous programs. Walking and biking infrastructure is not listed as a category because it is only a sliver of overall federal transportation spending, primarily through the Transportation Enhancements Program.
Most of the money from the surface transportation bill is distributed to states in two ways—through formula grant programs and through competitive grant programs. Formula- funded programs are by far the largest portion of this funding. The Surface Transportation Program (STP)—the largest program authorized in the surface transportation bill, which many call the highway program—allocates funds directly to state Departments of Transportation using the following formula:
• 25 percent based on total lane miles of federal-aid highways
• 40 percent based on vehicle miles traveled on lanes of federal-aid highways
• 35 percent based on estimated state contributions to the Highway Account4
This program therefore rewards states and regions that drive more, build more highways, and use more gas—a combination that does little
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 3 2
>
>
C h
a p
te r
2
to promote health and environmental quality.
Another significant formula-funded program is the Urbanized Area Formula Grants Program (also called the Large Urban Cities Program), which allocates funds used for public transportation. Urbanized areas of 200,000 or more receive this money directly instead of having the funds go through state departments of transportation. The funds are distributed based on the following formulas:
For areas of 50,000 to 199,999 in population, the formula is based on population and population density. For areas with populations of 200,000 and more, the formula is based on a combination of: (1) the distance in miles that a revenue vehicle (a vehicle that is charging a fare) is operated while it is available for passenger service (also called bus revenue vehicle miles), (2) bus passenger miles, (3) revenue vehicle miles that run along exclusive or controlled rights-of-way or rails (also called fixed guideway revenue vehicle miles), (4) the number of miles of exclusive or controlled right-of-ways or rails for transit (also called fixed guideway route miles), and (5) population and population density.5
The Urbanized Area Formula Grants Program provides funds for public transportation, both rail and bus service. Transit dollars are explicitly prohibited from being used for operations in jurisdictions of 200,000 people and above. Therefore, most federal transit dollars can only be used on capital expenditures and not on operations. Many transit operators have huge gaps in their budgets and are raising fares and decreasing services—often at the same time—to stay afloat; many transit-dependent populations are suffering from this combination. Cutting routes that many residents depend on can create a situation where people cannot get to work or access goods and services. Raising fares particularly hurts low-income people who comprise the majority of the transit-dependent population. Many find themselves struggling even more to budget their transportation costs.
Another important formula program, the Highway Safety Improvement Program, is allocated via formula. The program was specifically created to improve highway safety. Funds are distributed to states based on the following three factors, all of which are weighed equally: (1) lane miles of Federal-aid highways, (2) vehicle miles traveled on Federal aid-highways, and (3) the number of fatalities on the Federal-aid system.6 Thus, the program awards more money to states which drive more, have more highways and more fatalities.
Some programs allow, encourage, or require a portion of the formula funds to be used for specific programmatic goals. For example, the Transportation Enhancements Program (TEP) is allocated using a portion of STP funds. TEP requires the use of a small percent of STP dollars for 12 eligible activities of which walking and biking infrastructure is a significant portion.
Competitive grants are also available for which states and locales can compete. These programs include money for specific program goals. For example, the Job Access and Reverse Commute Program (JARC) provides funding for projects that specifically help connect low-income workers to job centers.7 Another key example of competitive grant programs is the New Starts Program. This is the federal government’s primary financial resource for supporting locally planned, implemented, and operated major transit capital investments. It funds new and extensions to commuter rail, light rail, heavy rail, bus rapid transit, streetcars, and ferries, among others.8 Local entities must match the dollars provided by the Program. While the federal portion of the match can be up to about 80 percent, in reality locales have paid about 50 percent for projects funded by New Starts due to the high demand for this program and the competitive nature of funding. This adds a high financial burden on locales to support the creation of new transit projects.
Transportation Authorization 101
ch. 2
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 3 3
<
<
Tr a
n s
p o
rt a
ti o
n A
u th
o ri
za ti
o n
1 0
1
State a nd loca l Oversig ht
Federal dollars typically require a match by states or local agencies. The exact requirement of matching funds for competitive grants and formula grants varies by program.
Generally, transportation projects have been funded accordingly:
• Highways: 25 percent federal, mostly for capital investments; 50 percent states, for capital and maintenance; remaining 25 percent local governments9
• Transit: 25 percent federal, for largely capital investment; the remaining funds are split, 70–80 percent funded directly from transit users and local governments for operational costs; the remaining 20–30 percent is provided by state governments.10
At the local level, metropolitan planning organizations (MPOs) share $300 million a year in federal transportation funds. MPOs make policy at the regional level and work with state transportation agencies and regional officials to develop regional transportation plans. MPOs’ composition varies significantly from region to region, with representatives from local government, transportation authorities, and other stakeholders. About 385 MPOs operate in the United States. MPOs are required for urbanized areas with populations of more than 50,000 residents. The U.S. Secretary of Transportation can also designate transportation management areas (TMAs) for metropolitan areas with populations greater than 200,000.
While the needs of rural communities have been somewhat overlooked in transportation planning and decision making, rural planning organizations (RPOs)—consisting of networks of local planners, officials, and other stakeholders—do exist in smaller communities.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 3 4
>
>
C h
a p
te r
2
RPOs are not federally mandated. State DOTs control planning and project selection outside of MPO areas. Therefore, rural areas have very little say in how transportation investments are made in their communities. Previous transportation bills provided some flexibility for transferring funds and suballocating dollars to cities and regions, but they lacked federal direction on what kind of national objectives should be promoted through these investments. Local and regional empowerment has been stunted in most states, given the lack of authority at the regional or local level in the project selection process or the direct funding allocation decision making. The impending bill should seek to provide direction on national objectives and create opportunities for appropriate ways to empower regional and local decision making that is equitable and provides a voice for all residents.
a Time for r eform
There is no doubt that the U.S. transportation system critically needs reforming. Many of the most pressing issues and challenges our nation faces today—obesity, air quality, climate change, congestion, energy independence, lack of access, and sprawl—are linked to transportation.
Public health and equity advocates have vital roles to play among the many partners who will shape this new system. In fact, all of our transportation policies, programs, and decisions should be steeped in the understanding that safety, health, equity, and well-being of the general public is a national priority, that public health and equity must always be considered when creating transportation policy. National transportation objectives are being considered in the next surface transportation bill. Objectives would guide transportation investments to correspond with national goals of environmental quality, safety, equity and public health. National objectives also improve accountability of transportation investments by setting performance measures which help eliminate disparate funding between modes and ensure the country’s transportation system helps America move towards a healthy and sustainable future.
The coming authorization of the federal surface transportation bill affords the crucial opportunity to help shape and, more importantly, reform our transportation system. And this time around: public health and equity considerations must not be confined to a small number of specialty program areas; they should be an overriding theme throughout all transportation programming.
Transportation Authorization 101
ch. 2
How we get around—in cars or on foot, by bus, bicycle, light rail, or commuter train—affects public health, environmental quality, economic vitality, and social equity. The following section examines specific surface modes of transportation that have significant potential to improve health, reduce emissions, and increase access to jobs and other opportunities, particularly in underserved communities. These travel options also hold enormous opportunity for reform through the upcoming authorization of the federal surface transportation bill.
The chapters in this section cover:
>> Public transportation
>> Walking and bicycling
>> roadways
While modes of travel are important to highlight in debates over the bill and in the national priorities it will ultimately reflect, federal transportation policies and funding should not fall into mode silos. Rather, policies and funding should be driven by performance measures that hold states and locales accountable for creating transportation systems that promote health, environmental quality, and opportunity for all.
Modes of travel should not compete with one another. Instead, each mode should be placed on equal footing to allow American cities and towns to incorporate and connect various modes of travel in order to meet the needs of diverse and changing populations.
TrAnSPOrTATIOn OPTIOnS
chapter name
p g
. 3 6
>
>
Public Transportation and Health ch. 3 TODD LITMAN, M.E.S. Founder and Executive Director Victoria Transport Policy Institute Victoria, British Columbia
ABSTRACT >> Improving public transportation service, encouraging its use, and integrating it into community development plans can make Americans healthier by reducing per capita automobile travel and associated risks, increasing walking and cycling activity, and improving mobility for disadvantaged people. Conventional transportation policies and planning practices tend to favor the automobile. Various reforms can help create more efficient and equitable transportation systems that, among other benefits, help improve public health. This paper investigates these issues, examines the role public transportation plays in an efficient and equitable transport system, and presents specific recommendations for transportation and land use policies to help achieve public health objectives.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 3 8
>
>
C h
a p
te r
3
Public Transportation and Health
CONTENTS
Introduction .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 39
Public Transportation's Roles.. .. .. .. .. .. .. .. .. .. 39
Public Transportation Health Impacts . .. .. .. .. .. 47
Traffic Crashes .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 47
Pollution Emissions .. .. .. .. .. .. .. .. .. .. .. .. .. 49
Physical Activity and Fitness . .. .. .. .. .. .. .. .. 49
Community Cohesion . .. .. .. .. .. .. .. .. .. .. .. 51
Mental Health Impacts .. .. .. .. .. .. .. .. .. .. .. 52
Basic Mobility.. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 53
Policy Opportunities and Barriers . .. .. .. .. .. .. .. 53
Recommendations .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 55
Convergence Opportunities .. .. .. .. .. .. .. .. .. .. 59
Conclusion.. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 60
LIST OF ILLUSTRATIONS
Figures
1. Transit Commute Mode Split in Selected Cities .. .. .. .. .. .. .. .. .. .. .. .. .. .. 40
2. Cycle of Automobile Dependency and Sprawl .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 43
3. International Vehicle Travel Trends . .. .. .. .. .. 44
4. Annual Change in Transit and Vehicle Travel . .. .. .. .. .. .. .. .. .. .. .. .. .. 45
5. Transport Fatalities .. .. .. .. .. .. .. .. .. .. .. .. .. 46
6. Annual Traffic Death Rates .. .. .. .. .. .. .. .. .. 47
7. U.S. Traffic Deaths .. .. .. .. .. .. .. .. .. .. .. .. .. 48
8. Daily Walking Trips and Transit Travel .. .. .. .. 50
9. Mode Split vs. National Obesity Rates . .. .. .. 52
Tables
1. Transit Level-of-Service Indicators.. .. .. .. .. .. 41
2. Personal Travel Mode Split of Various Countries .. .. .. .. .. .. .. .. .. .. .. .. 51
3. Scope of Conventional Planning Analysis .. .. 54
4. Healthy Transportation Policy Implementation . .. .. .. .. .. .. .. .. .. .. .. .. .. .. 58
ch. 3
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 3 9
<
<
P u
b li
c T
ra n
s p
o rt
a ti
o n
a n
d H
e a
lt h
introduction
Public transportation (also called public transit and mass transit) refers to various services using shared vehicles to provide mobility to the public, including buses, trains, and shared taxis. High- quality and affordable public transportation can help achieve various public health and equity goals by reducing traffic fatality rates, reducing air pollution emissions, increasing physical fitness, and improving nondrivers’ access to elemental goods and services—fresh, healthy food and healthcare—and reducing financial burdens on low-income households. In addition, public transportation can bolster a community’s quality of life by easing traffic congestion, energy costs, and pollution. Consequently, policies and investments that improve public transportation can be considered win-win strategies, providing diverse benefits and attracting broad support from a variety of interest groups.
However, current policies and planning practices fail to support public transportation to the degree justified by these benefits. Current evaluation practices overlook many benefits of public transportation, including many health benefits, and transportation financing systems provide inadequate funding. Without policy and planning reforms, public transportation will fail to provide its full potential benefits.
This paper examines the role public transportation plays in an efficient transportation system, the health benefits that can accrue from such a system, and models for creating a more equitable community by reforming transport policies and planning practices.
Public Tra nspor tation’s r oles
Public transportation plays multiple roles in an efficient and equitable transportation system. It provides basic mobility for people who cannot use or access an automobile; it provides
efficient transportation on major urban corridors; and it serves as a catalyst for more compact, walkable communities, called transit oriented development.
Public transportation consists of:
• Heavy rail—relatively large, higher-speed trains, operating on separate rights-of-way, with infrequent stops, providing service between communities.
• Light-rail transit—moderate-size, medium- speed trains, operating mainly on separate rights-of-way, with variable distances between stations, providing service within an urban area.
• Bus rapid transit—bus systems with premium features, including grade separation, quick boarding, and frequent service.
• Express commuter bus—direct bus service from residential to employment areas.
• Conventional urban bus transit— medium- and full-size buses on fixed route, scheduled service.
• Mini bus—smaller buses or large vans used for public transportation.
• Demand response paratransit—small buses or vans that provide direct (door-to- door) service, often intended primarily for people with disabilities.
Each type of public transportation has its niche. Bus rapid transit and light-rail transit are the most appropriate on major urban corridors connecting large activity centers. Express commuter service is most appropriate on longer-distance commuter corridors with large employment centers (such as between suburbs and downtown). Conventional buses are most appropriate on urban and suburban roadways. Demand response is most appropriate in lower-density areas as well as for serving people with special needs.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 4 0
>
>
C h
a p
te r
3
Although public transportation accounts for only a small portion of total travel in North America, it accommodates trips that are particularly important and costly to serve by other modes. In big cities, public transportation typically serves five to 15 percent of all commutes (figure 1) and as much as 20 to 60 percent of trips to major activity centers such as downtowns and university campuses. It provides mobility to people who are physically, economically, and socially disadvantaged and who would otherwise need to walk, bicycle, pay for a taxi, or simply not travel, sometimes to critical activities such as a doctor’s appointment, work, or school.
High-quality public transportation (either rail or
bus service that is convenient, fast, comfortable, and affordable) reduces automobile travel directly, by attracting travelers who would otherwise drive, and indirectly, by serving as a catalyst to help create more compact, walkable communities where residents drive less and rely more on alternative modes.2 These indirect, or leveraged, impacts often produce bigger results: studies indicate that each passenger-mile traveled in quality public transportation reduces the number of automobile vehicle-miles traveled by two to nine automobile vehicle-miles.3 As a result, residents of communities with access to good public transportation systems tend to drive 20 to 40 percent fewer annual miles than they would if they lived in more automobile- dependent communities.4
Public Transportation and Health
0%
5%
10%
15%
20%
25%
30%
35%
Figure 1. Transit Commute Mode Split in Selected Cities
3-1
M ad
iso n
Clev ela
nd
Den ve
r
M in
nea polis
Po ughke
ep sie
Durh am
Lo s A
ngele s
Po rtl
an d
Pit tsb
urg h
Balt im
ore
Conco rd
Honolu lu
Se at
tle
Brid gep
ort
Ph ila
delp hia
Chica go
Bosto n
W as
hin gto
n, D C
Sa n Fr
an cis
co
New Y
ork
31.1%
16.9% 15.8%
13.2% 12.6%
9.9% 9.9% 8.7% 8.5%
7.8% 7.6% 7.1% 6.5% 6.3% 5.9% 5.8% 5.3% 5.2% 5.1% 5.0%
Although public transit serves only a small portion of total travel, it serves a significant portion of urban trips.
Figure 1. Transit Commute Mode Split in Selected Cities 1
Although transit serves only a small portion of total travel, it serves a significant portion of urban trips.
ch. 3
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 4 1
<
<
P u
b li
c T
ra n
s p
o rt
a ti
o n
a n
d H
e a
lt h
Feature Description Indicators
Availability Where and when transit service is available
• Annual service-kilometers and service-hours per capita
• Daily hours of service
Frequency Frequency of service and average wait time
• Trips per hour or day • Headways (time between trips) • Average waiting times
Travel speed Transit travel speed • Average vehicle speeds • Transit travel speed relative to driving speed
for the same trip
Reliability How well service actually follows published schedules
• On-time operation • Portion of transfer connections made
Boarding speed
Vehicle loading and unloading speed
• Dwell time (time spent waiting at a stop or station)
• Boarding and alighting speeds
Safety and security
Users’ perceived safety and security
• Perceived transit passenger security • Number of accidents and injuries • Reported security incidents
Price and affordability
Fare prices, structure, payment options, ease of purchase
• Fares relative to average incomes • Fares relative to other travel mode costs • Targeted discounts or exemptions as
appropriate • Payment options (cash, credit cards, etc.)
Integration Ease of transferring between transit and other travel modes (bus, train, ferry, airport, etc.)
• Quality of transit service to transport terminals • Ease of accessing transit service information
from transport terminals
Comfort Passenger comfort • Seating availability and quality • Space (lack of crowding) • Quiet (lack of excessive noise) • Temperature (neither too hot nor too cold)
and air quality • Cleanliness
Accessibility Ease of reaching transit stations and stops
• Transit oriented development • Distance from transit stations and stops to
destinations • Walkability in areas serviced by transit
Baggage capacity
Accommodation of baggage • Ability to carry onboard baggage, including special items such as pets
• Ease and cost of carrying on baggage
Table 1. Transit Level-of-Service Indicators 5
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 4 2
>
>
C h
a p
te r
3
There are many ways to improve transit service and increase ridership (table 1). For instance, in the short-term, it is often possible to add new routes, increase service frequency, improve security, offer fare discounts, provide new amenities such as on-board refreshments and wireless Internet service (particularly for longer-distance express commuter service), and provide incentives such as parking cash out (offering commuters who currently receive subsidized parking the option of choosing its cash equivalent if they use alternative modes) and other rewards. In the medium-term, it is often possible to accelerate transit travel speeds,
increase reliability, improve stops and stations, provide real-time vehicle arrival information, upgrade vehicles for smoother and quieter rides, make trips more comfortable through better temperature control and fresh air, and provide park-and-ride facilities. In the long-term, it is often possible to create more transit oriented development so that more destinations (homes, worksites, and recreation and cultural centers) are located along major transit routes, with convenient pedestrian and bicycle access.
People sometimes mistakenly assume that these strategies are only feasible in large cities, but
Public Transportation and Health
Feature Description Indicators
Universal design
Accommodation of diverse users, including people with special needs
• Accessible design for transit vehicles, stations, and nearby areas
• Accommodation for people with limited language ability
User information
Ease of obtaining user information
• Availability, accuracy, and understandability of route, schedule, and fare information
• Real-time transit vehicle arrival information
Courtesy and responsiveness
Courtesy with which passengers are treated
• How passengers are treated by transit staff • Ease of filing a complaint • Responsiveness with which complaints are
treated
Attractiveness The attractiveness of public transportation facilities
• Attractiveness of vehicles and facilities • Attractiveness of documents and websites • Quality of nearby buildings and landscaping • Parks and recreational areas accessible by
transit • Provision of public art
Marketing Effectiveness of efforts to encourage using public transportation
• Popularity of promotion programs • Effectiveness at raising the social status of
transit travel • Increase in public transportation ridership in
response to marketing efforts
This table summarizes various factors to consider when evaluating public transportation services.
Table 1 continued
ch. 3
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 4 3
<
<
P u
b li
c T
ra n
s p
o rt
a ti
o n
a n
d H
e a
lt h
some alternative modes are suitable for use in suburban and rural areas.6 These include ridesharing (car- and vanpooling), demand response transit (shuttle vans and buses that operate on flexible routes to provide door-to- door service in more dispersed areas), improved walking and cycling facilities (such as wider road shoulders and separated paths), telework (use of telecommunications as a substitute for physical travel, such as improving Internet networks and having more online public services in rural areas), and delivery services.7 Rural and suburban areas can become more accessible and multi-modal by encouraging village
development, where shops, public services, and housing (particularly for older adults and other nondrivers) are located close together and served by regional public transportation.
Improving and encouraging public transportation is a timely issue. During the past century, transportation planning focused primarily on cars, and transit systems were evaluated primarily in terms of automobile travel speed, affordability, and safety. Transportation improvements consisted primarily of building more roads and parking facilities. Planners barely considered other modes, which were
Figure 2. Cycle of Automobile Dependency and Sprawl
This figure illustrates the self-reinforcing cycle of increased automobile dependency and sprawl.
Automobile-Oriented
Transport Planning
Reduced Travel
Options
CYCLE OF AUTOMOBILE DEPENDENCY
Alternative
Modes Stigmatized
Suburbanization and
Degradation of Cities
Automobile-Oriented
Land
Use Planning
Generous Parking
Supply
Dispersed
Development Patterns
Increased Vehicle Ownership
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 4 4
>
>
C h
a p
te r
3
Public Transportation and Health
A N
N U
A L
PA SS
E N
G E
R K
IL O
M E
T E
R S
PE R
C A
PI TA
1970 1980 1990 2000
YEAR 2007
0
5000
10000
15000
20000
25000
U.K.
Switzerland
Sweden
Spain
Portugal
Norway
Netherlands
Italy
Ireland
Greece
Germany
France
Denmark
Belgium
US
Per capita vehicle travel grew rapidly between 1970 and 1990 but has since leveled off in most OECD (Organizations for Economic Cooperation and Development) countries and is much lower in European countries than in the United States.
Figure 3. International Vehicle Travel Trends 8
ch. 3
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 4 5
<
<
P u
b li
c T
ra n
s p
o rt
a ti
o n
a n
d H
e a
lt h
considered of declining relevance in a culture increasingly dependent on automobile travel. The result was a self-reinforcing cycle of increasing automobile dependency and sprawl, as illustrated in figure 2.
But per capita automobile travel has peaked and has recently started to decline slightly in most economically developed countries, as illustrated in figure 3.
These changes reflect demographic and economic trends that are reducing demands for automobile travel and increasing demands for alternative modes9:
• Increasing health and environmental concerns. Numerous individuals, organizations, and jurisdictions are now committed to reducing pollution and increasing physical fitness.
• Aging population. As the baby boom generation retires, per capita vehicle travel will decline and their demand for alternatives will increase.
Figure 4. Annual Change in Transit and Vehicle Travel 10
Transit trips increased more than vehicle mileage during seven of the last 10 years. Note: Annual percent change in 2002 was zero. Therefore the chart does not include a visible bar for transit trips.
Transit Trips
Vehicle Mileage
A N
N U
A L
C H
A N
G E
-5
-4
-3
-2
-1
0
1
2
3
4
5
6
1999 2000 2001 2002 2003 2004
YEAR 2005 2006 2007 2008
Figure 4. Annual Change in Transit and Vehicle Travel
3-4
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 4 6
>
>
C h
a p
te r
3
• Uncertain future fuel prices. This uncertainty increases demand for energy- efficient travel options and more accessible, multi-modal locations for homes and businesses.
• Increasing urbanization. An increasing portion of households are choosing to live in existing cities, and many suburbs are becoming more urbanized. This increases demand for urban modes (walking, bicycling, and public transportation).
• Increasing traffic congestion and roadway construction costs. This increases the relative value of alternative modes that reduce congestion.
• Shifting consumer preferences. Various indicators suggest that an increasing number of consumers prefer living in more densely populated urban neighbourhoods and using multiple modes of travel.
As a result of these shifts, public transportation travel grew more than automobile travel during seven of the last 10 years and each of the last four years, as illustrated in figure 4. During this period, transit travel increased 24 percent compared to a 10 percent increase in automobile vehicle miles traveled. Many transit systems now carry their maximum capacity during peak periods, constraining further growth. Increasing capacity and improving service quality would allow further growth in
Public Transportation and Health
Figure 5. Transport Fatalities 13
Public transportation travel has lower crash rates than automobile travel, taking into account risks to all road users.
FA TA
LI T
IE S
PE R
B IL
LI O
N P
A SS
E N
G E
R -M
IL E
S
0
2
4
6
8
10
12
Passenger Car Light Trucks Intercity Bus Transit Bus Heavy Rail Commuter Rail
1.3
9.2
7.9
2.3
10.5
.3
5.0
2.2
8.2
8.2
.3
0.4
1.8
8.1
.1
4.4
.6
Other Road Users
Vehicle Occupants
Figure 5. Transport Fatalities
CH 3
ch. 3
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 4 7
<
<
P u
b li
c T
ra n
s p
o rt
a ti
o n
a n
d H
e a
lt h
public transportation ridership and additional reductions in automobile travel.
There is also growing demand for housing in multi-modal communities.11 The 2004 American Community Survey found that consumers place a high value on urban amenities such as shorter commute time and neighborhood walkability. Sixty percent of prospective homebuyers surveyed indicated that they preferred a neighborhood that offered sidewalks, a shorter commute, and amenities such as shops, restaurants, libraries, schools, and public transportation over more sparsely populated areas with larger lots but longer commutes and poorer walking conditions.12
Public Tra nspor tation Hea lth impacts
This section describes ways that improving public transportation can help achieve health objectives.
Traffic Crashes
Public transportation is relatively safe, as indicated in figure 5. Transit vehicle occupants have about one-tenth the fatality rate as car occupants, and even considering the risk to other road users, public transportation causes fewer than half the total deaths per passenger- mile as automobile travel.
Figure 6. Annual Traffic Death Rates 15
The smartest growth counties in the United States have one-fifth of the average per capita traffic fatality rate as the most sprawled counties.
Figure 6. Annual Traffic Death Rates
3-6
Most Sprawled
Smartest Growth
0
5
10
15
20
25
30
35
40
Gea uga C
ounty, O
H
Clin to
n C ounty,
M I
Fu lto
n C ounty,
O H
Gooch lan
d C ounty,
V A
Yad kin
C ounty,
N C
W alt
on C ounty,
G A
Isa nti
County, M
N
Dav ie
County, N
C
M iam
i C ounty,
K S
St oke
s C ounty,
N C
Balt im
ore C
ity , M
D
Rich m
ond C ounty,
N Y
Su ffo
lk County,
M A
Ph ila
delp hia
County, PA
Hudso n C
ounty, N
J
Sa n Fr
an cis
co C
ounty, C
A
Quee ns C
ounty, N
Y
Bro nx C
ounty, N
Y
Kin gs C
ounty, N
Y
New Y
ork C
ounty, N
Y
A N
N U
A L
T R
A F
F IC
D E
AT H
S PE
R 1
00 ,0
00 P
O PU
LA T
IO N
4.42 4.46 4.2 4.58 6.31 5.91
8.04
4.49 5.63
7.68
15.66
38.80
25.84
12.78
19.77
38.52
35.58
38.02
16.99
20.9
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 4 8
>
>
C h
a p
te r
3
High-quality public transportation provides even greater safety benefits than indicated by these distance-based fatality rates because it tends to leverage additional reductions in per capita vehicle travel. People who live or work in transit oriented areas tend to drive less (due to more accessible, multi-modal community design), drive at lower traffic speeds (due to more compact development), and do less high- risk driving (for example, teenagers are less likely to have a driver’s license and own a vehicle). 14 As a result, such communities have about one- fifth of the total per capita traffic fatality rate as sprawled, automobile-dependent communities, taking into account all traffic deaths, including risks to pedestrians, bicyclists, and public transportation travelers (figure 6). Traffic deaths
are a subcategory of violent deaths and overall, urban residents have significantly lower rates of violent deaths, even taking into account homicide risk.16
Per capita traffic fatalities decline as transit ridership increases in a community, as indicated in figure 7. The reduction in per capita crash rates is much larger than the reduction in per capita mileage in these cities, reflecting the combined effects of various transportation and land use factors associated with transit oriented development that increase safety, as previously described.
Public Transportation and Health
Figure 7. U.S. Traffic Deaths 17
Per capita traffic deaths (including transit and automobile occupants as well as pedestrians) tend to decline with increased transit ridership and are particularly low in cities with strong rail transit systems.
0
5
10
15
20
25
0 200 400 600 800 1000 1200
Strong Rail
Weak Rail
No Rail
T R
A F
F IC
F AT
A LI
T IE
S PE
R 1
00 ,0
00 P
O PU
LA T
IO N
ANNUAL PER-CAPITA TRANSIT PASSENGER-MILES
R2 = 0.352
ch. 3
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 4 9
<
<
P u
b li
c T
ra n
s p
o rt
a ti
o n
a n
d H
e a
lt h
Pollution Emissions
A second category of transport-related health impacts involves vehicle pollution emissions, including tailpipe emissions; also included are emissions from fuel production and distribution (“upstream” emissions), hot soak (evaporative emissions that occur after an engine is turned off), and particulates from road dust, brake linings, and tire wear.18
Many factors affect vehicle pollutant human health impacts, including emission rates per vehicle mile, per capita mileage, and exposure (the number of people located in areas where emissions are concentrated). Motor vehicle air pollution is estimated to cause a similar order of magnitude of total premature deaths as traffic crashes, although the victims tend to be older; thus air pollution causes smaller reductions in Potential Years of Life Lost (PYLL) than traffic crashes.19
Public transportation tends to produce less pollution per passenger-mile, particularly electric-powered trains and newer buses with state-of-the-art engines. And, as previously discussed, transit oriented development tends to reduce automobile travel and, therefore, emissions. On the other hand, older diesel buses tend to have high emission rates; public transportation tends to concentrate activity close to roadways; and bus depots are often located in low-income communities. Consequently, in some situations, increased transportation service and transit oriented development may increase human exposure to harmful air pollutants such as particulates and carbon monoxide unless implemented with bus emission reduction programs.
Physical Activity and Fitness
Another category of health impacts concerns the effects transport has on physical activity and fitness.20 Public health officials have become increasingly alarmed about declining physical fitness, increasing body weight, and
resulting increases in diseases associated with a sedentary lifestyle.21 There are many ways to be physically active, but many, such as team sports and gym exercise, require special time, skill, and expense, which discourage consistent, ongoing participation. Many experts believe that increasing community walking and bicycling (together called “active transportation”) are the most practical ways to improve public fitness, particularly for vulnerable populations— children, older adults, and people with low incomes who may be unable to participate in structured exercise programs due to financial and time constraints.22
Public transportation and active transportation tend to be complementary: most public transportation trips involve walking links; transit oriented development includes walking and biking improvements; and efficient transit systems incorporate amenities such as bike racks on buses and bike lockers at transit stations.23 As a result, increased transit travel tends to increase physical activity.
The National Household Travel Survey (NHTS) indicates that people who use public transportation on a particular day spend a median of 19 minutes daily walking to and from transit, and 29 percent achieve 30 minutes of physical activity during transit access trips—much higher than the rates by nontransit users.24 Using pedometers and surveys to track walking activity, Wener and Evans found that train commuters walked an average of 30 percent more steps daily, more frequently reported walking for 10 minutes or more, and were four times more likely than automobile commuters to achieve the 10,000 steps daily recommended for fitness and health.25
Similarly, a travel survey conducted in Atlanta, GA, found that public transportation users are more likely to walk, to walk longer average distances, and to meet recommended physical activity targets by walking than nontransit users.26 The study revealed that the chance a person meets minimum walking targets (2.4
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 5 0
>
>
C h
a p
te r
3
kilometers walked daily) increases by 3.87 for each transit trip taken and is 2.23 times greater for commuters who use an employer-sponsored public transportation pass. Public transportation travel increased walking activity for all income classes, as illustrated in figure 8, indicating that encouraging transit travel can support public health for a variety of demographic groups.
Residents of transit oriented communities tend to walk more and have lower rates of obesity and hypertension than residents in sprawled areas. A recent study collected transportation mode split and obesity rate data for various economically developed countries, as summarized in table 2 and figure 9. Two important points are illustrated: travel
patterns are highly variable, even among similar countries, and national obesity rates tend to be inversely related to rates of active transportation (walking and biking), suggesting that transport policy affects public fitness and health.
As a result, policies and planning practices that support public transportation tend to increase public fitness and health. Sturm estimates that shifting from a sprawled area such as San Bernardino, CA, to a areas which reflect smart growth principles such as Boston, MA, reduces chronic medical conditions about 16 percent, with greater reductions for older adults and low-income people because they tend to be most sedentary.30
Public Transportation and Health
Public transportation users are much more likely to take walking trips and walk much farther than nontransit users.
Transit Users
Nontransit Users
0%
10%
20%
30%
40%
50%
60%
70%
80%
59.6%
11.6%
60.9%
9.0%
56.3%
8.9%
58.9%
9.3%
Figure 8. Daily Walking Trips and Transit Travel
3-8
T O
O K
A T
L E
A ST
O N
E W
A LK
T R
IP
ANNUAL INCOME CLASS
Under $30k $30–60k Over $60k Total
Figure 8. Daily Walking Trips and Transit Travel 27
ch. 3
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 5 1
<
<
P u
b li
c T
ra n
s p
o rt
a ti
o n
a n
d H
e a
lt h
The total health costs that result from inadequate physical activity are far greater than those from traffic crashes. Cardiovascular diseases cause about 10 times the loss in productivity as do road crashes, and sedentary living contributes to a variety of other health problems—hypertension, non–insulin- dependent diabetes, colon cancer, osteoarthritis, osteoporosis, and probably depression. Even modest reductions in these illnesses could provide large health benefits. However, it is difficult to determine how a particular transportation policy will affect these diseases overall because it depends on the ability of otherwise sedentary people to increase their physical activity. The Health Benefits Economic
Model provides a methodology for valuing the health benefits of more active transportation.31
Community Cohesion
Community cohesion refers to the quantity and quality of positive interactions among residents in a local community.32 It affects human health in various ways, including the mental health benefits of friendly social interactions and the health benefits of increased neighborhood security.33 Although many demographic and geographic factors affect neighborhood interactions, cohesion tends to increase with walkability and local services.34 High-quality public transportation and transit oriented
Country Year Transit Bike Walk Obesity Rates*
Latvia 2003 32% 5% 30% (13.7%*)
Switzerland 2005 12% 5% 45% 8%
Netherlands 2006 5% 25% 22% 8.1% (11.2%*)
Spain 2000 12% N/A 35% 12.8%
Sweden 2006 11% 9% 23% 9.4%
Germany 2002 8% 9% 23% 12.1%
Finland 2005 8% 9% 22% 13.3%
Denmark 2003 8% 15% 16% 12.2%
Norway 2001 10% 4% 22% 14.3%*
U.K. 2006 9% 2% 24% 24%*
France 1994 8% 3% 19% 11%
Ireland 2006 11% 2% 13% 18%
Canada 2001 11% 1% 7% 15.2 (22.7%*)
Australia 2006 8% 1% 5% 16.2% (20.8%*)
U.S. 2001 2% 1% 9% 34.3%*
Table 2. Personal Travel Mode Split of Various Countries28
* Combined male and female obesity prevalence based on body mass index (BMI). Values in parentheses are from national health examination surveys. Other values are based on self-reported weight and height.
Source: D. Bassett et al., “Walking, Cycling, and Obesity Rates in Europe, North America, and Australia,” 2008.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 5 2
>
>
C h
a p
te r
3
development can increase community cohesion by creating opportunities for residents to interact while walking, waiting at transit stops, and riding on transit vehicles. Further, they reduce total automobile traffic, which improves the public realm, for example, by reducing traffic noise on sidewalks and front yards.35 This can increase connections and contacts among dissimilar groups, helping to bridge social distance and widening opportunities by introducing disadvantaged children to more affluent families and broadening the pool of role models and mentors available to low- income youths.36 Long-term social and economic benefits can result by increasing educational and
employment opportunities and reducing crime and dependence on social assistance.
Mental Health Impacts
Public transportation improvements such as increased service, improved climate control, more comfortable waiting conditions, and improved service reliability can improve mental health by reducing physical and emotional stresses (crowding, fear, and frustration), increasing affordability (and therefore reduced financial stress), influencing access to education and employment activities (and therefore long- term economic opportunities), and helping
Public Transportation and Health
Figure 9. Mode Split vs. National Obesity Rates29
This data set indicates that transportation mode split is highly variable, even among economically developed countries, and national obesity rates are inversely related to rates of active transportation (walking and bicycling).
Walk
Obesity Rates
Bike
Transit
0%
10%
20%
30%
40%
50%
60%
70%
80%
43%
32
67%
12
62%
22
52%
25
5 12
35
47%
5
8% 11.2%
12.8% 9.4%
12.1% 13.3% 12.2%
14.3%
24%
11%
18%
22.7%
34.3%
45
5
30
13.7% 9
11
23
40% 39% 39% 36% 35%
30%
26%
19%
14% 12%
23
9
8
22
9
8
16
15
8
22
4
10
24
2
9
19
3
8
13
2
11
7
1
11
5
20.8%
1
8
9
1
USA
Austr ali
a
Can ad
a
Ire lan
d
Fr an
ceUK
Norw ay
Den m
ar k
Fin lan
d
Ger m
an y
Sw ed
en Sp
ain
Net her
lan ds
Sw itz
er la
nd
La tv
ia
Figure 9. Mode Split vs. National Obesity Rates29
3-9
2
Source: D. Bassett et al., “Walking, Cycling, and Obesity Rates in Europe, North America, and Australia,” 2008.
ch. 3
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 5 3
<
<
P u
b li
c T
ra n
s p
o rt
a ti
o n
a n
d H
e a
lt h
to create more walkable communities, which increases physical activity and fitness.37 With high-quality service, many commuters find public transportation less stressful than driving.38 These mental health benefits are difficult to quantify but potentially large.
Basic Mobility
Basic mobility refers to people’s ability to access services and activities that society considers basic or essential, including medical and dental services, food and other basic goods, banking, education, and employment opportunities.39 Basic mobility is important for physical and mental health and is a critical equity objective. Public transportation provides basic mobility and accessibility, including access to medical services, affordable and healthy food, education, and employment. Inadequate transport options can result in patients missing appointments, which can exacerbate medical problems and waste medical resources, or force patients or medical service providers to pay for more costly transport services such as taxis.40 One survey found that four percent of U.S. children (3.2 million in total) either missed a scheduled healthcare visit or did not schedule a visit during the preceding year because of transportation restrictions.41 Although it is difficult to quantify the ultimate health benefits from basic mobility provided by public transportation, anecdotal evidence suggests that these impacts can be significant.
Policy Oppor tunities a nd Ba rriers
As noted, alternative modes—walking, cycling, and public transportation—can provide many economic, social, and environmental benefits. Yet current policy analysis and planning practices tend to undervalue alternative
modes and thus provide less support for and investment in them than is optimal.42 Some specific ways that alternative modes are undervalued are described below.
Conventional transportation planning analysis tends to focus on a limited set of impacts and objectives and overlooks others, as summarized in table 3. The impacts that conventional planning focuses on most—travel speed, congestion, and vehicle operating costs—tend to favor automobile transportation. Many benefits of public transportation, such as basic mobility for nondrivers and parking cost savings, are generally overlooked in conventional policy and planning analysis. Some of these omissions reflect the difficulty of quantifying impacts such as equity and sprawl costs, but others (parking costs and mileage-based depreciation, for example) are ignored simply out of tradition.
For example, when comparing highway expansion projects with public transportation improvements, conventional planning generally ignores the effects of generated traffic (the additional peak-period vehicle travel that results if congested roads are expanded), additional downstream congestion (additional traffic on surface streets), parking costs, vehicle ownership costs, traffic accidents, energy consumption, and pollution emissions—all costs that can be reduced if improved service allows the same trips to be made by public transportation. In addition, conventional analysis assumes that everybody (or, at least, everybody who matters) has a vehicle and can drive and thus assigns no explicit value to improving mobility for nondrivers.
Conventional analysis assigns no value to the fitness, health, and enjoyment benefits of increased walking and cycling activity44; conventional planning analysis would recognize
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 5 4
>
>
C h
a p
te r
3
the value of a motor vehicle trip to a gym to allow passengers to exercise on a treadmill, or to a park to walk or bike on public paths, but would not recognize the value of being able to walk or bike, rather than drive, for local errands.
Conventional planning tends to evaluate transport system performance based on the speed, convenience, and affordability of automobile travel, using indicators such as roadway level of service, average traffic speeds, congestion delay, parking supply per 1,000 square feet of building floor area, crash risk per 100 million vehicle-miles, and vehicle operating costs (particularly fuel costs). Comparable indicators are not usually provided for alternative modes, so it is more difficult to identify walking, cycling, and public transportation problems
as well as opportunities to improve these modes. For example, urban transportation models are often used to produce maps that show roadway congestion delays, indicated by roadway level-of-service grades from A to F, but no comparable indicators are provided for walking, cycling, and public transportation problems, putting these modes at a competitive disadvantage for investment.
This type of analysis often implies that public transportation investments are not cost effective, but this results, in part, from biases in conventional traffic models that tend to exaggerate the benefits of highway expansion and understate the benefits of improving alternative modes, particularly high-quality public transportation.
Public Transportation and Health
Usually Considered Often Overlooked
Financial costs to governments
Travel speed (reduced congestion delays)
Vehicle operating costs (fuel, tolls, tire wear)
Per-mile crash risk
Project construction environmental impacts
Downstream congestion impacts
Generated traffic impacts
Nondriver mobility, convenience, and comfort
Transportation diversity value (e.g., mobility for nondrivers)
Parking costs
Vehicle ownership and mileage-based depreciation costs
Project construction traffic delays
Total energy consumption and pollution emissions
Strategic land use objectives
Per capita crash risk
Impacts on physical activity and public health
Some travelers’ preference for transit (lower travel time costs)
Table 3. Scope of Conventional Planning Analysis 43
Conventional transportation planning tends to focus on a limited set of impacts, exaggerating the benefits of highway expansion and undervaluing transit improvements.
ch. 3
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 5 5
<
<
P u
b li
c T
ra n
s p
o rt
a ti
o n
a n
d H
e a
lt h
Transportation financing is also biased in favor of roadway improvements. A major portion of transportation funding is legally or practically restricted to automobile facilities and cannot be used to improve public transportation services, even when such improvements are more cost effective and beneficial overall.45 Thirty of the 50 states have constitutional amendments that limit fuel tax revenue to be spent only on highways, and most zoning codes require developers to provide generous amounts of vehicle parking—a large subsidy of driving that is difficult to convert into transit subsidy, even if preferred by some travelers (a concept called parking cash out). More neutral financing (sometimes called least cost planning) tends to increase funding for alternative modes and mobility management strategies.
Current transportation markets are further distorted in favor of automobile travel by underpricing. Although automobiles are expensive to own, they are relatively cheap to drive because most of the costs are either fixed or external. This gives motorists an incentive to drive more annual miles than optimal. An efficient transportation market would require increased road, parking, and fuel prices, along with distance-based insurance and registration fees, which would significantly increase the marginal cost of driving, particularly under urban peak conditions.
Together, these planning and market distortions increase automobile travel beyond what is economically optimal, reduce use of alternative modes, and stimulate more dispersed, automobile-oriented land use development. Described differently, with more optimal transport planning and pricing, consumers would choose to drive less, rely more on alternative modes, select more multi-modal communities, and be better off overall as a result.46 Although it is difficult to predict the exact magnitude of these changes, they are likely to be large, particularly over the long-term.
r ecommendations
Various transportation policy and planning reforms can improve public safety, fitness, and health by creating more efficient and multi-modal transportation systems where people drive less and rely more on alternative modes.47 Improved public safety, fitness and health are just three of many possible justifications for these reforms: they would help solve a variety of transportation problems, they reflect market principles and so increase economic efficiency, and they respond to changing consumer demands.48
The following are specific policies and planning strategies that can help create more diverse, more efficient, and healthier transportation systems:
• Educate decision makers concerning the relationships among transportation, land use, and public health; the full benefits of a more diverse, less automobile-dependent transportation system; and the trends that are changing future travel demands and strategic objectives.49 These all tend to increase the value of alternative modes, mobility management solutions, and smart growth land use development.
• Create a strategic vision of a more efficient and diverse transportation system and supportive land use development to accommodate changing demands and planning objectives, including public health objectives. This vision, which should be created by the federal government, should guide individual transportation and land use policies and planning practices, such as how transportation system quality is evaluated and how transportation funding is allocated.
• Increase public transportation funding for capital and operation costs. Transportation funding practices that currently favor investments in roads and parking facilities should be changed to allow significant new investments
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 5 6
>
>
C h
a p
te r
3
Public Transportation and Health
in public transportation. For example, economic stimulation and other economic development funds should be invested in public transportation. Transportation funds currently dedicated to roadways should be spent on public transportation improvements whenever it is more cost effective overall, taking into account all benefits and costs. Similarly, resources currently spent by governments and developers on parking facilities should be reinvested in public transportation whenever it is a cost-effective way to provide access. New funding sources should be developed to help finance public transit improvements, including parking taxes, congestion pricing, local property taxes, land value capture, and dedicated sales taxes.50 Higher levels of government (federal and state) should provide grants that leverage additional regional and local match funding. Regional and local governments must create stable sources of transit funding through dedicated fuel, sales, property, and parking taxes.
• Improve public transportation affordability. Insure that public transit services are affordable, particularly for lower- income users. This may include targeted discounts and exemptions, and research to identify better ways to meet the mobility needs of economically, physically and socially disadvantaged people.
• Establish transportation and land use policies that support transit oriented development so that more people are able to live and work in areas with high-quality public transportation services, good walking and biking conditions, compact and mixed land use development, and other supportive features.
• Implement transportation and land use policies that increase housing affordability in transit oriented communities.51 This includes changing development practices to encourage development of more compact and diverse housing types (small-lot single- family, townhouses, multi-family, etc.) with
unbundled parking in transit-rich, walkable areas with mixed land use and appropriate public services (schools, shops, parks, etc.), and employment.52 Public infrastructure investments and housing subsidies should be structured to support these objectives.
• Improve walking and bicycling conditions and promote active transportation. Encourage transportation professionals to recognize the importance of walking as a transport mode and to develop tools for evaluating the full benefits of improved walking and biking conditions and increased active transportation. Improve walking and bicycling access to transit stops and stations. Have bike racks on buses and trains, bike parking at stations, and bike rental services. Promote “walk and bike to school” and community walking and cycling events.
• Work to integrate affordable housing and affordable transportation so that physically, economically, and socially disadvantaged households can live in accessible, multi-modal communities. This requires a suitable mix of housing (affordable and subsidized housing included), public services (stores, medical and dental clinics, schools, parks, etc.), and high-quality public transportation located within convenient walking distance, with universal design features to ensure that everybody (including people using wheelchairs, walkers, pushing strollers, and hand carts) can easily travel to common destinations.
• Develop and apply multi-modal level-of- service standards to evaluate the service quality of various modes, including walking, biking, public transportation, taxi, car-sharing, and telecommunications within a community. Transportation agencies and professionals should use these to identify mobility and accessibility problems, particularly for the most vulnerable populations (children, older adults, people with disabilities, people with low incomes, immigrants, etc.).
• Apply least-cost planning so that transportation improvement resources (public funds and land) are invested in the most cost-effective improvements and consider all impacts and objectives, including public health objectives. Allow funds currently dedicated to roads and parking to be used for alternative modes and management strategies when they are more beneficial overall or support strategic planning objectives.
• Implement mobility management strategies and programs that encourage the use of alternative modes, such as efficient road and parking pricing, distance-based vehicle fees, and commute-trip reduction programs. Implement these in conjunction with transit service improvements.
• Develop and apply more comprehensive transportation planning tools for evaluating transit service quality,
transportation affordability, basic mobility, equity, affordability, and public health impacts.
• Sponsor research to improve public transit vehicles so that they are quieter, smoother, more spacious, climate controlled, less polluting, and easier to board; they should accommodate people with disabilities and offer amenities such as wireless Internet service. Give transit priority in traffic (bus lanes and signal control systems).
• Sponsor research and development to improve transit stops and stations so that they are more spacious, more comfortable, and safer; they should include amenities such as washrooms and refreshments.
• Develop convenient, integrated fares (for example, one payment system that can be used on various public transportation systems within a region) using electronic payment systems.
• Improve transit user information and marketing, such as real-time vehicle arrival signs, better-way finding, and culturally appropriate promotion programs.
• Apply more efficient parking management, such as efficient sharing, regulation, and pricing of parking facilities. Apply more flexible and reduced minimum parking requirements in transit oriented areas, particularly to increase housing affordability.
• Build coalitions involving public health and safety advocates and other interest groups that can benefit from transportation policy and planning reforms creating more efficient and diverse transportation systems—existing transit and community advocacy groups, transportation professionals, environmental organizations, local public officials, and economic development advocates. Use these coalitions to create the political support needed to achieve this vision.
ch. 3
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 5 7
<
<
P u
b li
c T
ra n
s p
o rt
a ti
o n
a n
d H
e a
lt h
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 5 8
>
>
C h
a p
te r
3
Public Transportation and Health
Reforms and Actions
Leaders Federal Legislative Role
Educate decision makers
Professional and advocacy organizations
Support policy analysis, research, and information sharing
Create a strategic vision
All levels of government; professional and advocacy organizations
Establish a national vision and encourage other levels of government to develop complementary visions
Increase public transportation funding
All levels of government Change transport funding to support public transportation, increase federal funding for public transportation programs, and use federal policies to leverage funding by other levels of government
Insure public transport affordability
All levels of government Provide funding, research and other support to insure that transit service is affordable and responds to the needs of disadvantaged people.
Support transit oriented development
All levels of government; transportation and land use planning agencies and professions
Change transport and land use policies to support transit oriented development and smart growth
Improve walking and cycling conditions
All levels of government; transportation and land use planning agencies and professions
Change transport funding and planning practices to support active transportation and walkable community development
Integrate affordable housing and affordable transportation
All levels of government Change transport and housing policies to support development of affordable housing in transit oriented areas
Apply multi- modal level- of-service standards
All levels of government; transportation agencies and professions
Change transport funding and planning practices so they are based on multi-modal performance evaluation
Apply least-cost planning
All levels of government; transportation agencies and professions
Change transport funding and planning practices to allow alternative modes and mobility management strategies to be funded whenever they are most cost effective, considering all impacts and objectives
Implement mobility management strategies and programs
All levels of government; transportation agencies and professions
Change transport funding and planning practices to support mobility management whenever it is cost effective, considering all impacts and objectives; support pricing reforms such as increased fuel taxes, road pricing, and distance-based insurance and registration fees
Table 4. Healthy Transportation Policy Implementation
ch. 3
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 5 9
<
<
P u
b li
c T
ra n
s p
o rt
a ti
o n
a n
d H
e a
lt h
Implementing these reforms will require action by various stakeholders, including federal, state, regional, and local governments, as well as diverse interest groups and advocates. Federal legislation can help support many of these reforms and actions by providing guidance and incentives. Such leadership and guidance can significantly accelerate the implementation of these reforms and avoid conflicts between existing and desired transportation policies. Table 4 indicates the level of government,
organization, or interest group that can provide leadership for implementing these recommendations and outlining the role of federal legislation.
Convergence Oppor tunities
Many interest groups and organizations with a wide range of objectives and perspectives have reasons to support policies to create a more efficient and diverse transportation system.
Reforms and Actions
Leaders Federal Legislative Role
Develop more comprehensive transportation planning tools
All levels of government; transportation agencies and professions
Support research for more comprehensive transport planning tools
Improve transit vehicles
Vehicle engineers, manufacturers, transit agencies, and governments
Support research; develop procurement guidelines
Improve transit stops and stations
All levels of government; transportation and land use planning agencies; private companies; and developers
Support innovative design and business models; support transit oriented development
Develop convenient, integrated fares
Regional governments and transit agencies
Support research, design, and implementation
Improve transit user information and marketing
Regional governments and transit agencies
Support research, design, and implementation
Apply more efficient parking management
All levels of government; transportation and land use planning agencies; private companies; and developers
Support transit oriented development and smart growth; provide incentives for local and regional governments to implement parking management
Build coalitions Professional and advocacy organizations
N/A
This table indicates how various stakeholders can help implement transportation policy reforms to improve public fitness and health. Public transit improvements can play a key role in many of these strategies.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 6 0
>
>
C h
a p
te r
3
Public Transportation and Health
This diverse interest offers an opportunity to build broader support for transit investments and supportive transportation and land use policies. For example, this is an ideal time to create collaborations among existing public transportation and community advocacy groups (wanting to achieve equity objectives), transportation professionals (wanting to reduce problems such as traffic and parking congestion), environmental organizations (wanting to reduce energy consumption, pollution emissions, and land use damages), local public officials (wanting to support urban redevelopment), senior advocacy groups (wanting to improve mobility options for nondrivers, to increase affordability, and to provide practical ways for older Americans to safely exercise), and health professionals (wanting to improve public fitness and health).
To fully achieve the potential benefits of high- quality public transportation, these diverse interest groups will need to overcome cultural and practical barriers. For example, correcting existing policy and planning biases that favor mobility over accessibility and automobile transportation over other modes will probably require a combination of professional education, planning agency reforms, and political advocacy to change laws and funding practices. No single interest group can achieve all these changes, but a collaborative effort can succeed.
Public transportation improvements can play a much greater role in creating a more diversified and efficient transportation system than indicated by its relatively modest share of total travel. High-quality public transportation often provides a catalyst for creating a more diverse transportation system and accessible, multi-modal land use development. Public transportation travel both supports and is supported by walking and biking trips. As a result, public transportation improvements can leverage large reductions in automobile travel and increases in walking and cycling activity.
The involvement of health professionals can significantly improve the chances for success because they can contribute a new sense of urgency, expertise, and leadership into transportation and land use policy reform debates. Previous public health successes, such as reduced tobacco use and increased breastfeeding, can provide models.
Conclusion
Transportation planning decisions impact public health in various ways: by affecting traffic risk, pollution exposure, physical activity and fitness, community cohesion, mental health, basic mobility, and affordability. Communities where people drive less and rely more on alternative modes are healthier places to live and work, particularly for physically, economically, and socially disadvantaged people. Transportation policy and planning reform improvements can play a significant role in creating healthier communities. High-quality public transportation (convenient, comfortable, frequent, fast, reliable, and safe) provides significant direct benefits when people shift from automobile to transit for individual trips. It provides even larger indirect benefits by providing a catalyst for development of more accessible, multi- modal communities where people own fewer
ch. 3
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 6 1
<
<
P u
b li
c T
ra n
s p
o rt
a ti
o n
a n
d H
e a
lt h
automobiles; drive less; and rely more on walking, biking, and public transportation for utilitarian trips and recreation.
This is a timely issue. Current demographic, economic, and market trends are reducing the demand for automobile travel and increasing the demand for alternative modes. This is not to suggest that Americans will give up driving altogether; but at the margin, that is, relative to current travel patterns, many people would prefer to drive less and rely more on alternative modes, provided that these alternatives are convenient, comfortable, safe, and affordable. This means that many consumers will choose healthier transport habits if given appropriate options, including high-quality public transportation and accessible, multi-modal communities.
Current transportation and land use planning practices favor automobile transportation and undervalue alternative modes and smart growth development. Various transportation policy and planning reforms can help achieve public health and social equity objectives by helping to create more diverse and efficient transportation systems. More comprehensive analysis is needed that accounts for the additional indirect costs of policy and planning decisions that increase automobile travel and sprawl and the additional indirect benefits of more compact, walkable, and transit oriented communities. Current funding is inadequate, causing public transportation service quality to decline and fares to increase in many communities. Budgeting practices must be reformed to provide adequate, reliable funding to ensure high-quality and affordable public transportation services. Land use development policies should change to better support smart growth and reduce sprawl.
These reforms are justified for a number of reasons, due to the diverse economic, social, and environmental benefits provided by public transportation improvements. When all impacts are considered, improving public transportation may be among the most cost-effective ways to improve public health, and improving public health is one of the best reasons to improve public transportation.
p g
. 6 2
>
>
Walking, Bicycling, and Health ch. 4 SUSAN L. HANDY, Ph.D. Professor, Department of Environmental Science and Policy University of California Davis, CA
ABSTRACT >> Walking and bicycling are efficient modes of travel and effective forms of exercise. Starting with the passage of the Intermodal Surface Transportation Efficiency Act (ISTEA) in 1991, the federal government has provided various forms of financial support for non-motorized transportation, but increasing walking and bicycling without increasing fatalities and injuries requires more than the limited federal resources to date. State, regional, and local policies determine the extent to which communities capitalize on the federal programs to expand walking and bicycling and help close the gap in health disparities between low-income communities and their more affluent neighbors. To increase non-motorized modes of travel—travel by walking and bicycling— safely, the authorization of the next federal transportation bill should:
• Assist: by providing state, regional, and local governments with the tools they need to plan for non- motorized travel
• Enable: by making it easier for state, regional, and local governments to spend federal funding on non-motorized modes
• Encourage: by providing incentives for state, regional, and local governments to pay more attention to non- motorized modes
• Require: by putting in place policies that compel state, regional, and local governments to improve conditions for non- motorized modes
Increased walking and bicycling would yield many health benefits and reduce disparities in health for low-income communities and others. The federal transportation bill can establish policies that will help to achieve the goal of increasing walking and bicycling safely.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 6 4
>
>
C h
a p
te r
4
Walking, Bicycling, and Health
CONTENTS
Introduction .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 65
Health and Non-motorized Transportation.. .. .. 68
Transportation Goals . .. .. .. .. .. .. .. .. .. .. .. .. .. 70
Strategic Targets .. .. .. .. .. .. .. .. .. .. .. .. .. .. 70
Measuring Progress.. .. .. .. .. .. .. .. .. .. .. .. .. 73
Transportation Policy: Opportunities and Barriers .. .. .. .. .. .. .. .. .. 74
Convergence Opportunities .. .. .. .. .. .. .. .. .. .. 77
Conclusion.. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 77
LIST OF ILLUSTRATIONS
Figures
1. Share of Trips by Walking, Bicycling, and Transit, by Country .. .. .. .. .. .. .. .. .. .. .. 65
2. Percent Usually Bicycling to Work in Selected U.S. Cities, 2000 .. .. .. .. .. .. .. .. 66
3. Cyclist Fatality and Injury Rates, by Country .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 69
4. Percent Walk and Bike Trips by Trip Length, Germany vs. United States .. .. .. .. .. .. .. .. .. 71
5. Trends in Mode of Travel to School in the United States, 1969–2001 .. .. .. .. .. .. 72
Tables
1. Factors Influencing Non-motorized Travel.. .. 67
2. Recommendations for Federal Policy on Walking and Bicycling .. .. .. .. .. .. .. .. .. .. 76
ch. 4
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 6 5
<
<
W a
lk in
g ,
B ic
y c
li n
g ,
a n
d H
e a
lt h
introduction
Walking and bicycling as modes of transportation—known as “non-motorized” or, more recently, “active” travel—are low- cost, low-polluting, calorie-burning, health- improving alternatives to driving. Despite these advantages, non-motorized modes represent a small share of all travel in the United States, or fewer than 10 percent of all daily trips in urban areas as of 2001.1 Increasing this number, without a congruent increase in fatalities and injuries, would yield considerable benefits, especially among low-income communities and people of color, the young and older adults, by helping to close wide gaps in health in this country. But what policies would achieve this aim?
For guidance, we can look to other developed countries, where rates of walking and bicycling are significantly higher than in the United States, particularly in Denmark, Germany, and the Netherlands (figure 1). We can also look to communities in the United States, where bicycle commuting is significantly more common than the national average of less than one percent of workers (figure 2). Common to these places is a supportive environment combined with a population motivated to walk and bicycle. These conditions have not come about by chance; they are the outcome of aggressive policies that address both environment and motivation.3
Figure 1. Share of Trips by Walking, Bicycling, and Transit, by Country 2
* work trips only ** walk and bike combined for Spain Source: D. Bassett et al., “Walking, Cycling, and Obesity Rates in Europe, North America, and Australia,” 2008.
Transit
Bike
Walk
0%
10%
20%
30%
40%
50%
60%
2
12%
1 9
8
14%
19%
26%
30% 30%
35% 36%
39% 39% 40%
42% 43%
44%
47%
52%
11
1 7
13
2
11
6
8
16
8
3
19
9
2
24
10
4
22
8
15
16
8
9
22
8
9
23
17
4
21
11
9
23
9
11
24
12
35
5
25
22
1 5
Net her
lan ds (
20 06
)
Sp ain
(2 00
0) **
Fin lan
d (1 99
9)
Sw en
den (2
00 6)
Austr ia
(2 00
5)
Ger m
an y (
20 02
)
Fin lan
d (2 00
5)
Den m
ar k (
20 03
)
Norw ay
(2 00
1)
UK (2 00
6)
Fr an
ce (1
99 4)
Belg iu
m (1
99 9)
Ire lan
d (2 00
6) *
Can ad
a (2
00 1)
*
Austr ali
a ( 20
06 )*
USA (2
00 1)
Figure 1. Share of Trips by Walking, Bicycling, and Transit, by Country
4-1
*Work trips only **walk and bike combined for spain
Source: J. Pucher and L. Buehler, “Making Cycling Irresistible,” 2008.
PE R
C E
N T
O F
T R
IP S
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 6 6
>
>
C h
a p
te r
4
A concerted and sustained effort is required to motivate people to walk and bike more and make their environment more conducive to doing so. The quality of the pedestrian and bicycle environment depends on several elements (see table 1), including land use patterns, network configuration, and facility design, all of which play an important role and are shaped by public investments and development policies over time. Natural features, particularly weather and topography, are also important, though obviously beyond the direct reach of policy. Motivation to
walk or bicycle also depends on personal characteristics—ability, comfort, confidence, habits, and perceptions—that can evolve over one’s lifespan but may also be modified by targeted intervention programs. Community norms also affect individual motivation but may be difficult to shift. Despite the challenges, a growing number of cities have demonstrated that it is possible to assemble a cost-effective package of policies, projects, and programs addressing both environment and motivation that significantly increases non-motorized travel.4
Walking, Bicycling, and Health
0
3
6
9
12
15
Irvine, CAIthaca, NYTuscon, AZMadison, WISanta Barbara, CASan Luis Obispo, CASanta Cruz, CAEugene, ORBerkeley, CAPalo Alto, CABoulder, COCorvallis, ORDavis, CA
4-2
Irv in
e, CA
Ith ac
a, NY
Tu cso
n, A Z
M ad
iso n, W
I
Sa nta
B ar
bar a,
CA
Sa n Lu
is Obisp
o, C A
Sa nta
C ru
z, CA
Eu gen
e, OR
Ber ke
ley , C
A
Pa lo
A lto
, C A
Bould er
, C O
Corv all
is, O
R
Dav is,
C A
14%
7% 7%
6% 6% 6%
4% 4%
3% 3%
2% 2%
1%
Figure 2. Percent Usually Bicycling to Work in Selected U.S. Cities, 2000
Source: 2000 U.S. Census, as compiled by the author.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 6 7
<
<
W a
lk in
g ,
B ic
y c
li n
g ,
a n
d H
e a
lt h
ch. 4
Two converging forces make this the right time to elevate non-motorized modes of travel. First, with health, economic, and environmental concerns on the rise, there seems to be a renewed interest in bicycling as evidenced by increased attention in the popular media. Second, Congress is now considering the authorization of the federal transportation bill, the Safe, Accountable, Flexible, Efficient Transportation Equity Act: A Legacy for Users, or SAFETEA-LU, which will set policy and dictate
funding levels for surface transportation well into the next decade. These forces together create an unprecedented opportunity to work toward the goal of increasing safe non- motorized travel.
Category Factor Definition Importance
Environmental Land use patterns The arrangement of land uses such as housing, shops, offices, etc., across the community
Determines the straight-line distance among different activities, such as housing, shopping, and offices
Network structure The layout of streets and trails throughout the community
Determines how direct the connections from one place to another are and thus influences the travel distance
Facility quality Characteristics of streets, including presence of sidewalks and bike lanes, widths, pavement conditions, crosswalks, signals, etc.
Influences how comfortable, safe, and attractive it is to walk or bicycle that route
Natural features Topography, weather, scenery
Influences the energy needed to walk or bicycle as well as comfort and enjoyment
Motivational Individual factors Ability, experience, comfort level, confidence, preferences, habits, etc.
Influences the willingness and desire of an individual to walk or bike
Community norms Social acceptability of bicycling, dominant attitude toward bicycling, bicycling culture
Influences the willingness and desire of an individual to walk or bike
Table 1. Factors Influencing Non-motorized Travel
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 6 8
>
>
C h
a p
te r
4
Hea lth a nd non-Motorized Tra nspor tation
Whether for transportation or recreation, walking and bicycling are important forms of physical activity. Federal guidelines categorize brisk walking and bicycling on level ground as moderate physical activity, while bicycling at more than 10 miles per hour qualifies as rigorous physical activity. The U.S. Department of Health and Human Services (DHHS) recommends that children engage in 60 minutes of physical activity each day and that adults engage in two hours and 30 minutes of moderate physical activity per week,5 a standard that more than one- third of all adults nationwide fail to meet.6 A 15-minute non-motorized commute twice a day for five days a week is enough to meet the adult recommendations. The DHHS identifies walking and biking as effective measures for increasing overall physical activity and notes that non-motorized commuting has a low risk of injury compared to many other forms of physical activity. Walking, in particular, has been described by health researchers as “near perfect exercise”7 and “a popular, familiar, convenient, and free form of exercise that can be incorporated into everyday life and sustained into old age.”8 The health benefits of achieving the recommended levels of physical activity are numerous: prevention of weight gain; improved cardio respiratory and muscular fitness; and lower risk of type 2 diabetes, heart disease, stroke, and other unhealthy conditions.
From an equity standpoint, non-motorized transportation presents both challenges and opportunities. Non-motorized modes can improve access to jobs, healthcare, and shopping for households with limited access to cars. Additionally, walking and bicycling reduce health disparities between low-income and more affluent communities. Safety, however, remains a significant concern: in 2007, there were 4,654 pedestrian and 698 bicyclist fatalities in the United States, with combined
injuries of more than 100,000.9 Indeed, public officials often use safety concerns to beat back arguments to do more to encourage walking and bicycling. The challenge is to increase non- motorized modes safely, primarily because the population groups that could most benefit from increased walking and bicycling are also the most vulnerable to traffic dangers.
Low-income and minority populations fall into this category. Ample evidence indicates that physical activity levels are lower among low-income and minority populations,10 despite the fact that only 73.5 percent of low- income households own cars and are more dependent on walking and public transit. That number compares with 91.7 percent of all U.S. households. Forty percent of the lowest-income transit users meet the recommended levels of physical activity solely from walking to and from transit.11 Without this, their total physical activity would be far less. However, the quality of non- motorized infrastructure is often lower in low- income and minority communities, contributing to higher pedestrian fatality rates.12 The confluence of these circumstances underscores the importance of improving walking and bicycling conditions in these communities.
Youth are also vulnerable. Across the country, adolescents depend on parents and other adults to drive them to school and other activities.13 If children were able to walk or bike more, they would get more physical activity and their parents (predominantly mothers) would have less need to drive them. Again, however, safety is a concern: rates of pedestrian and bicyclist fatalities and injuries per capita are highest for those under the age of 15.14 Parental fears about traffic as well as fear of abductions help explain why children now walk and bike less than in the past. Consequently, increasing walking and bicycling for children means removing threats— actual and perceived—to their safety.
Older adults, too, could benefit from increased walking and bicycling, but safety, once again, is an issue. One in five adults ages 65 years and
Walking, Bicycling, and Health
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 6 9
<
<
W a
lk in
g ,
B ic
y c
li n
g ,
a n
d H
e a
lt h
ch. 4
older does not drive, and more than 50 percent of the nondrivers stay home on any given day because they lack transportation options.15 For nondrivers, walking, bicycling, and transit can provide an important means of getting to the doctor’s office, the store, or a friend’s house. However, the decline in physical and mental abilities that make driving no longer safe can also make walking and bicycling less safe. Uneven sidewalks, for instance, can pose a perilous hazard to frail older adults. The highest rate of pedestrian fatalities per capita is for those over age 70.16 Where safe conditions exist, increased walking and bicycling can improve physical and mental health.17
The good news is that safety is likely to improve for low-income households, children, older adults, and others as more people walk and bicycle. Countries with high levels of non- motorized travel also have fewer fatalities and injuries per mile than does the United States (figure 3).In part, this difference is explained by better infrastructure, particularly the separation of pedestrians and bicyclists from motor vehicles. But the higher number of pedestrians and bicyclists using thoroughfares itself improves safety by heightening driver awareness and attentiveness.19 Larger numbers of pedestrians and bicyclists also spur elected officials to invest more in better, safer infrastructure, which, in turn, helps to encourage more walking and bicycling.
Figure 3. Cyclist Fatality and Injury Rates, by Country 18
Note: The symbol // in the graph represents a break in the consecutive numbering of the Y-axis. Source: Pucher and Buehler, “Making Cycling Irresistible,” 2008.
Cyclists killed per 100 million kilometers cycled
Cyclists injured per 10 million kilometers cycled
0
5
10
30
20
25
30
35 40
1.1 1.4 1.5 1.7 1.7
4.7
3.6
6.0 5.8
37.5
Figure 3. Cyclist Fatality and Injury Rates, by Country
Source: Pucher and Buehler, “Making Cycling Irresistable,” 2008.
Netherlands Denmark Germany United Kingdom USA
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 7 0
>
>
C h
a p
te r
4
The potential economic benefits of increased walking and bicycling are numerous. Improved health as a result of increased physical activity can reduce healthcare costs. Cheaper modes of travel can reduce household spending on transportation: the typical household in this country spent an average of $7,896 to own and drive their cars in 2005.20 Making walking and bicycling more viable, particularly in conjunction with improvements to transit, could increase access to jobs. Improvements to walking and bicycling facilities can contribute to economic development efforts by, for example, encouraging stores to locate within walking distance of residential areas, particularly in low- income areas.
The potential environmental benefits of non- motorized modes are also abundant and include reductions in air pollution, water pollution, noise, and greenhouse gas emissions. However, these benefits accrue only if the increase in the use of non-motorized modes comes with a reduction in the use of motorized modes. A substantial share of walking and bicycling in the United States is for recreation rather than for transportation, and even some non-motorized trips to destinations are made in addition to, rather than instead of, driving trips.21 Walking and bicycling trips that do not replace driving trips do not have a direct environmental benefit, though they still have important health benefits.
Tra nspor tation G oa ls
The goal for non-motorized modes is straightforward: increase walking and bicycling without increasing fatalities and injuries, particularly for low-income households, communities of color, the young, and older adults. But what is a realistic increase to aim for? Although walking and bicycling have virtually boundless potential as forms of recreational physical activity, their potential as modes of transportation are limited by practical constraints. Given the low levels of use in this country, significant increases as a percentage of
all travel may be possible even if they remain a relatively small share of all trips. The potential for the two modes is likely different: walking is possible for more people because it requires no equipment and less confidence and skill, but it is considerably slower than bicycling; bicycling is at least theoretically possible for more trips because it is considerably faster than walking, but it requires equipment as well as skills and confidence that many lack. Given the low- density patterns of development in the United States, which put destinations beyond walking distance in most places, bicycling seems to offer greater potential for expansion.
Strategic Targets
In aiming to increase safe non-motorized modes of transit, particularly among those with the greatest needs but also the greatest vulnerabilities, it makes sense to take a strategic approach and target the following: types of travel most conducive to non-motorized modes, communities with greater potential for change, and communities with greater potential benefits from change.
Short trips are an obvious target. According to the 2001 National Household Transportation Survey, 28 percent of all trips are less than one mile, a reasonable distance for walking, and 41 percent of trips are less than two miles, a distance that is reasonable for biking.22 The shares of these short-distance trips that are made by non-motorized modes are much lower in the United States than in European countries: 71.4 percent of trips shorter than one mile are by walking or bicycling in Germany versus 31.2 percent in America (figure 4). In other words, while trip distances are longer on average in the United States than in Europe, distance is not the only issue; environmental and motivational factors must explain differences in non- motorized rates at these short distances.
School trips are another obvious target and, indeed, the federal Centers for Disease Control and Prevention has set a goal of increasing
Walking, Bicycling, and Health
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 7 1
<
<
W a
lk in
g ,
B ic
y c
li n
g ,
a n
d H
e a
lt h
ch. 4
walking to school. This makes sense from a practical standpoint, given that these are frequent trips with regular routes and fixed destinations. Walking to school dropped from 40.7 percent of all school trips in 1969 to 12.9 percent in 2001, while bicycling remained roughly constant at around one percent (figure 5). Increasing walking and biking to school is generally a good starting point for increasing physical activity in children. For example, it could contribute to an increase in non-motorized travel to other destinations, as skills and habits change. Current efforts fall into two categories: changes in where schools are located to put more children within walking distances of school, and Safe Routes to School programs, which aim to improve safety around schools for walkers and bicyclists.
Some communities have greater potential for change than others. One target should be areas where walking and bicycling are already significant. For example, Davis, CA, has high levels of bicycling, but levels could clearly be even higher. The environment there supports bicycling, but not all residents take advantage of the opportunity: over three-fourths of children are driven to their Saturday morning soccer games.25 Motivational rather than environmental barriers are often the issue—habit, perceptions, confidence, etc. A second target should be places where land use patterns put destinations within walkable or bikeable distances of homes, that is, areas with higher densities and mixed land uses. In these places, the quality of sidewalks and other facilities may be a problem
Figure 4. Percent Walk and Bike Trips by Trip Length, Germany vs. United States 23
PE R
C E
N TA
G E
O F
T R
IP S
0%
10%
20%
30%
40%
50%
60%
70%
80%
U.S.A. Germany
Bike
Walk
< 1 MILE < 2 MILE < 3 MILE
U.S.A. Germany U.S.A. Germany
2.3
31.2%
28.9
14.5
71.4%
56.9
0.9
6%
5.1 0.5 2.4%
19.6%
1.9
9.3
10.3
13.5
31.9%
18.4
Figure 4. Percent Walk and Bike Trips by Trip Length, Germany vs. United States
4-4
Source: R. Buehler, “Transport Policies, Travel Behavior, and Sustainability,” 2008.
Source: R. Buehler, “Transport Policies, Travel Behavior, and Sustainability,” 2008.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 7 2
>
>
C h
a p
te r
4
in addition to motivational barriers.
Of lower priority, because they are harder to change, are low-density areas with limited walking and bicycling infrastructure, particularly rural areas. In these areas, however, it is still important to look for specific opportunities to reduce environmental barriers, e.g., by improving the shoulders of rural roads or through a trail project that connects rural residents to the town center. Finding such opportunities should be more of a priority in areas where residents have limited access to cars and where transit service is sparse or nonexistent.
Potential benefits from increases in non- motorized travel are greater in some areas than others. Increases are most important in low-income and minority communities, where efforts are needed to improve safety when residents of these communities do walk and bicycle and to make more places accessible by these modes. Bicycling, in particular, offers a way to fill the gap between places accessible by foot and those accessible by bus. Anecdotal evidence suggests that bicycles are an important mode for recent Hispanic immigrants in California, though bicycling often occurs in environments not designed for it.26 Hispanics walk and bike to work in greater shares than
Walking, Bicycling, and Health
1969
0
10
20
30
40
50
60
Walk/Bike
Public Transit
School Bus
Auto
1977 1983 1990 20011995
0
10
20
30
40
50
60
Public Transit
School Bus
Auto
Walk/Bike
Figure 5. Trends in Mode of Travel to School in United States, 1969–2001 24
Source: N. C. McDonald, “Active Transportation to School,” 2007.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 7 3
<
<
W a
lk in
g ,
B ic
y c
li n
g ,
a n
d H
e a
lt h
ch. 4
other Americans; not surprisingly, their rates of pedestrian and bicycle fatalities are also higher.27 Environmental improvements are essential in these communities.
Retirement communities, formal or informal, are another important target. It used to be that those who aged in place lived mostly in older communities that were designed for walking. Increasingly older adults now live in suburban environments that are not designed for walking. Improving the walking environment in these areas is not easy, though strategic projects coupled with programs to encourage walking or even bicycling could make a difference. In so- called active retirement communities, bicycling could be encouraged over golf carts as a way to get around within the community.
Measuring Progress
Achieving the goal of an increase in walking and biking safely requires development of new performance measures, both to assess current conditions and to monitor the effectiveness of new policies. Traditional transportation performance measures focus on vehicle traffic in support of the goal of maximizing vehicle flow and to the detriment of walking and bicycling. Without performance measures for non- motorized travel, policies are likely to continue to favor cars over pedestrians and bicyclists; transportation goals for which performance is not measured will get less attention in the planning process.28
Admittedly, developing such measures is difficult. If the goal—the desired outcome— is to increase walking and bicycling without increasing fatalities and injuries, then these factors are what should be measured. But increases in non-motorized travel are hard to measure.29 The best available data come from travel surveys, conducted at the regional or national level. Yet non-motorized trips have historically been undercounted in these surveys, which have primarily been concerned with driving trips. The surveys are also not frequent
enough to be useful for annual monitoring (the national survey occurs every five to seven years, while regional surveys are typically separated by 10 years or more). Although data on fatalities and injuries are arguably better than data on the amount of walking and bicycling, without the latter, it is impossible to adequately gauge the former. For example, the numbers of pedestrian and bicyclist fatalities and injuries have been going down on a per capita basis,30 but this likely reflects a decline in the use of these modes rather than a decline in danger. Improved data collection is needed.
As an alternative to measuring increases in non- motorized travel, performance measurement might focus on what might be called inputs rather than outcomes. One input is funding for bicycle and pedestrian projects. Another is the adoption of policies to promote non- motorized transportation, such as changes in zoning designed to bring about mixed-use land use patterns that reduce walking distances, or complete street policies that ensure that bicycles and pedestrians are given consideration in the design of all thoroughfares. Unfortunately, these inputs do not guarantee favorable changes in the environment, let alone the desired outcome of an increase in safe walking and biking. The input option for performance measures is the easiest to implement but the least effective in showing progress toward the goal.
An option that is better than measuring inputs but more feasible than measuring outcomes is to focus on outputs, that is, on changes in the environment that are expected to lead to increases in non-motorized travel, rather than changes in non-motorized travel that are difficult to measure. Outputs could be measured as projects actually constructed. However, non-motorized projects are not well tracked; categorizing such projects can be difficult, and bicycle and pedestrian improvements are often incorporated into larger road projects.31 Another option is to measure changes in the “walkability” or “bikeability” of a community. Many tools for measuring walkability and
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 7 4
>
>
C h
a p
te r
4
bikeability have already been developed,32 with increasingly frequent implementation in the transportation planning process. However, collecting data to calculate walkability and bikeability at a community scale can be labor intensive.
Tra nspor tation Policy: Oppor tunities a nd Ba rriers
The next authorization of the federal transportation bill offers a tremendous opportunity for non-motorized transportation. For almost two decades, federal policy has contributed to an expansion of investments in walking and bicycling infrastructure. However, many barriers have hindered progress toward the goal of increased walking and bicycling, including federal policy itself. The new transportation bill could overcome many of these barriers by putting in place stronger federal policy toward non-motorized modes.
Starting with the passage of the Intermodal Surface Transportation Efficiency Act (ISTEA) in 1991, the federal government has provided support for non-motorized transportation through a number of policies. Most importantly, federal transportation funding can be used for bicycle and pedestrian projects through the Transportation Enhancements (TE) Program, the CMAQ (Congestion Management and Air Quality) Program, the Surface Transportation Program (STP), the Safe Routes to School (SRTS) Program, the Non-Motorized Transportation Pilot Program, and several others, including the Highway Safety Improvement Program (HSIP).33
Other policies also support non-motorized modes. Federal policy specifies seven “planning factors” that must be considered in the development of long-range transportation plans at state and regional levels. These factors include increased safety and security for non- motorized users, increased mobility and
accessibility options, and increased integration of the transportation system across modes. States are also now required to have bicycle coordinators. Finally, the Federal Highway Administration has pushed the concept of context sensitive design, which has increased attention to bicycle and pedestrian needs.
Under current policies, however, the availability of federal funds is insufficient to ensure improvements to the walking and bicycling environment. State, regional, and local policy decisions determine the degree to which communities take advantage of the federal programs for bicycling and walking facilities. For example, through the regional transportation planning process, metropolitan planning organizations evaluate and prioritize regional needs and decide what share of federal funding in these categories will go to non-motorized projects. The availability of federal funds for bicycle and pedestrian facilities has created an important opportunity, but one that only some states and regions have taken advantage of. Indeed, spending on non-motorized projects has varied significantly across the major metropolitan regions, ranging from $0.20 per capita in Los Angeles to $2.32 per capita in Providence, RI, from 1992 through 2006.34
At the same time, many federal programs and policies hinder rather than support efforts to increase non-motorized travel.35 The TE program as administered by the states can present insurmountable bureaucratic hurdles, particularly for communities with limited resources. The CMAQ program requires proof of air quality benefits, yet the models used to forecast emissions are not usually sensitive to bicycle and pedestrian improvements. Most significantly, an overarching concern with congestion at the federal level as well as at state and local levels undervalues non-motorized projects relative to highway projects in the planning process. The current focus on job creation and economic stimulus also threatens to perpetuate the top priority given to highway projects.
Walking, Bicycling, and Health
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 7 5
<
<
W a
lk in
g ,
B ic
y c
li n
g ,
a n
d H
e a
lt h
ch. 4
One of the most intractable barriers to improving the walking and bicycling environment on a wide scale is local control of land use planning, a long-standing tradition throughout the country.36 The viability of non-motorized modes depends on land use patterns that put potential destinations within walking and bicycling distances of home. Similarly, transit viability increases as population and employment densities increase. These environmental characteristics are shaped by local policies such as zoning and subdivision ordinances. Investments in non-motorized infrastructure will be of little benefit without concomitant changes in local land use policies. Although land use planning authority is likely to remain at the local level for the foreseeable future, federal policy can and does influence the decisions of local governments, and this influence can be channeled toward the support of non-motorized modes.
Thus, federal policy alone will not bring about the needed changes, but it can help to expand non-motorized transportation by assisting, enabling, encouraging, or requiring agencies at the state, regional, and local levels to both improve the environment and
motivate people. To safely increase walking and bicycling, the upcoming authorization of the federal transportation bill should include the following policies, focusing on types of travel most conducive to non-motorized modes, communities with greater potential for change, and communities with greater potential benefits from change (see also table 2).
Assist: provide state, regional, and local governments with the tools they need to plan for non-motorized modes. Funding for more frequent and standardized travel surveys and for development of survey methods that collect more accurate and more comprehensive information on non-motorized modes would provide for better monitoring of progress. Such data could also provide a means of calibrating improved travel forecasting models that incorporate non-motorized modes. Resources should especially be directed towards low- income communities that may have a greater need for planning assistance.
Enable: make it easier for state, regional, and local governments to spend federal funding on non-motorized modes. Reducing bureaucratic barriers in current programs, particularly in the TE program, would likely increase the use of these funds for non-motorized projects, such as sidewalks and bicycle paths, particularly in low-income communities with fewer resources available for overcoming these barriers. Further increasing flexibility in federal programs would enable communities to give greater priority to non- motorized modes. In addition to infrastructure projects, educational and promotional programs should be eligible for funding.
Encourage: provide incentives to state, regional, and local governments to pay more attention to non-motorized modes. Specialized funding programs, such as Safe Routes to School, encourage spending on non-motorized modes. Targeted incentives, such as supplemental grants, could encourage attention to pedestrian and bicyclist needs, with
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 7 6
>
>
C h
a p
te r
4
priority given to low-income areas. Incentives that encourage coordination of land use and transportation planning could also enhance the viability of non-motorized modes; for example, jurisdictions that adopt land use policies promoting greater densities and mixed land uses might earn bonus funding for bicycle and pedestrian projects.
require: put in place policies that compel state, regional, and local governments to improve conditions for non-motorized modes. A federal complete streets policy would require that the needs of bicyclists and pedestrians are considered in all federally- funded projects. Federal transportation funding could be allocated based on the degree to which jurisdictions meet performance requirements for non-motorized modes. These
requirements could use the performance measures described earlier, such as increases in safe walkability and bikeability, with extra weight given to performance in lower-income areas and for key segments of the population. Performance standards could also be set with respect to land use policies; for example, jurisdictions might be eligible for funding only if they have adopted land use policies that are supportive of non-motorized modes.
As outlined, these approaches progress from least to most forceful; some combination of all four would have the best chance at success. But they must be accompanied by a shift in the focus of the federal program away from congestion reduction to goals related to health, equity, economic, and environmental benefits. Tying federal funding to demonstration of
Walking, Bicycling, and Health
Assist Help provide state, regional, and local governments with the tools they need to plan for non-motorized modes: fund travel surveys; support development of improved planning tools
Enable Make it easier for state, regional, and local governments to spend federal funding on non-motorized modes: reduce bureaucratic barriers; increase funding flexibility; expand eligibility of promotional programs
Encourage Provide incentives for state, regional, and local governments to pay more attention to non-motorized modes: continue and expand specialized funding programs; target incentives for prioritizing bicycle and pedestrian projects and for supportive land use policies
Require Put in place policies that compel improvements in conditions for non-motorized modes on the part of state, regional, and local governments: adopt federal complete streets policy; tie funding to performance requirements; tie funding to supportive land use policies
Table 2. Recommendations for Federal Policy on Walking and Bicycling
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 7 7
<
<
W a
lk in
g ,
B ic
y c
li n
g ,
a n
d H
e a
lt h
ch. 4
progress toward these goals would ensure that the shift in focus is not just rhetorical. Such an approach could provide a powerful mechanism for improving walking and bicycling conditions.
Convergence Oppor tunities
Credit for the existence of federal policies supporting non-motorized modes goes to a strong coalition of bicycle and pedestrian advocacy groups operating at the national level. This coalition is increasingly working in partnership with other interest groups, including those focused on public health, social equity, and environment issues, reflecting the broad benefits of non-motorized travel in all these realms, as described previously. This effective coalition is well positioned to influence the authorization of the upcoming federal transportation bill, though it must
continue to battle the traditional focus on congestion reduction and the new emphasis on highway investments as a way to stimulate the economy. Making the case that bicycle and pedestrian projects create jobs, too, while also helping to reduce our economically detrimental dependence on fossil fuels will be important for this coalition.
Because federal policy alone does not determine improvements to the bicycle and pedestrian environment, effective coalitions are also needed at the state, regional, and local levels. The local scale is especially important but also especially challenging, and the potential for building the needed partnerships varies from community to community. The Active Living by Design program, among others, has helped to foster such partnerships in communities throughout the country, including many low- income communities.37 The evaluation of this program should yield important lessons for other communities in their efforts to build partnerships in support of improvements to the bicycle and pedestrian environment.
Conclusion
A “perfect storm” of higher gas prices, strained household budgets, and declining public resources, coupled with emerging mandates to reduce greenhouse gas emissions and deepening concerns about the growing obesity epidemic, could produce a surge in interest in non-motorized travel modes. Indeed, recent media reports suggest that a new bicycling culture has begun to take hold. Surveys also suggest a growing interest nationwide in walkable communities.38 If federal, state, regional, and local lawmakers follow the public’s lead, walking and bicycling could move the United States toward a healthier, more equitable future.
p g
. 7 8
>
>
Roadways and Health: ch. 5 Making the Case for Collaboration CATHERINE L. ROSS, Ph.D. Director, Center for Quality Growth and Regional Development (CQGRD), and Harry West Chair
MICHELLE MARCUS, M.P.H. CQGRD Graduate Research Assistant Georgia Institute of Technology Atlanta, GA
ABSTRACT >> Our streets and highways are inextricably linked with the very fabric of America. Roadways are used for many different modes of transportation, and constitute a major portion of the public space in our towns and cities. The limited inclusion of health considerations in the operation and construction of our roadways results in negative health outcomes. Lack of safe, convenient walking and bicycling routes have led to sedentary lifestyles, feeding a massive epidemic of obesity and chronic diseases. Motor vehicle emissions contribute to many negative health outcomes including asthma, lung disease, and cardiovascular disease. Transportation is the fastest-growing source of green house gases in the U.S., adding to climate instability which can result in natural disasters, food scarcity, and premature deaths. In addition to environmental impacts traffic crashes result in nearly 42,000 deaths and three million injuries every year. The authorization of the federal transportation bill is an opportunity to increase resources and focus on improving the negative health consequences associated with roadway construction and use. Fundamental changes in the way we measure and rank mobility needs, distribute funding, design, construct, operate and evaluate our roadways are possible and necessary.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 8 0
>
>
C h
a p
te r
5
Roadways and Health
CONTENTS
Introduction .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 81
Connecting Roadways, Health, and Equity.. .. .. 82
Injury . .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 83
Impact .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 83
Mechanism .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 84
Mitigation: Reducing Injury .. .. .. .. .. .. .. .. .. 85
Environmental Quality .. .. .. .. .. .. .. .. .. .. .. .. .. 85
Impact .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 85
Mechanism .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 86
Mitigation: Improving Air Quality and the Environment .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 87
Mode Share .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 87
Impact .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 87
Mechanism .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 88
Mitigation: Diversifying Mode Share and Reducing Automobile/Roadway Use. .. .. .. .. .. .. .. .. .. 89
Federal Legislation: Equity, Health, and Highways . .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 89
Transportation Policy Barriers . .. .. .. .. .. .. .. .. .. 90
Transportation Policy Opportunities .. .. .. .. .. .. 91
Convergence Opportunities .. .. .. .. .. .. .. .. .. .. 92
Conclusion.. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 93
Appendix A: Policies and Strategies for Healthy Transportation .. .. .. .. .. .. .. .. .. 94
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 8 1
<
<
R o
a d
w a
y s
a n
d H
e a
lt h
ch. 5
introduction
A vast proportion of travel—and life—in the United States occurs on our roadways. This travel is made by car, foot, bicycle, wheelchair, bus, and streetcar. “Roadway” refers to the entire right-of-way—sidewalks, roadside, medians and verges, and in-street rails; it constitutes a major portion of the public space in our towns and cities. Roadways are used not only for transport, but also for socializing and support of public life. Our streets and highways are inextricably linked with the very fabric of America and impact our lives, cities, and environment in complex and pervasive ways. They have considerable impact on health and can be harmful if potential negative impacts are not mitigated.
Roadways, including highways, streets, and parkways, are linked to health outcomes in numerous ways. Foremost are physical inactivity, crashes, vehicle emissions, and equitable access to jobs and services. Lack of safe, convenient places and ways to walk and bicycle have led to sedentary lifestyles, feeding a massive epidemic of obesity and chronic diseases. Current levels of motor vehicle emissions contribute to many negative health outcomes, including increased incidence of asthma, lung disease, and cardiovascular disease. Increased levels of greenhouse gases, to which cars and trucks are a major contributor, are causing climate instability resulting in natural disasters, food scarcity, unhealthy ecological and weather patterns, and premature deaths. Traffic crashes result in nearly 42,000 deaths and three million injuries every year on American highways. Even the economic health of a community and its residents is affected by the cost, availability, and mode of transportation used for daily activities. Emotional well-being is challenged by traffic congestion, long and stressful commutes, and noise. Every community is affected, and often vulnerable populations face the greatest risk.
There is compelling evidence that poverty, race, ethnicity, disability, age, and urban or
rural setting are correlated with persistent and expanding health disparities among U.S. populations. The pursuit of good health requires safe and convenient access to a source of steady income, goods and services, and a wholesome environment. However, nearly one-third of Americans do not drive due to disability, age, financial constraint, or other personal circumstances. The majority is located in metropolitan areas, but even in rural areas about 14 percent of trips are made by those without access to a car.1 These Americans live in an automobile-oriented society without access to an automobile and are therefore both socially and economically disadvantaged. Their access to goods and services and their inclusion in the larger society are dependent on greater accessibility in the transportation system. The impending increase in the proportion of older Americans, constituting 20 percent of the population, will only add to this dependency. Without roadway system design and funding priorities that accommodate their travel needs, these individuals and their families often have limited access to jobs, hospitals, supermarkets, and more. Their level of access is also affected by land use patterns that have been formed by decades of automobile-oriented road planning and engineering.
Major roads and highways have turned into barriers as they become more difficult to cross by foot or by vehicle. Homes and stores have tried to withdraw from heavy motor vehicle traffic through use of the cul-de-sac and large setbacks from the edge of the street, reducing overall connectivity. Limited street connectivity forces use of a few heavily used, congested roadways, exposing travelers to greater risk from air pollution and car crashes. Cities have given over large tracts of valuable—and taxable—land to pavement for roads and parking that have depleted “Main Street,” drained the tax base, and created sprawling regions where businesses are dwarfed by their parking lots and roadways are often barren and dangerous. Designing for automobile use on every trip, no matter how short, has evolved into a self-reinforcing
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 8 2
>
>
C h
a p
te r
5
spiral of decentralizing communities, expanding pavement, and increasing per capita vehicle miles traveled (VMT). This trend has created many of the issues contributing to poor health outcomes.
Health is influenced by roads, but roads are influenced by infrastructure construction programs, public policy, and funding practices. A large proportion of funding and policy for roads is determined at the federal level; much of it is contained in the Safe, Accountable, Flexible, Efficient Transportation Equity Act: A Legacy for Users (SAFETEA-LU), which expires on September 30, 2009. The impending new authorization is an opportunity to make fundamental changes in the way we measure and rank mobility needs, the way we distribute funding, the way we conduct and design projects, and the way we evaluate our results. Ultimately, it is a chance to adopt powerful strategies that will help us achieve a healthy, equitable, and sustainable national infrastructure that supports robust economic development and the well-being of people and communities.
The upcoming authorization presents the opportunity to rethink transportation system design and operation in ways that are more supportive of positive health outcomes. Many of the policy changes that help achieve health objectives also address other planning objectives, including congestion reduction, road and parking facility cost savings, energy conservation, and economic development. For instance, designing our transportation system for shorter travel distances to enable walking and bicycling would increase physical activity, curb foreign oil dependence, and reduce the need for new or upgraded transportation facilities to accommodate vehicular travel. Given the importance of health to a viable, productive nation, and given the effect of transportation on health, we cannot reasonably design and fund our transportation system without addressing its health impacts.
Connecting r oa dways, Hea lth , a nd Equity
The impact of roadways on health is summarized by examining the level of injury (intentional and unintentional), environmental impact (climate change and air pollution), and mode share (including level of access, physical activity, and mental/social health). The mechanism, extent, and mitigation of roadway- related health impacts are detailed below, with additional attention to the distribution of these impacts across the population. The major principles for mitigating the health impacts of roadways are to reduce injury, improve air quality and the environment, diversify mode share, and reduce automobile dependency.
The following characteristics of roadways all have an impact on health:
• Modal Level of Service—refers to the proportion of roadway dedicated to each travel mode (automobile, bus and light rail, truck, bicycle, and pedestrian). While general
Roadways and Health
Highways built to past standards are unable to support safe multi-modal travel.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 8 3
<
<
R o
a d
w a
y s
a n
d H
e a
lt h
ch. 5
purpose lanes can be used by cars, trucks, buses, and bicycles, the inclusion of facilities intended exclusively for one of these modes can greatly modify the user behavior and utilization of the road. For instance, bicycle lanes or bus-only lanes may increase the safety and speed of travel in a corridor, and more people may choose these modes.
• roadway Design—focuses on features that impact behavior and safety. It addresses speed limit and design speed for motor vehicles, number and width of general purpose lanes (in each direction), presence of medians, and intersection design, including turn lane and free-flow turn/merge lane usage, corner radii, signal phasing, robustness of bicycle and pedestrian facilities, and more. A roadway will typically carry pedestrian and bicycle traffic, even if no facilities are provided for them.
• Access Management—refers to the regulation of interchanges, intersections, driveways, and median openings on a
roadway. Prohibiting turns or prohibiting certain users from part of or the entire road can improve operations. For instance, a left turn may be restricted to buses only, one leg of an intersection may be closed to pedestrians, or the quantity and placement of driveways along the roadway may be restricted. In doing so, conflicts between road users are reduced sometimes at the expense of freedom of movement. The right balance of access management can improve safety and level of service (LOS) for all road users.
• Streetscape—measures the degree of treatment of the roadway with trees and other plantings; placement of amenities such as lights, benches, and garbage cans; and general roadside appearance, including placement of buildings, artwork, or plazas. These influence motorist behavior, transportation access, and pedestrian and bicycle LOS.
• Density, Land Use, and Connectivity— refers to the types and intensity of uses along the roadway and the connectedness of the streets that support it. Research indicates that mixed land uses, higher land use density, and short block lengths have a strong relationship with higher levels of physical activity and social capital, as well as with lower levels of air pollution, greenhouse gas emissions, and fatal crashes.
injur y
Impact
There were 41,059 traffic-related deaths reported in the United States in 2007.2 This constituted the leading cause of death for individuals ages one to 34.3 After age 34, deaths from heart disease, stroke, and cancer—which are largely affected by physical activity levels, another outcome of transportation practices—exceed deaths due to traffic crashes. Additionally, crashes result in almost three million injuries per year. This creates an economic burden of about
Road policies have impacted land use.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 8 4
>
>
C h
a p
te r
5
$150 billion each year, including $52 billion in property damage, $42 billion in lost productivity, and $17 billion in medical expenses.4 Of the 41,059 traffic fatalities, 4,654 were pedestrians, 5,154 motorcyclists, and 698 bicyclists.5
Crashes were more likely to occur at an unsignalized intersection than a signalized one.6 Rural crashes were more likely to occur away from an intersection and appear to be most attributable to speed or driver distraction. In 2002, more than one-third of pedestrian travel took place on a roadway or shoulder. Crashes in urban areas alone result in about $160 billion in expenses (according to 2005 data) and may be responsible for half of the roadway congestion there.7
Vulnerable populations typically have a higher risk of unintentional injury.8 There are disparities by income, age, ethnicity, gender, and urban or rural residency. People of color and those earning less than $25,000 per year are much more likely to walk or bicycle.9 Traffic-related crashes are the leading cause of death for children,10 and poor children die at higher rates. The pedestrian victim of a car collision is statistically more likely to be a person of color.11 Higher pedestrian fatalities have also been noted around low- income neighborhoods. Schools with a high proportion of students of color are less likely to have continuous, well-maintained pedestrian facilities. Older adults and people with disabilities are at greater risk because of physical or mental limitations on their perception and movement. Pedestrians, bicyclists, and motorcyclists (including mopeds and scooters) are much more vulnerable than car or truck occupants in a crash. Recent studies have shown that per- cyclist risk of crash is reduced as the proportion of bicycle mode share increases.12 There is a similar effect for pedestrians.13 Although less than one-quarter of all driving takes place in a rural setting,14 more than half of all fatal motor vehicle crashes occur there.15 Rates of pedestrian fatalities are higher in urban areas.16
Mechanism
Collisions or crashes involving road users often result in physical traumas, which can lead to disability or death. A crash may involve a single bicycle or motor vehicle, multiple vehicles, or any number of vehicles and pedestrians. Conventional wisdom has held that roads can be made safer for motor vehicles by moving fixed objects back from the roadside; widening travel lanes; and employing channelization, acceleration lanes, and grade separation at intersections. However, researchers are finding that this type of design may not provide the anticipated safety benefits. Health professionals now believe that such designs promote speeding and reduce driver awareness, leading to much higher rates of pedestrian and bicycle fatalities.17
Road design can increase crash risk by determining where and how traffic movements will occur. This can exacerbate conflicts between two or more road users; changes in speed or direction; safety of at-grade rail crossings; and road user speeds, visibility, and attentiveness. Designing a road to control traffic flow as well as to accommodate all of the movements that any user might want to make, safely and without excessive delay, is the key. In urban areas, access management plays a large role. In a rural setting, the challenge can be accommodating slow or non-motorized traffic without promoting higher speeds. It even appears that rural roads with many curves have fewer crashes than flat, straight roads, perhaps due to increased vehicle speeds on the latter. Areas on the metropolitan fringe may be particularly vulnerable as they begin to carry more traffic on roads intended for rural use. While each road is different, users of all types must be anticipated, and design should be context sensitive. The principles of injury mitigation are outlined below.
Roadways and Health
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 8 5
<
<
R o
a d
w a
y s
a n
d H
e a
lt h
ch. 5
Mitigation: reducing Injury
• Base road design decisions on state-of-the- art transportation and health research and ensure that such research is disseminated to both planning and engineering staff.
• Constrain vehicle speeds as appropriate to the road context.18
• Incorporate treatments to control conflict points, such as medians, alleys, traffic signals, and movement restrictions.19
• Design roads to reduce risky driving behavior, rather than to accommodate it.
• Increase the share of bicycle facilities to reduce per-cyclist risk.
• Increase the share and quality of pedestrian facilities to protect pedestrians from traffic, reduce individual risk, and minimize fear of crime.20
• Include public transportation facilities and shift travel to this mode, reducing risk of injury.
• Provide sidewalks and frequent crosswalks to improve pedestrian safety.21
• Reduce corner radii where possible to minimize pedestrian exposure and reduce vehicle speed.22
• Provide more transportation choices to reduce vehicle volume.
• Utilize a network of streets to disperse traffic volume and provide smaller, safer roads for pedestrians and bicyclists.23
• Create landscaped, tree-lined roads.24
• Reduce roadside distractions such as billboards.
• Improve street and roadside lighting, especially at conflict points.25
• Review universal design standards and seek to implement road design that accommodates all users safely, regardless of their limitations.
• Institute and enforce maintenance schedules for all facilities.
Env ironmenta l Qua lity
Impact
Motor vehicle traffic presents a unique public health risk because of the toxicity of its emissions and its extensive integration within communities. Recent research links diesel exhaust to lung cancer, cardiopulmonary disease, and other causes of death. More than 42 percent of Americans live in places that exceed national air quality standards for ozone or fine particulate matter. Asthma affects nine percent of U.S. children and seven percent of adults.26 Climate change may already be responsible for more than 150,000 deaths per
Context sensitive roads designed for all users can enhance safety.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 8 6
>
>
C h
a p
te r
5
year and is expected to have a devastating effect on global climate patterns. Vehicle- related fine particulate matter becomes highly concentrated in areas immediately adjacent (200 meters) to major roadways. Outdoor particulate matter concentrations (PM2.5 and PM10) are an estimated 15 to 20 percent higher at homes located on high-traffic intensity streets compared to homes located on low-traffic intensity streets and at intersections.27
Children, older adults, pregnant women, and low-income households are especially vulnerable.28 Vehicle-related pollutants have been associated with increased respiratory illness, impaired lung development and function, and increased infant mortality. Also, pregnant women living within 200 to 300 meters of high-volume roads face a 10 to 20 percent higher risk of early birth and of low-birthweight babies. Children living near busy roads are six to eight times more likely to have certain forms of cancer. Additionally, fine particulate matter (PM2.5) has an adverse effect on lung development in adolescents that can lead to lifelong lung deficiency,29 and even small amounts of air pollutants are associated with small changes in cardiac function in older adults.30 In addition, low-income and minority communities are more at risk for higher levels of pollutant exposure, as their homes are more likely to be located near busy roadways.31
Mechanism
Road-based airborne emissions result from tailpipe exhaust, fuel delivery, road surface wear, deterioration of vehicle parts, and electricity production for electric-powered vehicles. Particulate matter (PM), carbon monoxide, nitrogen oxides (NOx), and volatile organic compounds (VOCs) are all major concerns, as well as ozone, which form from NOx and VOCs, and black carbon and sulfur dioxide, which are emitted by diesel-burning vehicles. Exposure to these pollutants significantly increases the incidence of asthma, respiratory diseases, lung cancer, and cardiovascular disease. Additionally,
carbon dioxide and other greenhouse gas (GHG) emissions cause climate instability and stimulate natural disasters, food scarcity, and unhealthful weather and ecological patterns such as heat waves and the spread of disease-carrying insects.
The actual level of pollution from all cars and trucks is a function of vehicle miles traveled, the number of trips, the condition of the vehicle, the weather, and the driving conditions. In particular, traffic congestion can increase emissions because it leads to extra accelerating, braking, and idling. The highest level of tailpipe emissions is generated when the vehicle is started, making even short motor vehicle trips a culprit in air pollution. Additionally, large expanses of pavement for highways and parking can exacerbate emissions by increasing air temperature, which facilitates ozone formation; trees, shrubs, and some plantings can reduce pollution by keeping the area cooler and by absorbing some carbon dioxide and VOCs from the air. Both passenger and freight movement are relevant to emissions levels, as freight transport accounts for a large percentage of air pollution.
Motorists experience high exposure to vehicle emissions while driving, especially in stopped
Roadways and Health
This congested roadway is exposing individuals on or near it to air pollutants, including children on a school bus. Alternative modes are often lacking, even for short trips.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 8 7
<
<
R o
a d
w a
y s
a n
d H
e a
lt h
ch. 5
traffic. People living in immediate proximities (200 meters) of major diesel thoroughfares are more likely to suffer from respiratory ailments, childhood cancer, brain cancer, leukemia, and higher mortality rates than those who live farther away. Adults with asthma who walk along these thoroughfares are more likely to suffer acute symptoms.32 Airborne outdoor pollutants can penetrate any building through small gaps, ventilation systems, and open doors or windows.
Mitigation: Improving Air Quality and the Environment
• Increase the level of service for non- motorized travel to reduce automobile trips.
• Use roadway design and transportation alternatives to reduce congestion and make motor vehicle travel more efficient.
• Avoid road projects that compete directly with existing or planned lower-emission freight and passenger rail transport.
• Seek alternatives to road projects that will increase motor vehicle traffic near populated areas.
• Manage access to control congestion and freight traffic.
• Permit trees and plants along roadways to provide cooling, shelter for pedestrians, and capture some emissions.
• Promote higher-density land use to reduce the distances traveled by motor vehicle.
• Promote a connected network of streets to allow bicyclists and pedestrians to avoid using major thoroughfares.
Mode Sha re
Impact
Physical inactivity and elevated body mass index (BMI) are among the most pressing health concerns today. Thirty-four percent of Americans are obese, and more than two-thirds are overweight or obese. Obesity, defined as a BMI over 30, leads to elevated risk for heart disease, type 2 diabetes, cancer (including breast cancer and colon cancer), high blood pressure, stroke, liver disease, sleep disorders, arthritis, and infertility. Obese individuals are twice as likely to die prematurely as their non-obese counterparts. Sixteen percent of American children are obese, many of them already at risk for heart disease and type 2 diabetes.33 Physical inactivity is a primary factor in obesity, and it is thought to contribute to approximately 30 percent of all U.S. deaths. Physical inactivity is estimated to have cost the United States more than $250 billion in 2006.34
Social capital—the collective benefits conferred by social networks—decreases 10 percent for each additional 10 minutes spent commuting35 and is lower for people who live on streets with high traffic volume.36 Mental health is assailed as traffic congestion, traffic danger, and commuting add to daily stress and prevent people from spending enough time with their families or engaging in more productive and enjoyable activities.37 Transportation expenditures are the second-largest expense for an American household, and some households spend more than 22 percent of their income on transportation. In 1998, this expense approached $9,000 per household.38
Low-income households are more affected by transportation expenses than others and can spend up to 40 percent of their income on transportation. These underserved populations tend to be minority or of lower economic status.39 Affected by high unemployment rates
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 8 8
>
>
C h
a p
te r
5
and lack of services, these populations rely on walking, bicycling, and public transportation to achieve economic stability. In many low-income communities, transportation to a hospital or medical office is completely lacking, except by ambulance. Additionally, almost one-third of Americans do not drive.40 This group includes children under age 16, older adults who can no longer drive safely, people who cannot afford to own and operate a car, and people with disabilities, among others. These individuals constitute a significant part of the economy, as both workers and consumers. Without transportation, they experience difficulty accessing jobs, healthcare, churches, stores, government services, and friends or family.
Mechanism
Over-reliance on private motor vehicle travel eliminates a major source of regular physical activity. Average BMI has increased as walking and bicycling trips have declined, but a greater share of pedestrian or bicycle travel leads to gains in physical activity. In many localities, it is unsafe, unpleasant, or simply impossible to walk, even across the street or to an adjacent property. Excessive travel times decrease social capital, which can lead to mental health issues, substance abuse, and degraded relationships between family members or neighbors. Increased pedestrian travel contributes to overall lower household transportation costs and gains in social capital. Additionally, a greater share of transportation facilities increases transportation ridership, which increases pedestrian travel and enhances physical activity levels. The more time an individual spends driving a car, the more likely that driver is to have an elevated BMI.41 Automobile transportation is vastly more expensive than walking or bicycling and generally much more expensive than mass transit. Therefore, families in automobile- dependent regions may have to spend more money on transportation.
Wide, continuous sidewalks increase the comfort and efficiency of walking, especially for groups or people employing wheelchairs or strollers, and lead to more people walking.42 Planting zones or furniture zones improve the comfort and efficiency of walking by buffering pedestrians from traffic, leaving room for pedestrians to pass behind turning vehicles, and removing obstacles from the main walkway. Good aesthetics, amenities, and sidewalk-oriented building frontage and design create a lively social environment and increase personal safety. Sidewalk-oriented building frontage and design improves access to homes, stores, and services for persons on foot. Street lighting increases walking43 and improves actual and perceived personal safety. Shorter distance to destinations has a strong correlation with increased walking and bicycling,44 and higher connectivity has a strong correlation with increased walking and bicycling. Trees provide shade, without which walking or bicycling may be unbearable on warmer days. Greater intensity of usage can also increase actual and perceived personal safety for non- motorized transport, while actual or perceived
Roadways and Health
Roads can accommodate the needs of all road users, regardless of travel mode and ability.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 8 9
<
<
R o
a d
w a
y s
a n
d H
e a
lt h
ch. 5
danger from high-speed or high-volume traffic discourages walking and bicycling. Shorter distance to destinations improves access for low- income families and people with disabilities, while higher-density retail and commercial development is linked to more pedestrian travel. ADA-compliant facilities allow persons with disabilities to travel along the sidewalks.
Mitigation: Diversifying Mode Share and reducing Automobile/ roadway Use
• Control speed and conflict points to improve the pedestrian and bicycle environment.
• Design intersections to serve all types of users with an equal degree of priority and minimum delay.
• Develop more accurate ways to evaluate level of service for all travel modes and road users, and use them to increase and improve bicycle, pedestrian, and transit travel as appropriate to location (including lower- volume rural roads).
• Enhance access to transportation services and eliminate roadway barriers such as infrequent pedestrian crossings or turn lanes that affect bus access to a bus stop.
• Promote higher-density land use to increase the number of destinations in walking or bicycling distance.
• Ensure that the entire roadway, including sidewalks and bicycle lanes, is adequately cleaned and maintained.
• Enhance street networks to minimize wide or high-volume roadways.
• Keep block lengths short and well-connected.
• Create pedestrian-friendly environments: wide sidewalks, planting or furniture zones between the vehicle lanes and the sidewalk, benches, waste and recycling receptacles, shade trees, sidewalk-oriented building frontage and design, street and sidewalk lighting, and pleasant streetscape.
federa l leg islation: Equity, Hea lth , a nd Hig hways
It is appropriate to argue for a redefinition of highways. Historically, the highway system has been designed to move large numbers of passenger and freight vehicles at fast speeds. It connects homes and jobs for motorists but is not sensitive to other needs of highway users. Highways define the travel experience of people with diverse backgrounds, socioeconomic status, and lifestyle preferences. They disrupt communities and begin to structure the social interaction of residents. Highways must become entities that integrate physical activity, minimize negative health impacts, enhance social interaction, preserve environmental quality, promote community health, increase safety, and promote sustainability even as they
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 9 0
>
>
C h
a p
te r
5
become more responsive to global demands providing equitable access and participation in daily life.
The current federal transportation bill— SAFETEA-LU—has an enormous influence on roads throughout this country. Approximately 40 percent of the transportation dollars spent nationally emanate from the U.S. Department of Transportation (DOT) and the Federal Highway Administration (FHWA).45 It comes with extensive stipulations, but very little evaluation or enforcement. It both sustained and introduced a number of notable programs, including the Highway Safety Improvement Program, various highway safety grants, Congestion Mitigation and Air Quality (CMAQ) funds, Safe Routes to School, and Transportation Enhancement funds. It promoted the Environmental Review Process, routine consideration of non-motorized travel needs, funding for routine maintenance, endorsement of standards for roadway design, and endorsement of the Americans with Disabilities Act Accessibility Guidelines; it added flexibility to National Highway System and Surface Transportation Program funds. SAFETEA-LU reinforced coordination, public participation, and planning requirements for states and metropolitan planning organizations (MPOs). These have been notable because they introduce the possibility of integrating comprehensive health considerations into transportation planning.
Roadway funding in the next federal authorization will need to place transportation in a larger context, rather than focusing narrowly on the movement of people and goods (or even more narrowly on the movement of cars and trucks). The legislation must explicitly address ways to mitigate climate change. It must continue to address casualties on our highways through requirements to restrict alcohol- impaired driving and seat belt legislation. And it must expand this effort through evidence- based road design, increased funding flexibility, and increased monies for research. As stated
in the final report of the National Surface Transportation Policy and Revenue Study Commission, highway policies should not conflict with other national policy goals.46
SAFETEA-LU implemented many initiatives aimed at making roads safer, less harmful to the environment, more equitable, and more efficient, yet such initiatives have only tinkered with the edges of highway policy and had little impact on the overall results. The current challenge is to strengthen these goals, integrate them into every decision, and provide a much wider set of mitigation options—all in a situation of shrinking fuel tax revenues and widespread economic decline.
Tra nspor tation Policy Ba rriers
Although SAFETEA-LU included a number of well-intentioned programs and policies addressing safety, environmental quality, and effects on vulnerable populations, it also contained fundamental operational practices that prevented these initiatives from being truly effective. An important first step in the new authorization will be to eliminate these barriers.
For example, transportation funding intake and allocation has been too heavily based on motor vehicle travel, motorized-vehicle lane miles, and trucking. Approximately 50 percent of the monies received by the states are based on VMT (vehicle miles traveled), arterial lane miles, diesel fuel usage, and the ratio of lane miles to population.47 It may not be desirable to link funding to increased VMT. Compare two states or localities that have created different road systems. One has roadways that primarily serve motor vehicle traffic; the other has constructed a complete, quality travel environment for pedestrians, bicyclists, cars, trucks, and buses. In this example the second location may be able to move as many people and goods at a comparable or better level of service and may do so with greatly reduced
Roadways and Health
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 9 1
<
<
R o
a d
w a
y s
a n
d H
e a
lt h
ch. 5
externalities (emissions, crashes, and inequities for nondrivers). While they may have similar amounts of total infrastructure to maintain, the second location may have lower lane miles and lower VMT, thus receiving less funding. In this example, the community with a roadway system more supportive of positive health outcomes would be penalized. Congestion Mitigation and Air Quality (CMAQ) funding share, which is based on air quality non-attainment, and Minimum Guarantee share, which is based on the states’ tax contribution (which is a function of the amount of fuel consumed) do little to rectify the situation.
Currently, very limited resources are allocated to non-motorized transportation, while enormous sums are committed to motor vehicle movement. A particularly large share goes to limited-access highways such as the Interstate Highway System (IHS). While the IHS fills a necessary transportation role, it is not sufficient to meet current or future travel and mobility needs. SAFETEA-LU and its predecessors have not allowed the flexibility in funding, nor the guidance, to allow more context sensitive, equitable funding of transportation projects. Local fund match requirements have not been equitable across travel modes, and previous transportation bills have not provided good mechanisms for assessing the effects of proposed highways on the roadside environment, on overall connectivity, or on the level of service for bicycles, pedestrians, or public transportation.
Overall, the use of federal transportation allocations has not been closely monitored. Although Environmental Impact Statements (EIS) are required, they have not adequately assessed health impacts (they are not sufficiently explicit on health). The needs of low-income communities and nondrivers have been routinely overlooked without consequence. In general, the entire bill has failed to sufficiently evaluate the outcome of the projects it has funded, especially with regard to vulnerable populations.
Tra nspor tation Policy Oppor tunities
A handful of policies are in use today to create healthy roads that function well for all users. These policies can be found at the federal, state, and local levels. The most relevant policies are Health Impact Assessment (HIA), Context Sensitive Design, Complete Streets, Local Area Traffic Management (LATM)/Traffic Calming, Environmental Review Toolkit, Livable Centers Initiative (LCI), Road Diets, and Green Streets (see appendix A for more detail about these policies). These policy examples go far beyond vehicle level of service to consider a project for its comprehensive effect on the immediate area and the region, often creating extra opportunities to consider equity and health concerns and to implement more meaningful public participation.
A $3.2 billion deficit is forecast for the highway trust fund in 2009, presenting both a challenge and an opportunity to revisit our transportation strategy. It is also likely that fuel purchases will decline or grow less quickly. The National Surface Transportation Policy and Revenue Study Commission final report, Transportation for Tomorrow, suggests increasing the highway trust fund revenue tax from 25 to 40 percent a gallon over the next five to eight years and indexing it to inflation. However, the report also champions environmental stewardship and the development of alternative and renewable fuels.48
Many other strategies are being put forth to help finance the priorities to be set in the upcoming authorization. Prioritizing long- term investment, developing more accurate and comprehensive cost-benefit analyses, and reducing earmarks can all help to control transportation financing. Another option is increasing collaboration with local and national advocates, planning organizations, and others to take advantage of innovations and research and facilitate private-sector funding of some initiatives. Finally, the cost-reduction benefits
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 9 2
>
>
C h
a p
te r
5
associated with other modes, travelways, and strategies are potentially substantial. There are already some innovative proposals, including:
• The Lieberman-Warner Climate Security Act, which includes some transportation funding that might be appropriate for Health Impact Assessment.
• Senator Benjamin Cardin of Maryland and others have recommended a Transportation Sector Emissions Reduction (TSER) Fund that would permit the auctioning of emission allowances. Approximately five percent of TSER funds would be available to state and local authorities for transportation alternatives that reduce travel demand, including regional planning organizations.
• Senator Tom Carper of Delaware has proposed CLEAN TEA (Clean Low-Emissions Affordable New Transportation Equity Act). This act reduces greenhouse gas emissions by promoting alternatives to driving. CLEAN TEA provides low-emissions transportation options by directing cities with more than 200,000 residents and state departments of transportation to review their transportation plans and determine how they could reduce greenhouse gas emissions. Federal funding for projects in those transportation plans would be distributed to states and localities based on the expected reductions in greenhouse gas emissions in each plan. States and cities with more ambitious plans would receive greater funding.
Convergence Oppor tunities
The upcoming transportation authorization presents many opportunities to create partnerships and take advantage of mutual interests to create healthier road networks. A number of innovative policies have been identified above. A small cross-section of entities and programs representing convergence opportunities follow:
• Medicare and Medicaid programs spend almost 10 percent of their budget each year treating conditions related to obesity and physical inactivity.
• State and local police departments incur significant costs responding to crashes. Many have already funded their own road safety programs.
• State and local tax dollars are being used to bus students, even though many live within walking distance. Some are participating in the federal Safe Routes to School program to reconstruct the road infrastructure near school property and develop programs to encourage physical activity.
• High-cost roadway capacity projects are becoming less feasible for transportation department budgets and less popular among taxpayers and residents.
• Industry, freight, and automakers will bear the brunt of climate change legislation without more opportunities for change in personal travel behavior.
• Emergency services for crash victims are overwhelmed and strapped for cash.
• Health insurance providers spend billions each year treating conditions related to physical inactivity, air pollution, and roadway casualties.
• Labor departments are aware that transport and child care are the biggest barriers to employment and are seeking solutions.
• Federal and state agriculture and environmental protection divisions are devoting resources toward environmental quality.
• The federal Centers for Disease Control and Prevention (CDC) and countless public and nonprofit organizations are investing
Roadways and Health
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 9 3
<
<
R o
a d
w a
y s
a n
d H
e a
lt h
in physical activity programs, road emission mitigation programs, and more.
• Public health research is building evidence for design of safe and healthy road environments, but the work may not be translated to engineering and planning practices.
Many of these opportunities involve various branches of the federal government, if only as a funding source. They allow addressing multiple issues at once by including health, equity, and road programs in the same planning process. This would prevent duplication of activities, take advantage of existing expertise, and avoid having federal programs work at cross-purposes to one another.
Conclusion
Roadway systems are set in a context of towns and cities, commerce and agriculture, ecological systems, neighborhoods, regions, state and local governments. Our public spaces and our travel along them have a profound effect on all of these settings. They are extensive and
thoroughly integrated into all aspects of the American landscape. As a result, they play a large role in the health and quality of life of the general population.
While the purpose of the upcoming authorization is to address highway funding and the movement of people and goods, within the entire national context, it plays a much larger role in the health outcomes of citizens. The biggest impacts result from crash- related injuries, vehicle emissions that pollute the air and contribute to climate change, automobile dependency leading to sedentary behavior, and the lack of equitable access for all Americans. The implementation of the mitigation strategies, policies, programs, and design guidelines outlined earlier result in significant improvement in the positive effect of roadway systems on health. The recommended steps to improve safety, reduce emissions, and create high levels of service for all travel modes change the role of the roadway system, causing it to be more supportive of good health and increased prosperity. In this way, it expands its contribution to improving the health status of Americans.
ch. 5
Appendix A. Policies and Strategies for Healthy Transportation
Health Impact Assessment (HIA)
Principles Addressed: • Injury Scope: • Local • Environmental Quality • State • Mode Share • Regional • Federal Description: A combination of procedures, methods, and tools by which a policy,
program, or project may be judged as to its potential effects on the health of a population and the distribution of those effects within the population. Public participation is an important part of health impact assessment.
References: http://www.cdc.gov/healthyplaces/hia.htm http://www.hc-sc.gc.ca/ewh-semt/pubs/eval/handbook-guide/vol_4/
table-tableau-3-eng.php#Table-3-1a
Context Sensitive Design
Principles Addressed: • Injury Scope: • Local • Mode Share • State • Regional • Federal Description: A collaborative, interdisciplinary approach that involves all stakeholders
to develop a transportation facility that fits its physical setting and preserves scenic, aesthetic, historic, and environmental resources while maintaining safety and mobility. An approach that considers the total context within which a transportation improvement project will exist.
References: http://www.cnu.org/streets http://www.fhwa.dot.gov/context/index.cfm http://www.contextsensitivesolutions.org/content/topics/css_design/
design-examples/
Complete Streets
Principles Addressed: • Injury Scope: • Local • Environmental Quality • State • Mode Share • Regional Description: Complete Streets are designed and operated to enable safe access for
all users. Pedestrians, bicyclists, motorists, and bus riders of all ages and abilities are able to safely move along and across a complete street.
References: http://www.completestreets.org/ http://www.completestreets.org/federal.html (S. 584/H.R. 1433)
Local Area Traffic Management (LATM)/Traffic Calming
Principles Addressed: • Injury Scope: • Local • Mode Share • Federal (non-U.S.) Description: Traffic calming is a system of design and management strategies that
aim to balance traffic on streets with other uses. The tools of traffic calming provide an example of a different approach from treating the street only as a conduit for vehicles passing through at the greatest possible speed.
References: http://www.cochrane.org/reviews/en/ab003110.html http://www.fhwa.dot.gov/environment/tcalm/part3.htm http://www.pps.org/info/placemakingtools/casesforplaces/
livememtraffic
Environmental review Toolkit
Principles Addressed: • Environmental Quality Scope: • Federal • Mode Share Description: Environmental stewardship and streamlining resources for FHWA offices,
state departments of transportation, resource agencies, and consultants. The website includes a guide to practices by state, links between planning and the environment, and the National Environmental Policy Act (NEPA).
References: http://www.environment.fhwa.dot.gov/
Roadways and Health H
e a
lt h
y, E
q u
it a
b le
T ra
n s
p o
rt a
ti o
n P
o li
c y
p g
. 9 4
>
>
C h
a p
te r
5
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 9 5
<
<
R o
a d
w a
y s
a n
d H
e a
lt h
Livable Centers Initiative (LCI)
Principles Addressed: • Environmental Quality Scope: • Local • Mode Share • Federal Description: The LCI is a program offered by the Atlanta Regional Commission. It
is an example of promoting local strategies to plan and implement a link between transportation improvements and land use development policies to create sustainable, livable communities consistent with regional development plans.
References: http://www.atlantaregional.com/html/308.aspx
road Diets
Principles Addressed: • Injury Scope: • Local • Mode Share • State • Regional Description: "Road diets" are typically conversions of four-lane undivided roads into
two through lanes and a center turn lane or two through lanes and a median. The fourth lane may then be converted into bicycle lanes, sidewalks, or on-street parking. "Road diets" are an example of service reevaluation for all users.
References: http://www.walkable.org/assets/downloads/roaddiets.pdf http://www.tfhrc.gov/safety/hsis/pubs/04082/index.htm http://www.contextsensitivesolutions.org/content/reading/road-diets-2/
Green Streets
Principles Addressed: • Injury Scope: • Local • Environmental Quality • State • Mode Share Description: Sustainable practices associated with the design and construction of
roadways, such as use of recycled or sustainable construction materials, ecologically-sensitive storm water management, and extensive use of vegetation.
References: http://www.lowimpactdevelopment.org/greenstreets/
ch. 5
p
g . 9 6
>
>
ch. 2
Transportation is the lifeline of communities. It connects residents to jobs, stores, family, friends, doctors, schools, parks, clubs, religious institutions, volunteer commitments—everything that allows people to participate and prosper in society. Transportation policy bears on every critical issue facing neighborhoods, regions, and the country.
The chapters in this section cover:
>> Economic development
>> access to healthy foods and healthy food systems
>> Traffic safety
These are by no means the only issues that should be considered in crafting the new transportation bill. Healthy, equitable, forward- thinking transportation policy must address a number of urgent and interconnected issues, among them climate change, environmental justice, freight transport, and workforce development.
KEY ISSUES
chapter name
p g
. 9 8
>
>
Breaking Down Silos: ch. 6 Transportation, Economic Development, and Health TODD SWANSTROM, Ph.D. E. Desmond Lee Professor of Community Collaboration and Public Policy Administration University of Missouri, St. Louis
ABSTRACT >> Transportation policy in the United States has historically emphasized automobile use and steered land use, development, and investments in infrastructure toward low-density suburbs. This approach has left low-income communities in aging city centers poorer, sicker, and increasingly immobile, unable—more and more—to get to work, their doctor, parks, gyms, or even grocery stores that sell fresh, healthy food. This paper explores an alternative transportation policy designed to create healthy, productive metro regions by closing the gap between affluent, mobile communities and their less mobile, disadvantaged neighbors.
By reconfiguring how we use available land, we can create densely populated, mixed-use communities that expand access to transportation and improve health outcomes. With a focus on equity, these policies can also support economic development that reduces poverty and economic and racial segregation.
This paper considers two approaches: creating mixed-income, transit oriented villages and using transportation funds to promote local workforce development. While the goals of equity and environmental sustainability are not mutually exclusive, the paper concludes by cautioning activists against ignoring the short-term needs of low-income families who live in built environments dominated by the automobile.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 0 0
>
>
C h
a p
te r
6
Breaking Down Silos
CONTENTS
Introduction .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . 101
Unhealthy Effects of the Highway Policy Silo.. . 102
New Transportation Policies for Healthier Economic Development . .. .. .. .. .. .. .. .. .. .104
Mixed-Income Transit Oriented Development 104
Policy Recommendations .. .. .. .. .. .. .. .. .. . 107
Transportation . .. .. .. .. .. .. .. .. .. .. .. .. .. . 107
Housing.. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . 107
Transportation and Local Workforce Development .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .108
Policy Recommendations .. .. .. .. .. .. .. .. .. . 110
Transportation . .. .. .. .. .. .. .. .. .. .. .. .. .. . 110
Labor .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . 110
Conclusion.. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. ..111
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 0 1
<
<
B re
a k
in g
D o
w n
S il
o s
ch. 6
introduction
The United States is in the midst of a shift in transportation policy—from mobility to individual and community accessibility. Traditionally, transportation choices in this country have been made inside policy “silos” that isolate decisions on how we commute and travel from decisions on how we live. By making these decisions in a vacuum, transportation policies have promoted sprawl, or low-density patterns of housing that favor automobile use over public transportation and that exact a huge toll on the health of our metro regions, particularly low-income communities.
The goal of making transportation more efficient is not to move people faster and farther but to give them wider access to all the things that are necessary for a good life: jobs, education, family, friends, recreation, culture, etc. Under this approach, for example, it might make sense to spend transportation funds on housing construction near major employment centers. This kind of planning can be especially beneficial for low-income families who don’t own a car. But for it to happen requires a more democratic decision-making process in which all community stakeholders have input. This broad- based effort can produce more environmentally sustainable regions.
The focus of this paper is on vertical equity, or policies that provide the most benefits to the most people, including those at the bottom of the socioeconomic ladder. Equity should not be understood simply in terms of income or wealth, but in terms of what Amartya Sen calls “functionings and capabilities.” According to Sen, “relevant functionings can vary from such elementary things as being adequately nourished, being in good health, avoiding escapable morbidity and premature mortality, etc., to more complex achievements such as being happy, having self-respect, taking part in the life of the community, and so forth.”1 Capabilities refer to the ability to have choices. Other things being equal, people are better off
if they have choices in how they want to live their lives.2 To achieve transportation equity, not all low-income people should be treated alike because, depending on where they live, some people have greater transportation needs than others.3 For example, using transportation funds to develop pedestrian-friendly, transit-rich villages will enable people to have acceptable “capabilities and functionings” without building expensive highways.
This essay will not examine the direct effects of transportation services on health. Providing more bus routes for low-income communities, for example, would help people to access medical care or healthy foods. Instead, the focus here is on how transportation influences economic development that in turn affects health. By facilitating market exchanges, transportation influences what kind of economic development occurs (single use or mixed use), where it occurs (on the suburban fringe or near the center), and who benefits (rich or poor, white or black). The type of economic development that occurs has direct effects on health. Compact, mixed-use developments that rely more on public transportation, walking, and biking support better health outcomes, other things being equal, than auto-dependent, low- density economic development that separates residential, retail, and office functions.4
Besides these direct effects, there are also many indirect effects of transportation systems on health. Transportation policies encourage economic development that either worsens or lessens poverty, inequality, and economic and racial segregation. All of these factors—poverty, inequity, and segregation—are associated with poor health outcomes (see endnotes five and six). The link between poverty and poor health outcomes is well documented, but less well known is that income inequalities across class and space are also associated with poor health.5 Moreover, residents of areas with concentrated poverty not only have little access to health services, but also experience other factors that undermine health,6 including:
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 0 2
>
>
C h
a p
te r
6
1. Less Exercise: Because people are afraid to go outside in high-crime areas and because high-poverty areas often lack good walking infrastructure, such as parks and sidewalks, living in poverty-impacted neighborhoods discourages physical activity and therefore increases obesity and other negative health outcomes.
2. Poor Air Quality: High-poverty neighborhoods are more likely to be the locations for toxic waste dumps, garbage transfer stations, bus depots, highways and ports, and truck facilities, and therefore suffer from inferior air quality due to toxic fumes as well as gasoline and diesel exhaust.
3. Inadequate Diet: Residents of high-poverty neighborhoods often lack access to low-cost, high-volume grocery stores with fresh fruits and vegetables.
4. High Stress: Finally, residents of poor neighborhoods suffer from the withering effects of stress. High crime, overcrowding, noise, unemployment, lack of retail outlets, and poor public services are all stressful. Chronic stress damages our organs and immune systems and is associated with cardiovascular disease, asthma attacks, and premature death.
The paper concludes with recommendations for transportation policies that can reduce economic inequalities and improve the access of disadvantaged populations to all those things that are necessary for a good life and good health. It cautions that we need both long-term policies—to reduce automobile dependency by changing land use patterns over time—and short-term policies—to meet the needs of low-income families who live in automobile- dependent environments.
Unhea lthy Ef fects of the Hig hway Policy Silo
Until the 1990s transportation policy in the United States was dominated by what political scientists call a policy monopoly, or silo—an arena of government decision making controlled by industry insiders and insulated from demands by other stakeholders.7 A steady stream of funding for transportation was guaranteed by federal- and state-earmarked gasoline taxes, and decisions about spending that money were made largely by highway engineers within state departments of transportation (DOTs).
The transportation policy silo was influenced by market principles intended to maximize mobility. Building more and more roads was the market’s response to meet demand of customers who had the most money to spend. Highway engineers in state DOTs based their decisions to extend roadways on mathematical projections for increasing automobile travel, and the central tenet was increased mobility—moving more people over greater distances at higher speeds. Highway engineers were not trained to think about how land use patterns influenced travel demand but to focus on how to move people in the most efficient manner given the infrastructure that was in place.
Rather than simply respond to demand, however, highway building created demand for more roads and cars. This is called traffic generation or induced demand: expanding road capacity on the urban fringe promoted low- density suburban sprawl that in turn generated demand for more highways.8 Reinforced by suburban zoning codes, auto-centered transportation policy promoted economic development that separated residential, retail, office, and wholesale functions into distinct geographic zones. Instead of a market equilibrium or balance between different transportation modes and land use patterns, silo-driven transportation policy generated a positive feedback mechanism that encouraged
Breaking Down Silos
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 0 3
<
<
B re
a k
in g
D o
w n
S il
o s
ch. 6
one mode (automobiles) and one land use pattern (suburban sprawl) to expand unchecked.
The white middle-class families that moved out to the suburbs to live in single-family homes on large lots generally inhabited environments with plenty of green space, sunshine, low crime, and low stress.9 Most of the negative effects of highway-oriented economic development fell on those left behind by suburban sprawl. Highway construction encouraged the movement of jobs away from the urban core.10 Largely because of suburban zoning codes, lack of access to federally guaranteed mortgages, and racism in housing markets, inner-city working class and minority households were unable to follow jobs out to the suburbs. Unusually long distances between home and jobs for low-income and minority workers are well documented by researchers and are a cause of poverty.11
Auto-driven urban sprawl has also been a mighty engine of economic segregation. Since the 1950s, new home construction on the suburban fringe has shifted from the middle to the top of the income distribution.12 The correlation between new housing and economic segregation is strong: the newer the housing in a neighborhood, the higher the average income in that neighborhood.13 By subsidizing the flight of the middle class out of central cities
and inner-ring suburbs, the auto-dominated transportation system left behind pockets of concentrated poverty, with the negative effects on health cited earlier.
Using the power of eminent domain, state DOTs displaced millions of households to build new highways.14 Highway engineers typically located highways connecting suburbs with central business districts through low-income, usually minority, neighborhoods to save money on land acquisition. Involuntary displacement from highway building severed social connections, which have been shown to be crucial for good health.15 Forced moves can be life threatening for older adults. At the same time that urban neighborhoods were disrupted by highway building, the highway construction jobs went overwhelmingly to white, often suburban, construction workers.16
The highway-dominated transportation system also puts pressure on family budgets, especially among low-income families. The general standard is that no family should spend more than 20 percent of income on transportation; after that, transportation expenditures will begin to eat into other necessities, such as housing and healthcare.17 The average American household devotes about 18 percent of its after-tax income to transportation, but this varies by income and by place of residence. Overall, transportation expenditures are regressive with regard to income.18 Low-income households, and especially those who live in areas without good public transportation, spend a much higher percentage of their incomes on transportation. For example, households earning between $20,000 and $35,000 and living far from employment centers spend 37 percent of their income on transportation.19 To have access to jobs, they must own a car. The necessity of car ownership exacerbates poverty. In 2007 the annual cost of owning an automobile averaged $9,498 (for insurance, gas, maintenance, and the average annual cost of purchasing or leasing an automobile).20
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 0 4
>
>
C h
a p
te r
6
new Tra nspor tation Policies for Hea lthier Economic Development
The 1991 Intermodal Surface Transportation Efficiency Act (ISTEA) was designed to break open the policy silo that had dominated transportation policy for so long.21 As the name suggests, ISTEA aimed to create intermodal systems that balance highways with transit, walking, and bicycling. ISTEA made it easier to “flex” funds from highways to transit. By encouraging the coordination of land use and transportation, ISTEA began the shift from a mobility policy paradigm to an accessibility policy paradigm. It changed the way decisions were made, removing some decision-making power from highway-dominated state DOTs and giving metropolitan planning organizations (MPOs) veto power over projects in their area. ISTEA began to open the transportation policy silo. For example, decisions for spending Congestion Mitigation and Air Quality (CMAQ) funds had to be approved by the air quality district, thus ensuring that environmental interests would be at the table when some transportation decisions were made. ISTEA also required MPOs to publish an overall plan for citizen participation. The intent was to have a broad array of stakeholders at the table when transportation decisions were made.
Although ISTEA and its successor acts (the Transportation Equity Act for the 21st Century, or TEA-21 (1998), and the Safe, Accountable, Flexible, Efficient Transportation Equity Act: A Legacy for Users, or SAFETEA-LU (2005) have activated new networks around transportation policy, the results on the ground have been disappointing. With the exception of California, relatively few dollars have been flexed from highways to other transportation modes.22 Even though transit ridership is up, the proportion of all trips made by public transportation declined steadily from 1990 to 2001.23 In 2007 10.3 billion trips were taken on public transportation, the highest level in 50 years; the third quarter
of 2008 reported the largest annual increase in transit ridership in 25 years.24 In 2009, just as public transportation is serving record numbers of people, many transit agencies are facing deep cuts. Efforts to coordinate transportation investments and land use continue to be halting and fragmented, and in most metropolitan areas, federal dollars are still going to highways, subsidizing energy-intensive, low-density sprawling patterns of land use that shift jobs away from needy urban communities.25 State DOTs still dominate decision making; only about six percent of federal funds are actually controlled by MPOs.26 Even within MPOs, citizen participation is often ritualistic.27 Citizen groups are put in the position of responding to decisions rather than being at the table when the agenda is set.
The upcoming authorization of federal transportation policy needs to take bold steps to correct these problems, completing the transition from a mobility policy paradigm to a focus on accessibility. All major stakeholders— drivers, transit users, local residents, environmental groups, civil rights organizations, pedestrians, and bicyclists—should have a say in how federal transportation dollars are spent in their areas. Above all, federal transportation policy needs to be more equitable. The next two sections examine areas where transportation policy can improve the health and well-being of disadvantaged groups at the same time that it builds a more efficient and environmentally sustainable transportation system. This requires transportation policymakers to step out of their policy silos and talk to those who formulate housing policy and workforce development policy.
Mixed-income Tra nsit Oriented Development
Transportation policy and housing policy tend to be developed in separate policy silos; DOTs don’t talk to HUDs. This is a mistake. Transportation investments shape housing demand and housing shapes transportation demand. Low- density suburban development would have
Breaking Down Silos
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 0 5
<
<
B re
a k
in g
D o
w n
S il
o s
ch. 6
been impossible without massive investments in suburban road capacity. Similarly, investments in new light-rail systems open up possibilities for higher-density development around transit stations. Well-planned development around such stations can produce broad benefits for society as well as targeted benefits for low- income persons, but only if equity is made a priority in the transportation-housing nexus. The result will be healthier communities, especially for low-income persons.
Starting with San Diego in the early 1980s, a new generation of fixed-rail transit systems has emerged in the United States. The new light- rail systems are faster than trolleys but stop more frequently than the heavy-rail suburban commuter trains. Bus rapid transit (BRT) lines, in which buses are given dedicated lanes and priority at traffic lights, are being developed in many cities and, if properly constructed, can provide many of the same benefits as light rail. Substantial new investments are being made in new light-rail systems. The federal New Starts program, which provides capital funds for light-rail systems, is funded at only about two billion dollars out of the approximately
$50 billion spent by the federal government on transportation each year. Only a handful of metropolitan areas get assistance in any year.
Many metropolitan areas have taken matters into their own hands, passing local taxes to pay for expansion. In 2004 Denver voters passed a half- cent sales tax to fund a $4.7 billion expansion of their light-rail system; Charlotte voters also approved a half-cent sales tax to finance a nine billion dollar light-rail system planned to be completed by 2030. Light-rail systems are sold to the voters for a wide range of benefits, including cutting traffic congestion, reducing gasoline consumption, improving air quality, and attracting new investment to the region.
All of these benefits are enhanced by transit oriented development (TOD), defined as development within a half-mile of a transit station (about a ten-minute walk) that is high density, pedestrian friendly, has mixed use, and includes station-focused public spaces. The development of new light-rail systems opens up possibilities for more efficient, more environmentally sustainable, and more equitable development. The land around light-rail stations increases in value because it is more accessible to housing, jobs, and shopping.28 Higher land values justify denser development. Drawing on these increased land values, public policies can leverage funding for affordable workforce housing with little or no cost to taxpayers. Developers can be offered density bonuses in exchange for building affordable housing. The profits they make by building more units on each plot of land will be used to fund the affordable housing, typically with money left over as additional profits. In weaker markets, mixed-income TOD may need to be subsidized by housing policies.
The demand for housing near light-rail station lines soared until the recent housing crisis, and it will rise again when the economy recovers and gas prices escalate. Today, about six million households live within a half-mile of a transit station. The demand for housing adjacent to
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 0 6
>
>
C h
a p
te r
6
transit is projected to reach 16 million by 2030.29 To meet this demand, 10 million housing units will need to be built within a 10-minute walk of transit stations. This movement toward denser, mixed-use forms of development presents a golden opportunity to create mixed- income transit villages, providing healthier environments, especially for low-income families. Enabling low-income households to live in TODs will give them access to pedestrian-/ bicycle-friendly environments that encourage an active, healthy lifestyle and that are closer to amenities, such as full-service grocery stores offering fresh fruits and vegetables.
TOD is built primarily by private developers, but it has extensive public benefits that justify government support: TOD increases property values around stations and therefore enhances tax revenues; well-designed TOD reduces crime by creating “eyes on the street” and 24-hour activity; TOD increases transit ridership and reduces traffic congestion by giving residents access to more destinations by transit and on foot; TOD reduces air pollution by cutting down on the need for automobile use; TOD saves infrastructure costs by reducing the need for parking; and TOD promotes active lifestyles that reduce obesity and improve health.
By including affordable housing, TOD can also improve equity and health. As we noted earlier, transportation costs are an onerous burden to low-income families, especially those that must own a car to get to work. TOD can reduce that burden. Higher levels of accessibility enable families to substitute more affordable and healthier forms of transportation—public transit, walking, and bicycling—for more expensive automobiles. A new tool, the Affordability Index, shows how much a household can save by living in a transit-rich environment. In Minneapolis-St. Paul, monthly costs of transportation varied from $446 to $941. Moving from a transit-poor to a transit- rich neighborhood would save the average household $5,940 a year.30 For a low-income
family, this savings would be huge. Locating jobs within TODs can help overcome the job-housing mismatch discussed earlier.
Planners may be tempted to include only higher- income housing in TODs on the ground that it will maximize property values. But this is not necessarily true. Smaller, more affordable rental housing and condos can be quite profitable. Moreover, low-income households are good to have in TODs because they tend to use transportation more than high-income households. In 2001 those earning less than $20,000 a year accounted for 38 percent of all transit riders, far more than their 14 percent share of the urban population.31 Low- income households are less likely to own a car; therefore, the zoning code can reduce the parking requirement by up to 75 percent (from one parking space per middle-income unit to one-quarter of a space per low-income unit).32 At $10,000–$30,000 per parking space, this can be a powerful incentive for developers to include affordable housing.
One of the barriers to realizing the savings of living in transit-rich environments is that it is very rarely possible for households to entirely give up access to a car. Automobile use has high fixed costs, and those costs are more burdensome to low-income households that drive fewer annual miles. Low-income drivers often pay high insurance rates, even though they drive less.33 Even if low-income households can use public transportation to get to work, in most American metropolitan areas, they will still need a car to transport major purchases or to visit friends or relatives in other parts of the region.
The root of the problem is that there is no easy way to own “part” of a car. The invention of car- sharing solves this problem by enabling access to an automobile on a pay-as-you-drive basis. A nonprofit in the Bay Area, City CarShare, opened for business in 2001, and subsequently private companies—such as ZipCar—have entered the business. Flex cars are parked on
Breaking Down Silos
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 0 7
<
<
B re
a k
in g
D o
w n
S il
o s
ch. 6
city streets and, after undergoing a background check, people can join the system and use the cars on a per-hour basis, usually for less than $10 an hour. A study of CarShare members found that nearly 30 percent of them had gotten rid of one or more cars and nearly two-thirds said they had decided not to purchase another car.34 This system could be adapted for low-income persons; used cars could be employed instead of new cars. Imagine what it would mean to a family of three earning the federal poverty cutoff ($17,600 in 2008) if they could dispense with the cost of owning a car (average cost $9,498) and instead use public transportation and car-sharing at one-half that amount or less.
To realize the full benefits of mixed-income TOD, new policies are needed to break down the silos that have encased transportation and housing policies and prevented the synergies that would result from coordinating them.35 The upcoming authorization of federal transportation policy presents an opportunity to connect transportation to economic development and health. When energy prices rise, as they will when the economy recovers, the motivation to coordinate housing and transportation policies to reduce energy consumption will also rise. The Obama administration and the new congressional leadership have expressed a desire to overcome policy silos and to begin planning transportation and housing policies together.36
Policy r ecommendations
Transportation
• Authorization of the upcoming federal transportation bill should enable MPOs to flex funds from transportation funding to subsidizing mixed-income TODs.37
• Funding for the New Starts program should be increased and the Federal Transportation Administration (FTA) should give priority to applications that incorporate plans for mixed- income TODs.
• Funds should be set aside in the next bill to provide technical assistance to local governments and community-based organizations (CBOs) to plan mixed-income TODs.
• U.S. DOT should develop a model overlay zoning code that encourages mixed- use, denser, more pedestrian-friendly development around transportation stations and disseminate best practices for TOD from around the country.
• DOT should require that MPOs’ Transportation Improvement Plans (TIPs) report on how transportation investments will address the need for affordable workforce housing near transit.
• DOT should develop a competitive grant program to subsidize car-sharing for low- income households living within half-a-mile of transit stations.
• DOT (or HUD) should develop an affordability index for housing that includes transportation costs to monitor the progress of metropolitan areas, especially for low-income households.
Housing
• The Low-Income Housing Tax Credit (LIHTC) and New Markets Tax Credit programs should be amended to incentivize projects that are located within half-a-mile of a transit stop; the U.S. Treasury should increase the LIHTC bonding cap for states to undertake mixed- income TOD projects.
• HUD should write regulations for the Community Development Block Grant (CDBG), and other grant programs, to give high priority to mixed-income TODs.
• The federal government should enact a homeownership tax credit targeted to low- and moderate-income homes located within half-a-mile of a transit station.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 0 8
>
>
C h
a p
te r
6
• HUD should create a program to preserve affordable housing within half-a-mile of a transit station and that is threatened by expiring use restrictions.
• State and local governments should allocate a portion of tax-increment financing (TIF) and other local incentives to mixed-income TODs; economic development incentives should be targeted on jobs that are accessible by transit (“location efficient job incentives”).38
• In strong market regions, local governments should enact TOD overlay zoning districts that reward developers with density bonuses if they include workforce housing.39
Tra nspor tation a nd loca l Workforce Development
Just as transportation policy needs to be coordinated with housing policy, it also needs to be coordinated with workforce development policy. Transportation expenditures generate hundreds of thousands of jobs each year in the construction industry. When these jobs are targeted to the neediest communities, transportation policy helps to lift up poor communities and, in the process, improve health outcomes. In effect, connecting transportation to workforce development enables the taxpayers to get “more bang for their bucks.”
The loss of well-paying manufacturing jobs has been devastating to many inner urban, heavily minority communities, creating pockets of concentrated poverty with all of the negative effects on health discussed earlier.40 One of the causes of entrenched poverty is the lack of decent-paying jobs for workers without a college education. The jobs they can get usually pay low wages, have few benefits (including no health insurance), and lack job ladders for advancement. Dead-end jobs offer little hope.
Construction is one industry where a worker
without a college education can get a job with good pay, decent benefits, and the prospects of advancing up a clear job ladder. Even though fewer than 10 percent of construction workers have college degrees, the average wage in construction in 2006 was $18.29 an hour, well above the minimum wage.41 Wages and benefits vary significantly in the industry.42 Unionized construction workers who have access to joint union-contractor apprenticeship systems can advance from apprentice to journey-level status, earning at least $30–$40 an hour. The apprenticeship system is paid for by a modest surcharge on all wages that are part of the collective bargaining agreement. Workers do not need thousands of dollars to access excellent job training services; in construction apprentice programs they can “earn while they learn” on the job.
Unfortunately, blacks and women have historically been blocked from skilled, unionized jobs in the construction trades. According to a recent study of the core counties in the 25 largest metropolitan areas, if blacks were employed in construction in 2006 at the same rate they were employed in the general workforce, an additional 137,044 blacks would be working in construction. In 2005 women represented only 2.6 percent of production workers in construction.43
Successful programs have been set up around the country involving collaboration among unions, community groups, and end users of construction to bring minorities, women, and low-income persons into skilled construction trades. With the exception of the recent downturn in the homebuilding industry, construction jobs are growing, offering the opportunity to bring new workers into skilled construction trades without displacing present workers. Based on retirements, transfers, and job growth, the federal government estimates that the industry will need to recruit 245,900 skilled construction workers each year between 2004 and 2014.44 With guaranteed funding
Breaking Down Silos
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 0 9
<
<
B re
a k
in g
D o
w n
S il
o s
ch. 6
of $244 billion over five years, SAFETEA-LU should have created more than 1.9 million person years of on-site construction jobs by its 2009 expiration.45
The 1931 Davis-Bacon Act, as amended, requires that all workers on federally funded construction projects be paid the “prevailing wage” in each region, which is usually close to the union wage in construction.46 The potential of targeting jobs from transportation projects to disadvantaged communities is illustrated by the Alameda Corridor project. In 1998, a coalition of community groups won a local hiring agreement on a $2.4 billion transportation project serving the ports of Los Angeles and Long Beach, called the Alameda Corridor.47 The project used a combination of federal and state monies. A coalition of 40 community- based organizations negotiated a community benefits agreement (CBA), requiring that at least 30 percent of all the hours on the project be performed by disadvantaged persons from the surrounding low-income zip codes. During the CBA negotiations, the federal government maintained that targeted hiring was prohibited on both statutory and constitutional grounds. The project was able to get around this prohibition by using only state funds for the targeted hiring program. CBOs were funded to run pre-apprenticeship programs to prepare applicants for the rigors of construction. Of the 880 graduates of the pre-apprenticeship programs, 373 were ex-offenders. Eventually, 710 local residents were placed in construction jobs.
The Transportation Equity Network (TEN)—a coalition of 300 grass-roots community groups working to make transportation policies more responsive to low-income persons, minorities, and disadvantaged communities—wanted to spread the Alameda model around the nation. In 2005 it was able to get a “Sense of Congress” inserted into SAFETEA-LU, which specifically upholds the Alameda Corridor project as a model and states that “federal transportation projects should facilitate and encourage” collaboration between state
departments of transportation and other interested parties “to help leverage scarce training and community resources to help ensure local participation in the building of transportation projects” (Public Law 109-59, Stat. 114. Section 1920: Transportation and Local Workforce Investment).
Using this provision, TEN and its allies have negotiated local workforce agreements in states and metropolitan areas around the nation.48 In one successful example community groups in St. Louis used a little-known provision in federal transportation law (23 USC 140) that allows state DOTs to use up to one-half of one percent of surface transportation funds for workforce development. The groups negotiated an agreement with the Missouri Department of Transportation that devoted $2.5 million from the $535 million I-64 project to local workforce development and reserved 30 percent of the work hours on the project for women, minorities, and low-income persons. A similar agreement was negotiated in 2008 for the Kansas City Paseo Bridge Project. In May 2008 Governor Tim Pawlenty of Minnesota signed a law that directs Minnesota’s DOT to spend the maximum amount feasible on job training and supports. Also in 2008 Michigan passed a law that directed $15 million of highway funds into job training over four years.
Successful state and local experiments show that transportation projects can successfully target jobs to needy communities. Federal prohibitions against race- or place-based targeting have been overcome by recruiting participants through “first-source” job training centers. Under first-source hiring provisions, apprenticeships are required to be filled by job training centers that are located within, and have close ties to, low-income and minority neighborhoods. These job training centers provide pre-apprenticeship training that prepares workers for the rigors of the construction trades. Many applicants lack the basic math skills, work habits, and knowledge of the construction industry to succeed
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 1 0
>
>
C h
a p
te r
6
in an apprentice program. Successful pre- apprenticeship programs impart these skills and weed out those who are unprepared, including those with drug or alcohol problems. The best pre-apprenticeship programs have high success rates placing their graduates in the construction trades, but they cost between $6,000 and 8,000 per participant.49
Successful experiments in local workforce development in the construction trades are encouraging, but they do not come close to meeting the need. This is where transportation policy can make a difference. Current federal transportation law permits states to use federal highway funds for local workforce development; it does not require them to do it. Local workforce development should be mandatory on all large federal transportation projects. The federal departments of transportation and labor should collaborate to develop joint programs on workforce development. Transportation expenditures will generate a steady demand for skilled construction labor, which could be met by targeted job training programs.
Policy r ecommendations
Transportation
• Section 1920 should be changed from a “Sense of Congress” to a mandate requiring that 30 percent of all hours on all large federal transportation projects (over $10 million) be performed by women, minorities, ex-offenders, and low-income persons from the local communities where the project is located.50
• One percent of all funding on large federal transportation projects, transit as well as highways, should be set aside to fund pre- apprenticeship programs and to subsidize the wages of apprentices.51
• State DOTs should be directed to facilitate negotiations among unions, contractors, community groups, local job training agencies, and other interested parties to negotiate agreements to implement mandated local hiring.
Labor
• The U.S. Department of Labor (DOL) should establish a program under the Workforce Investment Act to provide grants in metropolitan areas with demonstrated shortages of skilled construction workers for pre-apprenticeship programs run by unions, community-based organizations, high schools, or community colleges.
• DOL should fund a program to evaluate pre- apprenticeship programs around the country and spread best practices, including offering technical assistance to providers of such programs.
• DOL should gather data on the supply and demand for skilled construction labor
Breaking Down Silos
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 1 1
<
<
B re
a k
in g
D o
w n
S il
o s
ch. 6
in each metropolitan area for each major construction trade to guide local workforce development planning.
In short, health, environmental, and equity concerns can and must be addressed at the same time. Win-win policies can help to cement the so-called blue-green alliance between workers and environmentalists. For example, a recent Public Interest Research Group (PIRG) report showed that investment in public transportation produces 19 percent more jobs than equivalent investments in roads and bridges.52 We have shown that mixed-income development around transit stations can address poverty and improve health outcomes at the same time. Equity and health advocates have a natural convergence of interests here.
To realize these policy objectives, we do not need government agencies to just break out of their policy silos; we need citizens to break out of their advocacy silos. Transportation equity advocates need to understand the health implications of the policies they recommend, and health advocates need to be mindful of the impacts of their policies on equity—on the ability of people everywhere to access opportunities. Health advocates need to understand the key role played by land use reform in creating healthier environments and giving low-income persons access to jobs. There is a convergence of interests here that could build powerful coalitions for reform—only if advocates in each area set aside narrow definitions of self-interest and open themselves to new perspectives.
Conclusion
It is exciting to develop policies that can shape a new built environment that is healthier and more equitable than today’s norm. This will require working across the silos that have too often constrained effective public policies. For example, Secretary of HUD, Shaun Donovan, and Secretary of Transportation, Ray LaHood, have begun to collaborate on how to coordinate housing and transportation policies (see endnote 36). Using transportation policies to promote affordable housing and housing subsidies to support public transportation will reduce our over-reliance on automobiles and create healthier environments.
Unfortunately, most people today live in a built environment that requires extensive use of cars or buses. To devote the vast bulk of our resources to public transportation in order to shape the built environment in a more progressive direction would be shortsighted.53 We must continue to invest resources in maintaining and improving bus service for low- income persons and people with disabilities (including making buses less polluting), even though buses, unlike light-rail systems, do not create powerful incentives for higher-density TOD. Indeed, we may need to subsidize vans and even car ownership for some people who live in areas not serviced by mass transit.54
Ultimately, we need short-term policies to accommodate the transportation needs of people where they presently live at the same time that we advocate for long-term policies that will shape living patterns to reduce automobile dependence and create healthier environments for everyone.
p g
. 1 1 2
>
>
Sustainable Food Systems: ch. 7 Perspectives on Transportation Policy KAMI POTHUKUCHI, Ph.D. Associate Professor of Urban Planning, Wayne State University Detroit, MI
RICHARD WALLACE, M.S. Senior Project Manager, Center for Automotive Research Ann Arbor, MI
ABSTRACT >> Global agri-food and transportation systems have dramatically expanded food production and distribution worldwide. This integration, however, also adversely affects human health. The negative effects arise from unequal access to healthy food, unequal access to transportation for agri-food workers, increasing geospatial and economic concentration in the agri-food industry, and an emerging competition between food and fuel. Because the health of individuals is inextricably tied to the health of communities, regions, and ecological systems, health and transportation professionals need to act to both mitigate current disparities and enhance the future viability and sustainability of these systems. This paper offers numerous, specific recommendations for improving health through transportation policy and programs as they relate to agri-food systems.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 1 4
>
>
C h
a p
te r
7
Sustainable Food Systems
CONTENTS
Agri-food Systems, Health, and Transportation: An Overview.. .. .. .. .. . 115
Disparities in Urban and Rural Communities’ Access to Healthy Foods . .. .. .. .. .. .. .. .. .. . 116
Lack of Grocery Stores in and near Low-income Neighborhoods .. .. .. .. .. .. .. . 116
Increased Dependence on Use of an Automobile for Grocery Shopping .. .. .. .. . 117
Disparities in Affordable Transportation Alternatives for Agri-food System Workers . .. .. .. .. .. .. .. .. .. .. .. .. .. . 118
Transportation, Agri-food System Sustainability, and Disparate Community and Regional Impacts.. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . 120
Increased Road- and Air-miles in Food Transportation . .. .. .. .. .. .. .. .. .. .. .. . 120
Increased Consolidation of the Food Industry and Disparate Social and Spatial Impacts .. .. .. .. .. .. .. .. .. .. .. .. .. .. . 121
Food versus Fuel and Related Health Impacts . . 122
Elements of a Sustainable Agri-food System .. . 123
Transportation Goals . .. .. .. .. .. .. .. .. .. .. .. .. . 123
Transportation Policies: Opportunities and Barriers .. .. .. .. .. .. .. .. . 126
Convergence Opportunities .. .. .. .. .. .. .. .. .. . 128
Conclusion.. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . 129
LIST OF ILLUSTRATIONS
Tables
1. Energy Consumption and Emissions by Different Freight Modes .. .. .. .. .. .. .. .. . 119
2. Average Distance by Truck to Chicago Terminal Market . .. .. .. .. .. .. .. .. .. .. .. .. .. . 119
3. Estimated Fuel Consumption, CO2 Emissions, and Distance Traveled for Conventional, Iowa-based Regional, and Iowa-based Local Food Systems for Produce . .. .. .. .. .. . 120
4. Desired Policies and Programs to Address Transportation-Related Agri-food Problems: Opportunities for Success .. .. .. .. .. .. .. .. . 124
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 1 5
<
<
S u
s ta
in a
b le
F o
o d
S y
s te
m s
ch. 7
a g ri-food Systems, Hea lth , a nd Tra nspor tation: a n Over v iew
Agri-food systems include the production, processing, distribution, and consumption of food; the disposal of wastes; and the resources, actors, rules, and processes involved in the design, implementation, promotion, and regulation of these activities. These systems interact with communities to affect human health, both directly and indirectly. This paper explores these interactions to inform transportation policies that improve health, strengthen communities, and protect the environment.
As a result of linkages between the agri-foods industry and growing transportation networks, most U.S. households have ready access to large quantities of foods from all over the country and abroad; communities in crisis can quickly receive food aid transported from faraway countries; and exporters can efficiently reach grocery store shelves and markets around the world, positioning U.S. corporations at the helm of an international retail food enterprise pegged at four trillion dollars annually.1
But the integrated system for food production and distribution has left behind millions of Americans in low-income communities in the inner cities and sprawling rural areas. Women, people of color, and immigrants have been left particularly vulnerable. To reduce disparities and attendant costs; to distribute benefits more equitably; and to build more sustainable transportation, food, and community systems, transportation policy must focus on health concerns resulting from:
• Lack of access to grocery stores offering affordable, healthy foods. This imbalance is associated with higher rates of obesity, disease, food insecurity,2 and related stress;
• Lack of efficient, affordable transportation access for agri-food workers, such as farm workers and food service staff, whose wages are among the lowest in a region;
• A global agri-food industry that is fueled by cheap energy and transportation subsidies but, paradoxically, poses serious health risks to the community and exacerbates climate change; and
• Competitive market pressures to use crops for fuel, raising the price of food.
Transportation policy has not traditionally considered these issues, but it should, given the increasing rates of obesity and related health costs; climate change; threats to global food security; and inefficient, unsustainable food systems that rely on cheap energy to distribute food to faraway places.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 1 6
>
>
C h
a p
te r
7
Dispa rities in Urba n a nd rura l Communities ’ access to Hea lthy foods
Communities do not enjoy the same access to healthy foods, with inner-city neighborhoods and remote, rural areas faring the worst.3 This disparity occurs for several reasons, including a lack of grocery stores in low-income neighborhoods, a lack of affordable mass transportation, and lower rates of automobile ownership in low-income areas.
Lack of Grocery Stores In and near Low-income neighborhoods
Over the past five decades, the food retail industry has transformed itself in many ways, resulting in fewer corporate chains capturing a larger share of the retail market,4 more big- box stores opened in suburban locations and
fewer in urban and rural ones,5 and supermarket chains with consolidated food supply and distribution systems.6 These shifts, and increasing suburbanization, mean that fewer people now live within walking distance—or a short bus or subway ride—to the grocery store.7 This spatial dislocation has been made possible, in large part, by federal transportation policy that financed highway development, supported increased truck transportation of goods, and encouraged personal automobile use through subsidies that expanded roadways and parking. For example, one study puts the total “tax subsidy” to motor vehicle users in the range of $19–$64 billion per year.8
Today, inner-city9 and rural10 neighborhoods have fewer and smaller grocery supermarkets, with poorer selections of healthy foods and higher prices than their suburban counterparts. Urban neighborhoods, conversely, have an abundance of smaller convenience stores and fast-food outlets, which offer disproportionately higher amounts of foods of poor nutritional quality.11 A decline in wholesale and retail farmers’ markets12 also paralleled the decline of grocery supermarkets in urban and rural locations, although farmers’ markets have recently seen a dramatic rise.13 Nonetheless, farmland in metropolitan areas, where a majority of fruits and vegetables are grown, continues to be consumed by urban sprawl.14
For low-income and urban residents, for people of color, and for immigrants—all of whom tend to own fewer cars than affluent and middle-class whites,15 the paucity of nearby supermarkets leads to higher rates of diet-related morbidity and mortality,16 and even greater stress related to grocery shopping. Conversely, relatively easy access to supermarkets is associated with higher household consumption of fruits and other positive dietary behaviors.17 Disparities in the number and size of supermarkets have been documented by race even after controlling for income, with African American neighborhoods most adversely affected.18 Higher costs,
Sustainable Food Systems
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 1 7
<
<
S u
s ta
in a
b le
F o
o d
S y
s te
m s
ch. 7
poorer selections, and lower quality of foods in low-income neighborhoods mean that taxpayer-funded nutrition programs such as the food stamp program (more recently known as SNAP, or the Supplemental Nutrition Assistance Program) don’t go as far as in better-off neighborhoods. Lack of affordable, neighborhood-based food outlets also forces low-income households to rely more on emergency food programs such as food pantries that—dependent on private donations and government surpluses—stock little in the way of healthy foods. What’s more, poor diets conspire with poor air quality, fewer parks and fitness facilities, poor quality housing, high levels of crime, noise, and other social and environmental stressors in low-income neighborhoods.
Increased Dependence on Use of an Automobile for Grocery Shopping
Grocery shoppers tend to prefer to travel to supermarkets by car, in part because of the one- stop design of supermarkets and their proximity to large-scale shopping districts with abundant, available parking, all of which discourage walking or biking. Vehicles save time and can help shoppers reach more stores, combine trips, and transport heavy packages easily, including in inclement weather.19 One Austin, TX, study found that few people substitute walking for driving to the grocery store, even if pedestrian or cycling access is good.20 Even the poor who do not own cars often borrow them, ask for rides from friends, or take taxis to do grocery shopping21; however, transportation and walking remain critical in providing the mobility needed to access grocery outlets for these families.22
Public bus routes and schedules, even in well- serviced communities, are typically planned in ways that disadvantage food-shopping trips needed during weekends and evenings. A typical bus system is also planned around a central hub, a design that often lengthens travel time to more peripherally located supermarkets. And high levels of required parking for supermarkets may make them less of a priority
in transportation system planning. Perversely, such land use policies may exacerbate the peripheral location of supermarkets. Research from the United Kingdom suggests that when land use policies discourage new supermarket development on the urban fringe, stores invest more in expanding and refurbishing the older stores based closer to the urban core.23
People who live in low-income households are underserved by both the food24 and transportation25 systems. In 2007, food insecurity rates in the United States rose even before the sharp economic declines of 2007–08. Overall, 36.2 million persons—or 12.2 percent of Americans, mostly women, minorities, and children—struggled with hunger. In May 2008, more than 28 million persons participated in the food stamp program, a 32 percent increase in five years; yet the program reaches only two out of three eligible households.26 Access to food stamp offices for these populations often is undermined by the distances needed to travel, lack of evening hours of operation, and limited public transportation within communities.27
Food stamp recipients are also vulnerable to losing benefits due to lack of transportation to recertification appointments.28 For a variety of reasons, farm worker households face a higher risk of food insecurity.29 At the same time, the poorest Americans who have cars spend disproportionately more of their household budget than the national average on the purchase, operation, and maintenance of automobiles30; are subject to higher interest rates when attempting to purchase a car; spend disproportionately more on commuting to work31; and are more likely to miss work due to car problems.32
Low-income populations are comprised disproportionately of women, who also tend to make more trips related to childcare and household servicing—including 75 percent more grocery shopping than men do.33 Shoppers tend to mix and match stores for food shopping based on criteria related to product mix, price, quality, and quantities desired and also the
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 1 8
>
>
C h
a p
te r
7
Sustainable Food Systems
relative proximity of suitable outlets to their homes and workplaces.34 Rural residents shop for groceries at more stores than do urban residents and travel farther to reach the stores.35
Nonetheless, the scarcity of large supermarkets in poor neighborhoods and the economic pressures that force low-income residents to shop in smaller stores in their neighborhoods remain significant factors in why poor people pay more for food.36 Federal nutrition programs such as food stamps and WIC (Women, Infants, and Children) do not pay for transportation costs incurred by households to procure food.37 The Summer Food Service Program, which is under-enrolled in large part because of transportation barriers, provides small multiyear, competitive grants for innovative approaches to overcome such barriers.38
Although transportation costs represent only a modest share of the cost of food consumed at home—an estimated six to 12 percent39— energy disruptions can cause significant hikes in the price of food, as was experienced in the first half of 2008.40 This is because both the food and transportation systems are highly energy intensive. Also, declining diesel oil prices through the 1990s tended to restrain food transportation cost increases; this trend is unlikely to continue for long. Rising energy costs hit low-income households especially hard as they struggle with maintaining an automobile, higher utility costs, and buying enough food for their families.
Dispa rities in a f fordable Tra nspor tation a lternatives for a g ri-food System Workers
Low-income rural households also experience problems with access to affordable transportation.41 Agri-food workers’ burdens in this regard are especially heavy, and the least paid among them also tend to be predominantly members of groups that
are also vulnerable within communities: disproportionately younger (or older), female, immigrant (including those without legal residency status), and people of color. Most farm laborers and food service workers earn close to the minimum wage and get few additional benefits or perks. According to the U.S. Department of Labor, the national median wage in 2007 for waiters and waitresses was $7.62 per hour, and that for farm workers and laborers was $9.78 per hour. By comparison, the median for all occupations was $15.10 per hour. Dependence on public transportation reduces employment access far more than any other factor42; when people who work at or near the minimum wage must make longer journeys to work, their income does not rise.43
Agri-food workers also experience greater transportation challenges because of the dispersal of jobs across the metropolitan and rural landscape. As a subset, farm workers have special difficulties accessing transportation.44 In one study of farm workers in Mendocino County, CA, two out of five workers depended on rides from family members and other acquaintances; those who incurred transportation costs (i.e., were not living on farms) reported a mean cost of $40 per week—or roughly 16 percent of the average weekly wage—with a median of $30 per week.45 As other papers in this collection show, strong evidence exists of a correlation between lack of access to adequate mobility and lack of access to opportunities, social networks, and health-supporting services such as clinics and pharmacies. At the same time, anecdotal evidence suggests that farm workers with transportation issues are at higher risk for injury as a result of their greater reliance on older “junker” cars, traveling in the early hours of the morning, lower safety requirements (such as seatbelts) for farm-worker transport vehicles, and lax enforcement of safety regulations for such vehicles.46
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 1 9
<
<
S u
s ta
in a
b le
F o
o d
S y
s te
m s
ch. 7
Rail Water Truck Air
Fuel (kilojoules per ton-kilometer) 677 423 2,890 15,839
Emissions (grams per ton-kilometer)
Carbon Dioxide 41 30 207 1,260
Hydrocarbons 0.06 0.04 0.3 2.0
Volatile Organic Compounds 0.08 0.1 1.1 3.0
Nitrogen Oxide 0.2 0.4 3.6 5.5
Carbon Monoxide 0.05 0.12 2.4 1.4
Average distance by truck to Chicago Terminal Market (continental U.S. only)*
# States supplying this item
% Total from Mexico
Grapes 2,143 miles 1 7
Broccoli 2,095 miles 3 3
Asparagus 1,671 miles 5 37
Apples 1,555 miles 8 0
Sweet Corn 813 miles 16 7
Squash 781 miles 12 43
Pumpkins 233 miles 5 0
* Information for this chart is based on the weighted average source distance—a single distance figure that combines information on distances from production source to consumption or purchase endpoint. For more information on method, refer to Pirog and Van Pelt, 2002 (endnote 55).
Table 1. Energy Consumption and Emissions by Different Freight Modes54
Table 2. Average Distance by Truck to Chicago Terminal Market, 199855
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 2 0
>
>
C h
a p
te r
7
Sustainable Food Systems
Tra nspor tation , a g ri-food System Susta inability, a nd Dispa rate Community a nd r eg iona l impacts
In global commerce, the agri-food sector presents special opportunities and challenges when it comes to transportation. Food, especially produce, is different from other commodities in that it is perishable and requires timely delivery and careful handling—including temperature control and cooling—to prevent spoilage. Globalized transportation of food enables surpluses from one region to efficiently make up for shortfalls in other regions, and one hemisphere to continue to supply familiar foods to the other following the latter’s growing season; it also makes available new markets for local agriculture.
Because both modern agriculture and transportation today are more energy intensive than in the past, when energy costs go up, food costs rise dramatically, making the global food system especially susceptible to inflationary pressures and communities vulnerable to rising
energy prices.47 Additionally, the greater reliance on faraway sources for food has resulted in a loss of access to markets for many local and smaller-scale farmers, which, when combined with the loss of metropolitan farmland to urban sprawl, only exacerbates the vulnerability of food systems in many parts of the country.48 Increased truck-miles and air-miles in food transportation worsen air pollution and climate change; increased roadway congestion causes more accidents; the loss of nearby slaughter and packing facilities increases travel times and stress for animals. Together, these factors accumulate social, economic, and environmental costs that are greater than what food source communities get in return for their products.
Increased road- and Air-miles in Food Transportation
Environmentalists are increasingly concerned about the distance food travels from field to plate—typically 1,500 road-miles— which creates unsustainable demands on transportation, air quality, climate, and energy systems. One study revealed that the average distance for fruits transported to the Jessup, MD, terminal market was 2,146 miles, while
Table 3. Estimated Fuel Consumption, CO2 Emissions, and Distance Traveled for Conventional, Iowa-based Regional and Iowa-based Local Food Systems for Produce56
Food system type/type of truck Fuel consumption (gal/year)
$ value of fuel (2001 prices)
CO2 emissions (lb/year)
Distance traveled (miles)
Conventional/semitrailer 368,102 581,601 8,392,727 2,245,423
Iowa regional/semitrailer 22,005 35,208 501,714 134,230
Iowa regional/midsize truck 43,564 69,702 993,243 370,289
Iowa local–CSA farmers’ market/ small truck (gas)
49,359 78,974 967,436 848,981
Iowa local–institutional/ small truck (gas)
88,265 141,224 1,729,994 1,518,155
ch. 7
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 2 1
<
<
S u
s ta
in a
b le
F o
o d
S y
s te
m s
the average for vegetables was 1,596 miles.49 Transportation accounts for about 11 percent of the energy use in the food system.50 About 93 percent of fresh produce transported between cities in this country was carried by trucks, according to a 1996 USDA study.51 In addition to general emissions that affect our climate, truck emissions create disparate air quality- related health impacts on low-income and minority neighborhoods because of their greater proximity to highways and truck terminals.52 Causing even more concern is the rapidly growing air transport of food, which creates the highest CO2 emissions per ton.53
Table 1 shows the energy consumption and tailpipe emissions for different modes of transportation. Of course, the actual mode of transportation and the distance traveled varies by specific food product and its origin. Distances traveled by different products shipped from within the continental United States are given in table 2 (which also shows how much averages derived from travel within the continental United States may understate actual distances if a larger share of a product comes from Mexico). Energy consumption and emissions for different kinds of truck transportation participating in distinct local, regional, and the conventional national food system considered by Pirog et al. (2001) are given in table 3. This last table underscores the point that the sustainability of local food systems is mediated by the specific mode and fuel used in transporting foods.
Finally, the transportation sector is responsible for more than one-quarter of all emissions causing climate change.57 Many agri-food advocates are increasingly concerned about the implications of climate change for future agricultural productivity and food security in poorer regions of the world, given the greater likelihood of drought, soil erosion, extreme weather events, and higher pest prevalence.58 More sustainable transportation, together with an agri-food system that reduces energy and transportation demand, would help reduce burdens on future agriculture globally.
Increased Consolidation of the Food Industry and Disparate Social and Spatial Impacts
Industrial agri-food’s specialization in certain crops has concentrated food production in regions and uses large quantities of fossil fuels to ship food around the country and the world. For example, 95 percent of the nation’s processed tomatoes and just under one-third of the fresh tomato crops come from California.59 In 2007, nearly $152 billion of agricultural products crossed U.S. borders as imports and exports, representing more than half the value of agricultural products sold by U.S. farms that year.60 This specialization, however, has reduced many “receiving” regions’ previous diversity of production and made them more vulnerable to shocks in the system. For example, agricultural modernization has favored large farm size, crop monocultures, mechanization, and increased chemical inputs. Moreover, research points to rising food insecurity among low-income farmers in some countries as subsistence production has been replaced by export-oriented mono-cropping.61 These challenges, of course, affect rural communities and predominantly smaller-scale and low-income farmers whose market reach is hurt by the loss of localized infrastructure and support for logistics (management of the movement of goods). Cheap energy and transportation subsidies have therefore enabled the consolidation and globalization of the agri-food sector.
The case of retail supermarkets and resulting disparities in healthy food access was presented in the first section of this paper.62 The increase in food miles traveled results from: (a) restructuring of logistical systems due to stricter requirements from retailers’ management of inventories; (b) realignment of supply chains so that more of the product from farm to supermarket is owned by a single firm or a strategic partnership of firms (which has happened to reduce costs and risks and also increase responsiveness to consumers); (c) shifts in production and distribution scheduling
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 2 2
>
>
C h
a p
te r
7
Sustainable Food Systems
decisions, with negotiated coordination replacing market coordination; and (d) changes in management of transport resources such as increasing the use of air instead of road transport for food.63
The consolidation of processing, wholesaling, and distribution operations results in fewer, larger, and more efficient facilities and the closure of more local and regional processing plants, warehouses, and related facilities. As a result, the plant closures cause greater economic insecurity and health risks for nearby communities.
The transportation sector also has experienced consolidation, with somewhat similar results. Railroad consolidations, for example, have increased the number of captive customers and, while the monopolization helps railroads financially, it also tends to distort the location of economic activity, creating or exacerbating regional disparities64—and therefore vulnerabilities—in the food system.
food Versus fuel a nd r elated Hea lth impacts
The production of the most popular forms of biofuels—corn ethanol and palm oil—threatens to cause a major increase in greenhouse gas emissions.65 In the United States, corn ethanol poses special concern because of its net negative energy balance (that is, more energy is required to produce a gallon of corn ethanol than can be gained from it) and because its production and use contribute to air, water, and soil pollution.66 Some food security advocates worry that the continued expansion of biofuels is raising food prices in this country67 and elsewhere and causing malnutrition in many developing countries.68 Still others suggest that corn ethanol has a worse impact on the environment and human health than do conventional fuels such as gasoline and diesel.69 There are direct transportation impacts as well: as corn use shifts from exports and animal-feed use to ethanol production, grain transportation
is affected because of changes in quantities transported to diverse destinations and modes of freight used for raw and finished products.70
To summarize the paper’s analysis, transportation policies and subsidies—when combined with cheap energy over the past six decades—have thus created patterns of spatial dispersion of people and food outlets over the metropolitan landscape in ways that pose special hardships for low-income food shoppers as well as agri-food workers in urban and rural communities. Transportation has also enabled structural change in the agri-food sector so that decisions made in the name of economic efficiency have generated many negative environmental, social, health, economic, and spatial consequences, along with increased costs and risks to society as a whole. These consequences call for a review of the basic goals and purposes of transportation policy so that environmental, social, and health needs and goals take priority over private gain.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 2 3
<
<
S u
s ta
in a
b le
F o
o d
S y
s te
m s
ch. 7
Elements of a Susta inable a g ri-food System
A primary contribution of the agri-food system is to deliver adequate nutrition to support the health of human communities now and into the future. However, contemporary industrial agri- food practices also create direct health problems (such as through the effects of pesticides on farm workers or widespread obesity among youth and adults) and indirect health problems (through diminished quality of air and ground water and the pervasive use of antibiotics in meat production, for example). These practices also endanger the very base upon which the food system depends, thereby threatening future food security and health. That is, they are unsustainable.
A sustainable food system promotes the health of individuals, communities, and the ecosystem. As this paper shows, transportation is implicated in many of the pathways linking the agri-food system and health. Sustainable food systems are typically organized around the following principles, on which consensus more or less exists:
• produce and distribute food so that all persons have adequate access to nutritious foods within neighborhoods;
• respect and operate within the biological limits of natural resources such as soil, water, and species;
• minimize energy inputs, recycle resources, and use renewable energy and other resources;
• support vital and diverse urban and rural economies;
• enable viable livelihoods and fair trade among producers, processors, distributors, retailers, and consumers;
• provide safe, fair, and satisfying working conditions for workers;
• treat animals humanely;
• sustain the amount and quality of land needed for food production; and
• promote democratic processes in decision making related to food and nutrition.71
Tra nspor tation G oa ls
The following goals are proposed for transportation policy and programs to help build sustainable food systems that promote human, community, and environmental health in the United States and globally.
1. Healthy food access for all, with special focus on the needs of low-income communities and communities of color, through appropriate land use policies and affordable transportation alternatives.
2. Affordable and reliable transportation alternatives for low-income agri-food workers so that they may have access to employment, food sources, and other basic needs.
3. Transportation policies and programs that prioritize regional linkages over national and global ones as they relate to food systems so that local producers are connected with local eaters; regional economic development is promoted through localized networks and infrastructure; small-scale farms are supported; air pollution and climate change impacts are reduced; and risks associated with agri-food concentration, dependence on distant sources, and energy price hikes are mitigated.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 2 4
>
>
C h
a p
te r
7
Sustainable Food Systems
Goals Desired Policies and Programs
Reduce disparities in access to healthy foods
Support local and metropolitan land use policies and planning for increasing neighborhood-based access to food retail sites such as stores, farm stands, and urban agriculture sites72:
• Promote smart growth development that supports multiple transport modes and contains grocery stores, urban agriculture sites, and farm stands.
• Encourage transit oriented neighborhood design to include grocery outlets.
• Retrofit older neighborhoods for pedestrian, bike, and transportation access to food outlets and urban agriculture sites.
• Reduce required parking for grocery stores in exchange for public bus connectivity during peak grocery shopping times (weekends, especially).
Support policies and programs that promote transportation access for low-income residents to grocery outlets and other healthy food sites:
• Promote paratransit or public-private partnerships for shuttle programs sponsored by supermarkets,73 congregate (subsidized) housing facilities and community-based nonprofits to provide affordable rides for grocery shopping.
• Develop and promote “grocery bus” routes74 with weekend service to connect low-income neighborhoods to full-service supermarkets, food pantries, and urban agriculture sites.
• Support community-based programs to create mobile markets or grocery van-delivery in urban and rural communities.75
Require transportation support in federal nutrition programs:
• Include transportation support for WIC, food stamp (SNAP), Summer Food Service, and farmers’ market-related nutrition programs to access healthy foods.76
• Provide transportation support for small-scale farmers to sell at farmers’ markets in or near low-income urban or rural areas.
Table 4. Desired Policies and Programs to Address Transportation-Related Agri-food Problems: Opportunities for Success
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 2 5
<
<
S u
s ta
in a
b le
F o
o d
S y
s te
m s
ch. 7
Goals Desired Policies and Programs
Promote safe and affordable transit for agri-food workers
• Increase funding for job access and reverse commutes for low- income employees, including agri-food workers.
• Encourage metropolitan transportation system design to increase access for low-income agri-food workers in processing, wholesale, and retail jobs in metropolitan areas.
• Encourage paratransit options (vanpools) for farm workers.77
• Review rules related to vehicle conversion for farm-worker transportation and safety equipment/use to increase transportation safety and minimize accidents.
Promote agri-food sustain- ability
• Support within transportation law small-scale farmers’ and processors’ transportation of product to farmers’ markets and other local outlets.
• Encourage and support cleaner and more efficient vehicles, especially smaller trucks used for local food transportation.
• Review and adjust tax structure as it relates to overall transportation subsidy so that social and environmental costs associated with emissions in agri-food transportation are reflected in prices, especially in the case of air transportation of foods.
• Promote use of more sustainable modes of freight for long- distance food transportation, such as rail and water.
• Increase competitive access to rail for food transport (via separation of ownership of rail infrastructure from that of rolling stock, e.g. rail cars), increase subsidy for rail relative to road and air, and break up geographic concentration of control over railway infrastructure (e.g. tracks) to increase competition.
• Prioritize local and regional food transportation networks and infrastructure over long-distance ones.
• Support the development of mobile kitchens and processing facilities in urban and rural communities.
• Promote metropolitan planning to prevent sprawl, preserve farmland, and promote urban agriculture in transportation- related rights of way.78
Prioritize agriculture for food and promote sustainable biofuels
• Minimize competition in agricultural production between food and fuel (since most biofuel is used for transportation) by giving food a clear priority.
• Support the development and promotion of genuinely sustainable biofuels.
• Support the widespread conversion of waste cooking oil into biodiesel.
• Internalize social and environmental costs of corn-ethanol production and end subsidies for biofuels that are sourced from food grains.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 2 6
>
>
C h
a p
te r
7
Sustainable Food Systems
Goals Desired Policies and Programs
General recommendations • Promote greater coordination between transportation and agri- food policies and programs.
• Provide greater support for intra-regional (versus inter-regional) transportation.
• Encourage tighter links among transportation planning, policy, and programs and anti-sprawl and pro-urban planning.
• Facilitate improved regional coordination to support multiple transportation modes and programs and diverse trip purposes and needs.
• Develop transportation systems at the regional level to create positive economic impact, including through regional food systems.
• Consider USDA’s Community Food Projects Competitive Grants Program as a model to promote community- and region-based collaborative approaches to improve food access, market access to small-scale farmers, and affordable agri-food system transportation.79
4. The agri-food system reconfigured as a resource to reduce energy and transportation demands and related problems through the development of more local food systems and truly renewable fuels.
Tra nspor tation Policies: Oppor tunities a nd Ba rriers
Many of the problems outlined in the first part of this paper are rapidly turning into emergencies—if they are not already emergencies. Their simultaneous occurrence presents something of a perfect storm for health and sustainability concerns. The upcoming authorization of the federal transportation bill offers a significant opportunity to make headway in addressing—and correcting— these problems. The crises related to rising incidence of obesity and diet-related diseases, climate change, and national energy and food security provide impetus to increase access to healthy foods as part of a preventive
approach to improve health, build localized food systems, reduce the energy intensity of the agri-food system, and help the agri-food system contribute to the creation of sustainable transportation systems.
Specific recommendations that link policies and programs to emerging problems are presented in table 4.
Notwithstanding the policy and programmatic opportunities outlined in table 4, those seeking to meet health goals within transportation legislation face many barriers to success. These are outlined below.
The most obvious barrier lies in the structure of transportation funding, legislation, and governance—especially at the federal level. The majority of transportation funds are allocated by formulas tied to modes and trip purposes; this makes it hard to achieve the goals outlined here within the existing structure of transportation policy and policymaking. The
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 2 7
<
<
S u
s ta
in a
b le
F o
o d
S y
s te
m s
ch. 7
problem is that, at the national level, we fund and manage transportation programs primarily by mode, rather than by urgent societal needs or compelling national goals. We also allocate funding by state, making achievement of national goals even more difficult. This is further complicated by competition between donor and donee states (that is, states that send more gas taxes to the federal transportation budget than they receive in transportation funding, or vice versa), a situation made worse in the current recession because many of the donee states are in the hard-hit, former manufacturing belt of the Midwest. Moreover, we fund transportation through a myriad of other (non-Department of Transportation) agencies, including the departments of Agriculture (USDA) and Health and Human Services (HHS), leading to further fragmentation by sector. Such fragmentation of the program is the cause of many transportation-related problems experienced by communities and within metropolitan regions.
The problems posed by programmatic fragmentation suggest that addressing food- and health-related transportation problems, as recommended in this paper, could increase overall transportation inefficiency, if they are not coordinated well, that is, more silos are not the solution. Instead, the programs and policies recommended here must be tied to land use policies that reduce transportation demand, improve access and regional connectivity (regardless of trip mode or purpose), and improve coordination between transportation providers and the system as a whole. In addition, policy must prioritize regional food system transportation connectivity over national or international ones, support more energy- efficient and less polluting modes and vehicles, and more effectively use spare capacity in existing programs to support food access for low-income consumers and regional market access for small-scale farmers. This will require coordination across federal agencies such as Department of Transportation (DOT), USDA, and the Environmental Protection Agency (EPA).
Lack of precedence within transportation legislation for key asks: To date, there is little precedence for transportation legislation incorporating many of the policies recommended in this paper. Some policymakers may view the recommendation to increase transportation assistance to low-income households participating in federal nutrition programs as more appropriately falling within the agriculture law. USDA already funds transportation for rural providers of the Summer Food Service Program, which feeds low-income children.80 Similarly, the recommendation to prioritize agriculture for food over fuel may be viewed as falling under agriculture or energy, rather than transportation, even if most of the corn ethanol is destined for transportation-related uses.
Highways and roads (rather than access) as the primary orientation of transportation policy: Despite the progressive changes ushered in by ISTEA and its successors, transportation policy continues to be driven by a dominant orientation toward roads and highways, rather than toward multi-modality that provides access to goods, services, employment, healthy food, etc., thereby meeting community and regional needs and goals. Local land use decisions often follow, rather than drive, regional transportation planning by metropolitan planning organizations. Because land use decisions are local, more support is also needed than is available within the transportation legislation for transportation planning that effectively integrates land use and transportation to promote smart growth, that is, increase mixed-use, transit oriented development and neighborhood-based access to basic needs. Similarly, many advocates believe that transportation programs and funding tend to be designed to serve the interests of powerful groups—highway builders, auto manufacturers, and petroleum corporations— and that relationships of power and patronage, rather than systematically derived community needs, drive transportation policy.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 2 8
>
>
C h
a p
te r
7
Sustainable Food Systems
Impending revenue shortfalls from gas taxes: The expected shortfalls in the Highway Trust Fund present a challenge to funding new programs in the transportation legislation. Policymakers will need to find additional sources of funding that are adequate, sustainable, and fair. To this end, policies that improve health can result in savings in other areas, such as healthcare cost savings81 and can present new funding alternatives to fuel taxes. Such solutions go beyond the oft-suggested road and congestion pricing, both of which may further disadvantage the communities already at risk from current policies. More research is needed related to the net benefits and costs of transportation programs, including those suggested in this paper.
Convergence Oppor tunities
Efforts to build sustainable food systems are inherently boundary spanning and require work across disciplines, sectors, professions, and geographic scales. The federal transportation law authorization process provides unique opportunities to build partnerships among interests in sustainable agri-food systems, smart growth, public health, community economic development, anti-poverty and social justice, labor, energy security, and climate change mitigation.
Coalitions that have emerged to advocate for transportation policy reform, such as the Transportation Equity Network, Transportation for America, Surface Transportation Policy Project, Complete Street Coalition, and Smart Growth America, are calling for proposals with broadly similar goals as those suggested herein, even if they are largely silent on agri-food issues addressed in this paper.82 Among the coalitions advocating for more sustainable agri-food systems or elements thereof are the Community Food Security Coalition, National Sustainable Agriculture Coalition, Food Research and Action Center, National Family Farm Coalition, and American Farmland Trust.83 Past efforts by these
groups to bring attention to sustainable agri- food issues within the transportation law have borne little, if any, fruit. We hope that the broad health rubric under which these papers are assembled will help coalesce the many groups mentioned above and attract new groups into the fold to add power to related transportation advocacy.
Additionally, the specific proposals made by this paper call for greater collaboration and coordination among various departments at the federal and state levels. For example, the proposals in this paper could benefit from partnerships among:
• DOT and USDA (and Department of Health and Human Services or the Department of Education when applicable) to provide transportation assistance to nutrition program participants in order to procure food, to improve neighborhood-based access to healthy foods through the use of transportation resources, and to support small-scale farmers’ efforts to bring products to local markets in underserved areas. This would increase participation in nutrition programs such as SNAP, WIC, Summer Food Service, and Farmers’ Market Nutrition; it would also increase the benefits of participation, improve health, and reduce healthcare costs.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 2 9
<
<
S u
s ta
in a
b le
F o
o d
S y
s te
m s
ch. 7
• DOT, USDA, and the Department of Labor to provide affordable transportation for urban and rural agri-food workers to access jobs, food, healthcare, and other vital services.
• DOT, USDA, the Department of Energy, and the EPA to support the development of more truly renewable energy sources in environmentally sensitive ways, including through the use of switchgrass and waste cooking oil; to support the development of fuel-efficient vehicle and transportation systems; and to discourage the use of food grains for producing fuel. Such cooperation is sorely needed to eliminate the competition between food and fuel.
• USDA, DOT, and the EPA to mitigate the problems caused by long-distance transportation of food in international trade.
Conclusion
This paper presents four clear problems impacting the interaction between agri-food and transportation systems and suggests possible actions that could solve them. Some solutions can be addressed through transportation legislation, but clearly efforts need to extend to legislation that addresses energy, agriculture, child nutrition, labor, and health and human services.
Whatever the final mix of policies, successful efforts will result in affirmative responses to the following questions:
• Do neighborhoods provide convenient access for all residents to healthy foods and other basic goods and services? Do they allow food shopping without the need for a car?
• Beyond basic accessibility, do transportation policies and programs enhance local and regional quality of life through improved multi-modal access for all residents to the region’s resources and destinations and through reduced congestion?
• Does the regional transportation infrastructure support local food producers and processors to efficiently market to local consumers, in addition to national distribution channels?
• Do transportation policies support modes of freight, fuel choices, and vehicle designs such that air and water pollution, greenhouse gas emissions, and energy use are minimized?
• Are the currently externalized social, health, and environmental costs and increased risks posed by the global, industrial food system internalized in the price of food and transportation? Are associated costs and benefits fairly distributed across diverse income and racial groups in urban and rural areas?
• Does the agri-food system support transportation policies with renewable and efficient options for energy that reduce environmental impacts on air, water, and climate; minimize competition with food production; and reduce dependence on foreign sources for energy?
The transportation authorization process presents opportunities to break bad habits, extend positive developments from the past, and launch bold new initiatives that set us on a better course. Promising directions that build on positive aspects of SAFETEA-LU include, for example, correcting inequities in funding across states; providing dedicated funding to states to meet air quality requirements; and creating pilot programs to test alternative transportation funding schemes (which should be extended beyond tolling and road pricing schemes that may hurt the transportation-disadvantaged).
Clearly, other strategies are needed to eliminate disparities and problems caused by the current agri-food–transportation system linkage: extending transportation programs to increase access to healthy food and agri-food employment, reducing railroad concentration, ending competition between food and fuel, and more.
p
g . 1 3 0
>
>
Traffic Injury Prevention: ch. 8 A 21st-Century Approach LARRY COHEN, M.S.W. Founder and Executive Director
JANANI SRIKANTHARAJAH, B.A. Program Coordinator
LESLIE MIKKELSEN, R.D., M.P.H. Managing Director, Prevention Institute, Oakland, CA
ABSTRACT >> Traffic injuries and deaths exact a huge toll on our finances, our families, and our future. There are opportunities in the upcoming authorization of a new federal transportation bill to promote safety for all travelers. More broadly, safety for all travelers must become a national health and transportation priority. Advocates for injury prevention should collaborate with public health experts (specialists in chronic disease prevention, for example) and partners in other sectors (such as economic development) to promote a broad vision for health and equity in transportation policy.
The overarching policy goals that support traffic injury prevention are to: (1) promote the safe transportation of all travelers by improving infrastructure in communities; (2) reduce the number of vehicle miles traveled by promoting alternative modes of transportation, including public transportation, walking, and bicycling; and (3) protect drivers and passengers through continued improvements in vehicle safety, occupant protection, and road safety. This paper describes specific strategies to achieve these goals.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 3 2
>
>
C h
a p
te r
8
Traffic Injury Prevention
CONTENTS
Introduction .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . 133
Achievements in Traffic Injury Prevention . .. .. . 134
Prioritizing Traffic Injury Prevention for All Modes of Travel .. .. .. .. .. .. .. .. .. .. . 135
The Continuing Burden of Traffic Injuries and Deaths .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . 137
Disparities in Traffic Injuries and Deaths . .. . 137
Other Populations with Greater Risk . .. .. .. . 139
Transportation Injury Prevention Strategies . .. . 139
Land Use .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . 139
Road Design .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .140
Public Transportation .. .. .. .. .. .. .. .. .. .. .. .140
Speed Limits .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . 141
Impaired Driving Laws . .. .. .. .. .. .. .. .. .. .. . 141
Bicycle Helmet Laws . .. .. .. .. .. .. .. .. .. .. .. . 141
Vehicle Design Standards .. .. .. .. .. .. .. .. .. . 142
Seat Belt Laws. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . 142
Motorcycle Safety Laws . .. .. .. .. .. .. .. .. .. . 142
Child Safety Seat Laws .. .. .. .. .. .. .. .. .. .. . 142
Graduated Driver Licensing .. .. .. .. .. .. .. .. . 143
Truck Regulations.. .. .. .. .. .. .. .. .. .. .. .. .. . 143
Challenges to and Opportunities in Traffic Injury Prevention Policy .. .. .. .. .. .. .. .. .. .. . 143
Conclusion.. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . 145
LIST OF ILLUSTRATIONS
Tables
1. Traffic Injury Prevention Highlights . .. .. .. .. . 134
2. The Haddon Matrix (with examples) . .. .. .. . 135
3. The Spectrum of Prevention.. .. .. .. .. .. .. .. . 136
4. SAFETEA-LU Programs That Support Injury Prevention. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .144
5. Federal and State Government Support for Traffic Injury Prevention .. .. .. .. .. .. .. .. . 145
Graphs
1. U.S. Traffic Fatalities by VMT and Per 10,000 Population .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . 138
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 3 3
<
<
Tr a
ff ic
I n
ju ry
P re
v e
n ti
o n
ch. 8
introduction
While getting off a streetcar in New York City on September 9, 1899, Henry Hale Bliss was struck by an electric-powered taxicab and suffered injuries so severe—his skull and chest were crushed—that he died the next day. Bliss thus became the first person killed by a motor vehicle in the United States. The taxicab driver was arrested and charged with manslaughter but was later acquitted on the grounds that the death was unintentional. While the legal proceedings considered where responsibility for Bliss’s death lay, there was no discussion of what could have been done to prevent the crash.1
What was unprecedented in 1899 is unremarkable today. Traffic crashes are the leading cause of death in the United States for people ages one to 34,2 and by 2020, traffic- related deaths will be the third-leading cause of death worldwide.3
Traffic injuries and deaths exact an unnecessary economic toll. In 2000, motor vehicle crashes in the United States cost $230.6 billion in emergency services, medical treatment, legal procedures, insurance administration, property damage, lost workers’ productivity, and travel delays.4 That figure represents 2.3 percent of the nation’s gross domestic product.5
In 1900, motor vehicle travel was considered a novelty, and the risks to health and safety were largely overlooked. Subsequent improvements in manufacturing made cars more affordable and available, benefiting commerce, communications, and personal mobility. In 1900, an estimated 8,000 automobiles were registered in the United States. By 1950 there were 50 million, and by 2001, more than 230 million vehicles and 193 million licensed drivers were on the road.6 The current number of cars and drivers, along with the extensive networks of roads and highways around the nation, would have been inconceivable in 1899 but are accepted as norms of transportation today. Traffic injuries and deaths are frequently
considered uncontrollable aspects of America’s love affair with the car. This may account for the fact that traffic crashes are too often ignored as a major contributor of premature death and disability, the consequence of which is a missed opportunity to improve health and reduce costs.
In light of ever-shrinking federal, state, and local budgets, the authorization of a new federal surface transportation bill is an opportunity to structure transportation programs to reduce the burden on the healthcare system, the economy, and society at large. National and international experts on traffic injury prevention, including the U.S. National Highway Traffic Safety Administration (NHTSA), the U.S. Centers for Disease Control and Prevention (CDC), and the World Health Organization, increasingly reject the notion that traffic injuries are the inevitable price we pay for modern travel.7
Many transportation policies and practices that lead to traffic injuries also contribute to chronic diseases that result from physical inactivity, poor air quality, and other environmental factors that are the consequences of our car culture. Linkages between injury prevention and other health fields should be developed to foster a national transportation strategy that forges solutions to these intersecting problems. Such strategic partnerships can help catalyze a revamped national transportation strategy that is central to policymakers’ efforts to address a range of critical challenges: the economy, climate change, the limited supply of fossil fuels, and soaring healthcare costs. A transportation agenda that emphasizes health, equity, environmental protection, jobs, and an improved quality of life requires collaboration from all sectors.
The overarching policy goals that support traffic injury prevention are to: (1) promote the safe transportation of all travelers by improving the physical infrastructure in communities; (2) reduce vehicle miles traveled by promoting alternative modes of transportation, including public transportation, walking, and bicycling;
Traffic Injury Prevention H
e a
lt h
y, E
q u
it a
b le
T ra
n s
p o
rt a
ti o
n P
o li
c y
p g
. 1 3 4
>
>
C h
a p
te r
8
and (3) protect drivers and passengers through continued improvements in vehicle safety, occupant protection, and road safety.
achievements in Tra f f ic injur y Prevention
While it is impossible to forecast the exact circumstances of traffic crashes, these incidents are not isolated events but are both predictable and preventable. The news and entertainment media often speak of traffic “accidents,” but the word implies—erroneously—that the event is happenstance and arbitrary.
Dr. William Haddon, Jr., the first director of the National Highway Safety Bureau, which in 1970 became the National Highway Traffic Safety Administration, brought an emphasis on injury prevention to the government’s transportation policies and practices. Dr. Haddon is also recognized for developing the Haddon Matrix (see table 2).
By deconstructing the sequence of events contributing to traffic-related injuries, Dr. Haddon developed effective strategies to prevent crashes and limit injuries. By integrating education, legislation, and enforcement, health and safety advocates as well as government officials have bolstered Dr. Haddon’s research by requiring the
1923: Garrett Augustus Morgan, an African American traffic safety innovator, invents the modern traffic signal to reduce the high risk of collisions he observed on roadways shared by horse- drawn buggies, pedestrians, and automobiles.
1924: President Herbert Hoover convenes the National Conference on Street and Highway Safety, marking the first presidential initiative to bring attention to traffic safety.
1964: Ralph Nader’s book Unsafe at Any Speed: The Designed-In Dangers of the American Automobile is published—another milestone that attributes injuries not just to driver error but also to vehicle design flaws and describes auto executives’ resistance to vehicle safety features, most notably General Motors’ Chevrolet Corvair. Following the book’s release, public pressure mounts, forcing President Lyndon Johnson to call for tighter regulation.
1966: President Johnson signs The Traffic and Motor Vehicle Safety Act and The Highway Safety Act into law, authorizing the National Highway Safety Bureau (now the National Highway Traffic Safety Administration (NHTSA)) to set vehicle and road safety standards and to fund research and programs on traffic safety.
1967: The U.S. Department of Transportation (DOT) is created to oversee transportation issues, including traffic safety (NHTSA is housed within the DOT).
1979: Healthy People – The Surgeon General’s Report on Health Promotion and Disease Prevention is released and is the first call to attention that traffic injury prevention should be part of the country’s public health agenda.
1985: Under the direction of Congress, the National Academy of Sciences releases the report Injury in America which recommends a major national program of research to address injury as a health problem.
1986: Congress creates a center for injury research, surveillance, and education within the Centers for Disease Control and Prevention (CDC), now called the National Center for Injury Prevention and Control.
Table 1. Traffic Injury Prevention Highlights
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 3 5
<
<
Tr a
ff ic
I n
ju ry
P re
v e
n ti
o n
ch. 8
use of seat belts, infant car seats, and motorcycle helmets; implementing safe driving laws; and toughening drunk driving laws. The Spectrum of Prevention (table 3) provides a framework for developing comprehensive approaches to preventing injuries.8
Prioritizing Tra f f ic injur y Prevention for All Modes of Travel
Diversifying transportation options is emerging as a top priority for policymakers. Preventing injuries, improving air quality, encouraging physical activity, and promoting healthier lifestyles can be addressed by reducing miles traveled via automobile and increasing the use of public transportation, bicycling, and walking. This is no easy feat in a country where the car is king and where driving is central to our identity. Advertising campaigns that associate cars with the desire for affluence and independence reinforce the societal link between mobility and upward mobility. The car has historically been promoted as an instrument of sexuality and
power; it’s the guy with the “sexy car” who gets the girl. Driving is a rite of passage that marks our lives nearly from cradle to grave. It is an exuberant transition for a teen when he or she gets a driver’s license and a moment of loss or fear for the adult who must surrender the car keys. Cars will remain the major source of transportation and continue to pose increasing risks unless other safe and convenient forms of transportation are made generally available to the public.
Building transportation systems for all modes of travel promotes equity. Robert Moses, New York City’s storied planner known as the builder of the modern metropolis, reportedly constructed the overpasses on his Long Island parkways too low to accommodate buses as a means of preventing low-income residents of the city— especially blacks and Latinos—from visiting the beaches and parks.9 Thus, parkways like these served as tools for segregation and economic discrimination by putting suburban communities off limits as places of employment and recreation for someone from the inner city who had no car. Decades later, these thoroughfares
Host Agent/Equipment Physical Environment
Social Environment
Pre-Event Drinking Alcohol ignition lock
Alcohol outlets Drinking norms
Event Seat belts and Car seats
Airbags Safety rails Speeding
Post-Event Emergency phones Healthcare access
The Haddon Matrix delineates factors along the timeline of a traffic incident (pre-event through post- event) with four other elements involved in the occurrence of injury (host [e.g., driver], agent [e.g., vehicle], physical environment, and social environment). Prevention activities can be developed within any of these elements. For example, bicycle lanes separate bicyclists from motorized travelers and can thus prevent a crash in the first place. When a crash does occur, if the bicyclist is wearing a helmet, severe head trauma can be prevented. When trauma occurs, a fast and efficient emergency medical system and healthcare must be in place to treat the injuries and prevent death.
Table 2. The Haddon Matrix (with examples)
stand as monuments to transportation policies that divided the country rather than healed its divisions.
Generally, the safety of public transportation and non-motorized travel (i.e., bicycling and walking) has received relatively little federal support, yet communities with diverse transportation options have been shown to have fewer traffic injuries and deaths.10 Contrary to the widespread belief that increased bicycle
and foot traffic will lead to more cyclist and pedestrian injuries and deaths, increasing the numbers of non-motorized travelers may actually make walking and bicycling safer.11 There is also evidence that residents of transit oriented communities have lower per capita traffic fatality rates.12
Germany and the Netherlands illustrate the benefits of government support for safety improvements for pedestrians and bicyclists.
Traffic Injury Prevention H
e a
lt h
y, E
q u
it a
b le
T ra
n s
p o
rt a
ti o
n P
o li
c y
p g
. 1 3 6
>
>
C h
a p
te r
8
Levels of the Spectrum Description
Influencing policy and legislation Developing strategies to change laws and policies to influence outcomes in health, education, and justice
Changing organizational practices Adopting regulations and norms to improve health and safety; creating new models
Fostering coalitions and networks Bringing together groups and individuals for broader goals and greater impact
Educating providers Informing providers who will transmit skills and knowledge to others
Promoting community education Reaching groups of people with information and resources to promote health and safety
Strengthening individual knowledge and skills Enhancing an individual’s ability to prevent injury or illness
Table 3. The Spectrum of Prevention The Spectrum of Prevention* is a tool to guide development of comprehensive strategies that encourage movement beyond the educational or “individual skill-building” approach to address broader environmental and systems-level issues. The Spectrum builds on the Haddon Matrix by providing a method for developing strategies to address traffic safety that are beyond the incident itself and approaches that focus on the individual. The tool has been used across injury fields to integrate individual-oriented efforts with systems change to have the greatest overall effect.
Successful injury prevention strategies have been multifaceted and engaged efforts at multiple levels of the Spectrum of Prevention. In fact, traffic injury prevention has emerged as a model example of prevention.
*The Spectrum of Prevention was originally developed by Larry Cohen in 1983 while working as director of prevention programs at the Contra Costa County Health Department. For application of the Spectrum of Prevention to injury prevention: T. Christoffel and S.S. Gallagher, Injury Prevention and Public Health (Sudbury, MA: Jones and Bartlett Publishers, Inc., 2006).
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 3 7
<
<
Tr a
ff ic
I n
ju ry
P re
v e
n ti
o n
ch. 8
Per mile and per trip walked, Americans are roughly three times more likely to get killed than German pedestrians and more than six times as likely as Dutch pedestrians. Per mile and per trip cycled, Americans are twice as likely to be killed as German cyclists and more than three times as likely as Dutch cyclists.13 Furthermore, pedestrian and bicyclist deaths have declined far more in both countries than in the United States. The Netherlands and Germany have invested heavily in high-quality streetscapes for safe walking and bicycling, making non- motorized travel a norm compared to passenger vehicle travel. The United States has seen virtually the opposite—an interplay of land use, housing, and transportation patterns that have promoted low-density sprawl, high-speed roadways, narrow or no sidewalks, unsafe or no crosswalks, the absence of bicycle lanes, and inaccessible or no public transportation at all. All this makes alternatives to cars and driving not only impractical but also less safe.
With its promise of convenience and freedom, the car still has a strong allure. But a growing number of Americans say they want to drive less and walk, bicycle, and use public transportation more. Advocates can use this desire as momentum to raise public awareness about the benefits of these travel options that are good for better health, for the environment, and for the family budget.
The Continuing Burden of Tra f f ic injuries a nd Deaths
While there have been reductions in death rates per vehicle mile traveled (VMT) over the past four decades, the declines are far less when deaths are measured per capita because Americans drive more than ever (see graph 1).14
In 2007, traffic crashes accounted for 41,059 deaths,15 1,755,247 years of lost life,16 and 2.5 million nonfatal injuries.17 Bicyclists and pedestrians have a disproportionately higher risk
of death in a traffic crash compared to vehicle occupants.18 This greater vulnerability stems from the fact that bicyclists and pedestrians do not have the buffers and protective measures that vehicles offer drivers and passengers. An analysis of 1995 National Household Travel Survey data indicates that the rate of pedestrian fatalities is 36 times higher than car-occupant fatalities per mile traveled, and bicycling fatalities are 11 times higher.19
In 2007, there were 5,504 non-motorized fatalities.20 While walking and bicycling accounted for only 9.5 percent of all trips in 2001, non-motorized fatalities accounted for more than 13 percent of traffic fatalities nationwide.21 Pedestrian fatalities accounted for 84.5 percent of all non-motorized fatalities, bicyclist fatalities accounted for 12.7 percent, and the remaining 2.8 percent were skateboard riders, roller skaters, etc.22
Contrary to the belief that these statistics make a favorable case for continuing to travel exclusively by car, they highlight the lack of infrastructure to support safe non-motorized travel alongside motorized travel. By implementing strategies that reduce the amount of exposure non- motorized travelers have to moving vehicles and reducing the number of cars on the road, it is possible to dually promote alternative modes of transportation and mechanisms to improve the safety of these alternative modes.
Disparities in Traffic Injuries and Deaths
Traffic injuries and deaths are major health concerns for everyone but more so among society’s most vulnerable populations. National data from the Centers for Disease Control and Prevention (CDC) indicate that Native Americans are 1.5 times more likely to die from traffic crashes than other Americans.23 Data collection methods inhibit clarity about the disparate impact of traffic crashes on other racial/ethnic groups, and there is a dearth of data that looks at disparities by income. This is due to the fact
Traffic Injury Prevention H
e a
lt h
y, E
q u
it a
b le
T ra
n s
p o
rt a
ti o
n P
o li
c y
p g
. 1 3 8
>
>
C h
a p
te r
8
that the primary source of data comes from police reports, which do not collect race and ethnicity data. However, some studies seem to indicate the existence of such disparities across race/ethnicity. Between 1990 and 1998, death rates from motor vehicle crashes declined least for African Americans and Native Americans, who also continued to have higher age-adjusted death rates for motor vehicle crashes than any other racial or ethnic group.24 An analysis of North Carolina’s licensed drivers, ages 16 to 24, puts the fatality rate for Latinos at nearly 1.5 times greater than that for whites.25
Pedestrian safety is particularly important for populations that have less access to cars
and rely more on walking for transportation. For example, African Americans make up approximately 12 percent of the U.S. population, but they account for 20 percent of pedestrian deaths.26 Another CDC analysis suggests that the pedestrian fatality rate for Latino men in the Atlanta metropolitan statistical area was six times greater than that for whites between 1994 and 1998.27 While Latinos made up 28 percent of the population in Orange County, CA, they accounted for 40 percent of all pedestrian injuries and 43 percent of pedestrian deaths in 1999, according to a study done by the Los Angeles Times.28
While data comparing traffic injury rates by
0
1
2
3
4
5
6
1960 1965 1970 1975
YEAR
R AT
E
Graph 1: US Traffic Fatalities by VMT and per 10,000 population
8-1
1980
Per 100 Million Vehicle-Miles Traveled
Per 10,000 Population
1985 1990 1995 2000
5.06
5.3
4.74
3.35 3.35
2.47
2.08
1.73 1.582.03
2.46 2.59
2.07
2.25
1.84 1.79
1.59 1.53
Graph 1. U.S. Traffic Fatalities by VMT and Per 10,000 Population
Primary data collected by the Bureau of Transportation Statistics (2000), available at http://www.bts.gov/ publications/nts/index.html. This graph was originally compiled by Todd Litman, Victoria Transport Policy Institute.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 3 9
<
<
Tr a
ff ic
I n
ju ry
P re
v e
n ti
o n
ch. 8
income level are not readily available, people with low incomes may be more vulnerable to traffic injuries and deaths. Low-income often means less access to products that enhance safety, such as newer, safer vehicles or child safety seats; moreover, low-income communities have fewer resources for safe roads and sidewalks, crosswalks, lighting, and traffic enforcement.
Other Populations with Greater risk
Across all ethnic groups, more males than females die from motor vehicle crashes.29 Compared to females, males have lower rates of seat belt use; and are more likely to be involved in alcohol-related crashes and be alcohol- impaired (whether as drivers, passengers, pedestrians, or cyclists) at the time of the incident.30 Drivers under the age of 25 are also more likely to be involved in fatal traffic crashes than any other age group.31
Additionally, driving skills decline with age; with older adults representing the fastest-growing segment of the U.S. population, protecting them from injuries caused by collision should be a top priority on any health and safety agenda. Although older motorists drive fewer miles, they are more likely to be killed or injured in a crash of the same severity compared to other age groups.32 Not only are older drivers typically frailer than others, they also tend to drive older cars, which typically have fewer safety features.33 Even if older drivers in the future drive at the same modest rates as the current elder population, their growing numbers mean that total miles driven by people ages 65 and older would increase 50 percent by 2020 and more than double by 2040.34 While strategies can focus on mitigating risks for older drivers, the best safety approach is to provide safe pedestrian facilities and accessible, affordable public transportation.
Tra nspor tation injur y Prevention Strateg ies
Transportation safety practices and policies should be integrated into all relevant agency agendas and across all levels of government. The pending authorization of the federal transportation bill, the Safe, Accountable, Flexible, Efficient Transportation Equity Act: A Legacy for Users (SAFETEA-LU), is an opportunity to expand programs that have led to improvements in health and safety. Federal policy has historically succeeded in establishing national standards through a carrot-and- stick approach, encouraging state and local governments to comply with federal targets such as those on seat belt use or car seats by dangling federal funds as the carrot. The federal government thus effectively leverages its resources and expands safety targets.
Land Use
Deciding the best uses for our land has not traditionally been included among injury prevention strategies. However, land use issues strongly influence how we travel, which is a key component in determining our risk for getting hurt in a crash. Zoning laws and general plans influence population density within a community, how streets connect, and the distance between homes and key institutions such as schools and workplaces. These factors affect the feasibility, appeal, and safety of walking, bicycling, or using public transportation to get where we need to go. Smart growth strategies—which encourage compact development combining housing, shops, businesses, and parks—reduce our reliance on car travel, creating communities that are safer, more convenient, and more inclusive of low- income residents, older adults, and people with disabilities. One approach that utilizes smart growth elements is transit oriented development (TOD), which develops compact major activity centers around public transportation hubs.
Traffic Injury Prevention H
e a
lt h
y, E
q u
it a
b le
T ra
n s
p o
rt a
ti o
n P
o li
c y
p g
. 1 4 0
>
>
C h
a p
te r
8
By limiting the number of alcohol outlets, zoning laws can also help tackle the problem of impaired driving.
road Design
Road design influences driving behavior and is an important determinant of bicyclists’ and pedestrians’ exposure to traffic, and thus, risk of injury and death.
Road design strategies should emphasize the safety of both motorized and non-motorized travelers. Many road and street improvements can accomplish this: clear road markings and signage to designate crosswalks, bicycle lanes, demarcations between vehicle lanes, and adequate lighting alongside the road to ensure good visibility.35 Additionally, sidewalks, bulb- outs at street corners (which shorten crossing distances and slow the speed of traffic), curb cuts, and separate pathways for pedestrians and bicyclists can limit motor vehicle crashes. Road design strategies should also pay particular attention to improving safe access and mobility for older adults and people with disabilities, beyond Americans with Disabilities Act (ADA) street design requirements.36
Because the risk of death and severe injury in traffic crashes has a direct correlation to speed37 and because speeding is a factor in one-third of all crashes, environmental changes to encourage slower speeds on our roads are vital. Traffic calming, design approaches that acknowledge the relationship between environmental design and behavioral norms, is one of the most important injury prevention strategies in recent decades. Reducing lane widths, curving streets, and adding trees enhance the roadway experience and lead to slower, safer driving. The construction of raised islands, medians, and roundabouts in the roadway also reduces traffic speeds.
These design improvements must reach all neighborhoods. Funds should especially be targeted to low-income communities, where
residents are more likely to walk or bicycle for transportation.
Public Transportation
Safe, efficient, and easily accessible public transportation systems will reduce the frequency of injury and death caused by passenger vehicles and truck traffic. Public transportation systems can solve a number of transportation issues simultaneously, e.g., provide equitable access for vulnerable populations such as older adults, people with disabilities, and low-income populations as well as improve air quality by having fewer vehicles on the road.
Funding should be increased for public transportation improvements and expansions. Public transportation must be fast and affordable; it must link people with the places they need to go. Americans will not give up their cars in significant numbers without realistic public transportation alternatives, including safe routes for walking or bicycling to transit stops. Transit operators can help by providing bicycle lockers and racks, elevators, adequate lighting, and security guards or other safety monitors. Road design features such as crosswalks, sidewalks, and conveniently located transit stops (bus stops and transit lines positioned for easy pedestrian access) are also beneficial. Public transportation accessibility and safety will become increasingly important for older Americans as the U.S. population ages.
ch. 8
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 4 1
<
<
Tr a
ff ic
I n
ju ry
P re
v e
n ti
o n
Speed Limits
As noted earlier, speeding is an important factor in traffic injury and death. The 55-mile- per-hour highway speed limit, established by Congress in 1974 and later adopted by all states, was repealed in 1995. When speed limits are increased on major highways, motorists tend to drive faster on secondary roadways, a process known as “speed adaptation.”38 Reducing speed saves not only lives but also energy because speeding reduces fuel efficiency.
Automobile advertising tends to glorify high- speed driving and risky driving behaviors.39 Getting drivers to slow down may also require changes in automobile marketing practices.
Impaired Driving Laws
Alcohol-related motor vehicle crashes kill someone in the United States every 39 minutes.40 Several studies reveal that when alcohol plays a role, crashes tend to be much more severe.41 Strategies that are effective at preventing impaired driving include:
• Maintain strict enforcement of 0.08 percent blood alcohol content (BAC) laws.42
• Consistently enforce the national minimum legal drinking age law and adopt zero tolerance laws (i.e., revoking a driver’s license if impaired) for drivers younger than 21 in all states.43
• Establish sobriety checkpoints,44 coupled with extensive media campaigns to increase public awareness.
• Install alcohol ignition interlocks in vehicles.45
A number of impaired driving prevention strategies focus on organizational interventions such as alcohol licensing, alcohol availability, alcohol bans, reducing alcohol outlet density and server interventions.46 Other effective
strategies include economic interventions such as raising state and federal alcohol excise taxes and reducing the number of alcohol retailers.47
It must be noted that there are higher densities of alcohol retail in low-income communities and communities of color; consequently, strategies should address the saturation of liquor stores in these communities rather than relying exclusively on modifying consumers’ behavior.48
Driver or pedestrian alcohol use was reported in 47 percent of the traffic crashes that resulted in pedestrian fatalities, with pedestrians more likely to be intoxicated than drivers.49 As rates of driving continue to decline and other modes become more prevalent, specific solutions must be explored for preventing alcohol-related traffic crashes among bicyclists and pedestrians.
Bicycle Helmet Laws
More than a half-million people are treated annually in hospital emergency rooms in the United States for bicycle-related injuries.50 Approximately 60 percent of bicycle deaths involve a head injury; research indicates that a helmet can reduce the risk of head injury by up to 85 percent.51 In 1999, the U.S. Consumer Product Safety Commission issued a mandatory safety standard for bicycle helmets.52 Twenty- one states and the District of Columbia have helmet laws but require use only among young riders (often under the age of 16).53 Little political will exists at the federal and state levels to legislate helmets—despite their lifesaving value—for a greater percentage of bicyclists. Municipal ordinances remain the most promising policy approach.
Schools, businesses, and government agencies can also mandate that children and employees wear bicycle helmets when riding to and from school or work. Schools and offices can disseminate information about their importance and value. Stores that sell bicycles and helmets can also be productive partners in this effort, offering reduced-price or free helmets and
Traffic Injury Prevention H
e a
lt h
y, E
q u
it a
b le
T ra
n s
p o
rt a
ti o
n P
o li
c y
p g
. 1 4 2
>
>
C h
a p
te r
8
distributing information about their proper use and importance in preventing injuries or deaths.
Vehicle Design Standards
Vehicle design standards play a key role in increasing safety for drivers and their passengers and for bicyclists and pedestrians. Examples include improved braking systems, bumpers and external frame requirements, airbags, shatter- resistant windshields, shock-absorbing steering wheels, and automatic seat belts.
Seat Belt Laws
It’s been proven that seat belts save lives. Yet the United States ranks among the lowest nations in the developed world for seat belt usage—an 83 percent daytime use rate.54 Every state except New Hampshire has seat belt use laws, but only 25 states and the District of Columbia allow primary enforcement,55 which permits officers to ticket a driver for not wearing a seat belt without necessitating another traffic violation. Primary enforcement has been associated with lower fatality rates56; in states with such laws, seat belt use is typically 10 percent to 15 percent higher.57 SAFETEA-LU provided more than $500 million in incentive grant money to encourage states to pass primary enforcement seat belt laws, but only a few states have done so. In addition to incentives, federal transportation dollars should be withheld from states that do not adopt such laws. There should also be safeguards for uniform enforcement of primary seat belt laws to address the concern from many opponents that traffic laws have a history of discriminatory enforcement, with targeting of certain racial and ethnic groups.58 The National Organization of Black Law Enforcement Executives, the nation’s leading group of minority law enforcement executives, has recognized that large numbers of African Americans die because they don’t use seat belts or child safety seats (discussed below); it supports primary enforcement laws covering both strategies.
Motorcycle Helmet Laws
Motorcycles make up more than three percent of registered vehicles and only 0.4 percent of vehicle miles traveled but 11 percent of traffic fatalities.59 Helmet use is the most effective measure to protect motorcyclists. Although helmets do not prevent crashes, they offer significant protection against head and brain injuries. States with all-rider helmet laws have a use rate of nearly 100 percent. Twenty-six states have laws that cover only some riders (e.g., up to age 18), which are nearly impossible to enforce; the trend now is toward repealing such laws rather than enacting them. All states should be required to enact an all-rider motorcycle helmet law, and grant funding should provide incentives for promoting motorcyclists’ safety.
Child Safety Seat Laws
Child safety seats reduce the risk of death in vehicles by 71 percent for infants and by 54 percent for children ages one to four years.60 For the past 20 years, child safety seats have been tremendously successful with nearly 100 percent compliance. The CDC Guide to Community Preventive Services presents strong evidence that child safety seat laws, the distribution of safety seats, and education and enforcement
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 4 3
<
<
Tr a
ff ic
I n
ju ry
P re
v e
n ti
o n
ch. 8
campaigns are effective in increasing child safety seat use.61
But more work needs to be done to protect child occupants who remain at heightened risk. The next priority: enacting booster seat laws for children up to age eight, as recommended by the NHTSA. At present, 42 states and the District of Columbia have such laws.62
Lack of access to affordable child safety seats makes their use lower in rural and low-income communities.63 Research reveals, however, that 95 percent of low-income families who own a child safety seat use it.64 The federal surface transportation bill should help low-income families to purchase booster seats.
Graduated Driver Licensing
Graduated driver licensing (GDL) laws, which require newly licensed youth to “graduate” to full licensing, allow young people to practice before assuming the full rights and responsibilities of driving. Research suggests that comprehensive GDL programs can reduce fatal crashes among 16-year-old drivers by up to 38 percent.65
Truck regulations
Although this paper emphasizes safety for passenger vehicles, truck safety is another important area for injury prevention. Strategies include improving built-in truck safety features, regular inspections, restrictions on hours operators can drive without a break, and regulations limiting load size. Federal transportation policy can make roads safer for everyone by supporting expanded rail transport and reducing reliance on trucks.
Cha llenges to a nd Oppor tunities in Tra f f ic injur y Prevention Policy
The current federal transportation bill, SAFETEA- LU, includes programs that advance both health and safety. These programs can benefit greatly from additional funding in the pending authorization of a new bill and an emphasis on expanding best practices and promoting equity. Funding should be prioritized to ensure that injury prevention efforts are designed to benefit the most vulnerable communities. Notably, the Highway Safety Improvement Program (HSIP) was an unprecedented attempt to consolidate safety efforts. Other successes that should be expanded: the Safe Routes to School (SRTS) program, the Transportation Enhancements (TE) program, and The Non-Motorized Transportation Pilot program (see table 4 for details about these programs).
A well-thought-out federal health and safety framework for transportation policy and practice must be reflected at the local level as well. States and locales are the crucibles of change; they do most of the transportation planning and implementation. Yet the quality of safety efforts remains uneven. Without a sufficient federal mandate, some states ignore the imperative for traffic safety, and others have not implemented measures to their greatest potential. Federal mandates should be flexible so locales can choose strategies that best respond to community conditions. HSIP’s mandatory strategic highway safety plan process, which requires states to develop safety priorities and targets in order to receive safety funds from the program, is an opportunity for this type of coordinated traffic safety approach.
The federal government should also require states to include in their transportation planning a wide range of voices, including groups concerned with health and community well-being. An important model for this type of multi-sector collaboration is the Safe
Traffic Injury Prevention H
e a
lt h
y, E
q u
it a
b le
T ra
n s
p o
rt a
ti o
n P
o li
c y
p g
. 1 4 4
>
>
C h
a p
te r
8
Program Amount Description
Highway Safety Improvement Program (HSIP)
$5 billion over 5 years
Achieves a significant reduction in traffic fatalities and serious injuries on all public roads by implementing infrastructure- related highway safety improvements. A portion of these funds can be used for safe behavior enhancement programs.
Safe Routes to School (SRTS)
$612 million over 5 years
Funds infrastructure and programming projects to encourage children and their accompanying guardians to walk or bicycle safely to school every day. This program is one of a few existing models that jointly focuses on increasing rates of walking and bicycling and improving safety conditions for non-motorized travelers. It should be authorized with greater investment.
The Non- Motorized Transportation Pilot Program
$125 million over 5 years
Funds infrastructure and programming in four communities to increase bicycling and walking. Expanding it to fund more communities and conduct further evaluation is the next step. Its authorization should require funded communities to include safety goals in their transportation plans so that every new project focuses on reducing traffic injuries and deaths among bicyclists and pedestrians as well as infrastructure improvements that improve safety for all.
Transportation Enhancements (TE)
$3.5 billion Funds bicycle and pedestrian trails and rail-trail conversions, which include safety improvements to these environments; these conversions take up about 55% of TE funding. It is a 10% set-aside from another major program in SAFETEA-LU, the Surface Transportation Program. This is the largest source of federal funds for non-motorized projects and should be increased to reflect growing demand.
* Funds for agencies under the U.S. Department of Transportation that address traffic safety and for the State and Community Highway Safety Grant Program, described in table 5, were also authorized under SAFETEA-LU.
Table 4. SAFETEA-LU Programs That Support Injury Prevention*
Communities Program, funded through Section 402 transportation funds (described in table 5).
Moreover, the authorization should provide states with data, training, and technical assistance to ensure that plans are well tailored to community needs, that they effectively reach low-income communities and communities of color, and that they include a diverse and comprehensive set of strategies. HSIP currently focuses almost exclusively on the safety of
motorized travelers. To equitably distribute transportation safety funds, several advocates are calling for a “Fair Share for Safety” provision, requiring states to spend a portion of their funds, proportional to the percentage of non-motorized travelers’ deaths, on walking and bicycling safety projects.
A complete streets policy—which emphasizes safe, easy, and efficient mobility for all travelers through connected networks of roads, paths,
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 4 5
<
<
Tr a
ff ic
I n
ju ry
P re
v e
n ti
o n
ch. 8
• Federal: The U.S. Department of Transportation (DOT) is the agency responsible for the federal transportation system. One of its primary charges is to ensure the safety and security of the traveling public; safety is among its top three priorities. The National Highway Traffic Safety Administration (NHTSA), the Federal Highway Administration (FHWA), and the Federal Motor Carrier Safety Administration (FMCSA) are the three major agencies under the DOT umbrella that provide national leadership and support on transportation safety issues. The Federal Transit Administration (FTA) addresses safety related to public transportation. Congress has also created the National Center for Injury Prevention and Control (NCIP) within the Centers for Disease Control and Prevention (CDC); it funds injury research, provides grants to state and local health agencies, and works to increase awareness about injury prevention.
• State: In addition to the federal agencies and programs dedicated to traffic safety, states also have dedicated funding sources to improve traffic safety. This support comes primarily through Section 402 State and Community Highway Safety Grant Program, first authorized by the Highway Safety Act of 1966 and reauthorized in succeeding federal surface transportation bills. Most state public health departments also support ongoing injury prevention and control programs.
Table 5. Federal and State Government Support for Traffic Injury Prevention
and trails—is not included in SAFETEA-LU, but should be incorporated into the new federal transportation bill.66
Another policy issue that requires attention is deciding the appropriate mechanisms to distribute funds in order to encourage projects that promote safety and convenience by modes other than passenger vehicle travel. The new federal transportation bill should provide alternatives to the current funding formula, which bases allocations on a state’s total number of vehicle miles traveled. One option is to link transportation funds to land use patterns that encourage smart growth development and discourage development patterns that require passenger vehicles for the majority of local travel.
Conclusion
Twenty-first century transportation policy must reflect a new vision of mobility and accessibility. Safe travel for all road users and broader considerations of health and equity must be at the center of policy and practice, which would be a difficult task even without the entrenched interests invested in maintaining the status quo. It requires a strong, committed partnership that spans multiple sectors and disciplines.
Building this partnership requires moving beyond past differences and historical positions. Diverse groups must recognize their common interest in opposing policies centered on building more roads, highways, and sprawling developments at the expense of air quality, bicycle and pedestrian access, smart growth, and safety for everyone.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 4 6
>
>
Author Biographies
The Tra nspor tation Prescription: a Summa r y of findings a nd a fra mework for action
&
Chapter 1: Health Effects of Transportation Policy
Judith Bell, M.P.A., is the President of PolicyLink in Oakland, CA, and oversees policy development, strategic planning, program implementation, and policy campaign strategy; she leads projects focused on equitable development, such as the fair distribution of affordable housing, equity in public investment, and community strategies to improve health. Bell is a frequent speaker, trainer, and consultant on advocacy strategy. Her work at PolicyLink includes access to healthy foods, transportation, and infrastructure investment. In addition, Bell leads PolicyLink work with the Convergence Partnership, a multi-foundation initiative to support equity-focused efforts to advance policy and environmental changes for healthy people and healthy places. For more information: http://www.policylink.org/JudithBell.
Larry Cohen, M.S.W., is Founder and Executive Director of Prevention Institute, a nonprofit national center that moves beyond approaches that target individuals to create systematic, comprehensive strategies that alter the conditions that impact community health. With an emphasis on health equity, Cohen has led many successful public health efforts at the local, state, and federal levels on injury and violence prevention, mental health, traffic safety, and food- and physical activity-related chronic disease prevention. Prior to founding Prevention Institute in Oakland, CA, Cohen participated in passing the nation’s first multi-city smoking ban. He established the Food and Nutrition Policy Consortium, which catalyzed the nation’s food labeling law. Cohen also helped shape strategy to secure passage of bicycle and motorcycle helmet laws and to strengthen child and adult passenger restraint laws. For more information: http://www.preventioninstitute.org/larry.html.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 4 7
<
<
A u
th o
r B
io g
ra p
h ie
s
Chapter 2: Transportation Authorization 101: A Backgrounder
Susan Polan, Ph.D., is the Associate Executive Director, Public Affairs and Advocacy, for the American Public Health Association in Washington, DC. In this capacity, Polan oversees APHA’s government relations, communications, membership, affiliate affairs, and component affairs departments. She has more than 15 years of experience in health and public health issues. Her doctorate is in Social Ecology from the University of California, Irvine. For more information: http://www.apha.org/about/ board/aphastaff/biopolan.htm.
Tracy Kolian, M.P.H., is a Senior Policy Analyst in the Public Health Policy Center of the American Public Health Association and is responsible for the association’s environmental public health policy issues and initiatives. She holds a bachelor’s degree in Toxicology from Northeastern University and a master’s degree in environmental health from Tulane University, School of Public Health and Tropical Medicine. For more information: http://www.apha.org/ about/board/aphastaff/biokolian.htm.
Shireen Malekafzali, M.P.H., is a Senior Associate at PolicyLink, a national research and action institute dedicated to social and economic equity. She works across topics to create environmental and policy changes aimed at promoting health and equity. Shireen provides research, technical assistance, training, and policy development support to collaborative efforts intended to create health-enabling environments for all, regardless of race, class, or gender. Her expertise focuses at the intersection of health, equity, and the built environment. For more information: http://www.policylink.org/ ShireenMalekafzali.
Tra nspor tation Choices
Chapter 3: Public Transportation and Health
Todd Litman, M.E.S., is Founder and Executive Director of the Victoria Transport Policy Institute, an independent research organization in Victoria, British Columbia, that is dedicated to developing innovative solutions to transport problems. His work helps expand the range of impacts and options considered in transportation decision making, improve evaluation methods, and make specialized technical concepts accessible to a larger audience. His research is used worldwide in transport planning and policy analysis. For more information: http://www.vtpi.org/documents/ resume.pdf.
Chapter 4: Walking, Bicycling and Health
Susan Handy, Ph.D., is Professor of Environmental Science and Policy and Director of the Sustainable Transportation Center at the University of California, Davis. Her research focuses on the impact of land use on travel behavior, and she is internationally known for her work on the connection between neighborhood design and walking. She is a member of the Committee on Women’s Transportation Issues of the Transportation Research Board and the Institute of Medicine Committee on Childhood Obesity Prevention Actions for Local Governments. For more information: http://www.des.ucdavis.edu/ faculty/handy/.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 4 8
>
>
Author Biographies
Chapter 5: roadways and Health: Making the Case for Collaboration
Catherine L. ross, Ph.D., is Director of Georgia Tech’s Center for Quality Growth and Regional Development (CQGRD). A nationally recognized transportation expert, Ross is the college’s first endowed faculty member—the Harry West Chair for Quality Growth and Regional Development. She has held a variety of important leadership positions at Georgia Tech, including vice provost for academic affairs, associate vice president for academic affairs, co-director of the Transportation Research and Education Center, and director of the College of Architecture’s Ph.D. program. For more information: http://www.cqgrd.gatech.edu/ about/ross.php.
key issues
Chapter 6: Breaking Down Silos: Transportation, Economic Development and Health
Todd Swanstrom, Ph.D., is the E. Desmond Lee Professor of Community Collaboration and Public Policy Administration at the University of Missouri, St. Louis. A co-author of Place Matters: Metropolitics for the Twenty-first Century (University Press of Kansas, 2004), Swanstrom is presently working with the Transportation Equity Network (TEN) on local workforce development in the construction industry. His most recent report, The Road to Good Jobs: Patterns of Employment in the Construction Industry, is available at http:// www.transportationequity.org. He is also doing research, funded by the MacArthur Foundation’s Building Resilient Regions project, on local responses to the foreclosure crisis in six metropolitan areas. For more information: http://pprc.umsl.edu/base_pages/home/staff. htm#research.
Chapter 7: Sustainable Food Systems: Perspectives on Transportation Policy
Kami Pothukuchi, Ph.D., is Associate Professor of Urban Planning at Wayne State University, Detroit, MI. Her research examines the links between food and community and economic development, and the roles public and nonprofit agencies might play to foster these links. A policy guide, “Community and Regional Food Planning Policy Guide,” co- authored by her, was recently adopted by the American Planning Association (http:// www.planning.org/policyguides/food.htm). She serves on the Detroit Food Policy Council Convening Committee, the Detroit Food and Fitness Collaborative, and several other local and national committees related to community food planning. For more information: http://www. clas.wayne.edu/faculty/Pothukuchi.
richard Wallace, M.S., is a Senior Project Manager with the Center for Automotive Research (CAR) in Ann Arbor, MI. He plays the leading role in CAR’s work in the connected vehicle and transportation infrastructure realms. While with the Altarum Institute, he completed and served as Co- Principal Investigator of a study, “Cost Benefit of Providing Non-emergency Medical Transportation.” This groundbreaking study (TCRP B-27), completed under contract to the Transit Cooperative Research Program of the Transportation Research Board of the National Academies, compared the healthcare costs and benefits to the additional transportation costs of providing nonemergency medical transportation to transportation-disadvantaged persons that face transportation barriers to obtaining needed medical care. He holds a master’s degree in Technology and Science Policy from the Georgia Institute of Technology. For more information: http://www.linkedin.com/in/richardwallacecar.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 4 9
<
<
A u
th o
r B
io g
ra p
h ie
s
Chapter 8: Traffic Injury Prevention: A 21st Century Approach
Larry Cohen, M.S.W., is Founder and Executive Director of Prevention Institute. For more details, see his full profile under the “Framing and Summary” section above.
Janani Srikantharajah, B.A., is a Program Coordinator at Prevention Institute, where she supports the Institute’s built environment, transportation, and health reform efforts. Srikantharajah authored the American Public Health Association’s transportation and land use policy resolution in 2008. Prior to joining Prevention Institute, she spent two years with the Ernest Gallo Research Clinic, at UCSF, studying alcohol addiction pathways. For more information: http://www.preventioninstitute. org/staffbio.html.
Leslie Mikkelsen, R.D, M.P.H., is Managing Director of Prevention Institute, where she leads a team focused on environmental and policy approaches to promoting healthy eating and physical activity. Mikkelsen is a policy consultant to the Healthy Eating Active Living Convergence Partnership. She is also Co-founder and Project Director of the Strategic Alliance for Healthy Food and Activity Environments, a California coalition promoting a broad agenda that has influenced state legislation and the Governor’s California Obesity Prevention plan. For more information: http://www.preventioninstitute. org/staffbio.html.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 5 0
>
>
Acknowledgments
This publication is a collaborative effort including the insights and assistance of numerous individuals in addition to the authors.
Our sincerest thanks to the following for their contributions to the development of this report:
• Todd Litman, Manuel Pastor, Carli Paine, Jason Corburn, and Larry Frank for their careful reviews of various portions of the report.
• Fran Smith, for her skillful writing, editing, and research assistance, as well as her valuable input throughout the development of this report.
• Victor rubin, Janani Srikantharajah, and Leslie Mikkelsen for their insightful review and input.
• Paulette Jones robinson, Ariana Zeno, Erika Bernabei, and Emma Sarnat, for their thorough and diligent copyediting, fact-checking, and proofing.
• Annie Finkenbinder and Lili Shoup, for sharing their policy expertise.
• The members of the Convergence Partnership, for their guidance throughout this project:
Linda Jo Doctor, W. K. Kellogg Foundation
David Fukuzawa, Kresge Foundation
Allison S. Gertel-rosenberg and rich Killingsworth, Nemours
Laura Kettel Khan, Centers for Disease Control and Prevention
Angie McGowan and Maisha Simmons, Robert Wood Johnson Foundation
Brian raymond and Loel Solomon, Kaiser Permanente
Marion Standish, The California Endowment
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 5 1
<
<
N o
te s
1 John King, “Bus Route Closing Devastates Disabled Couple,” CNN, March 27, 2009, http://www.cnn.com/2009/POLITICS/03/27/ st.louis.no.bus/.
2 Henry K. Lee, “Diesel Exhaust Poses Health Risks in West Oakland, Study Finds,” San Francisco Chronicle, November 16, 2003, http://www.sfgate.com/cgi-bin/article. cgi?file=/chronicle/archive/2003/11/16/ BAGQE334JL1.DTL.
3 Jennifer Langston, “No Easy Access to Fresh Groceries in Many Parts of Seattle,” Seattle Post Intelligencer, May 1, 2008, http://www. seattlepi.com/local/361235_foodvoid01.html.
4 See http://www.investininfrastructure.org/.
5 President Franklin D. Roosevelt took a similar tack during the Great Depression. Addressing transportation needs accounted for much of the work of the WPA. By 1938, the WPA had paved or repaired 280,000 miles of road and had built 29,000 bridges and 150 airfields, according to Jim Couch, professor of economics and finance at the University of North Alabama and co-author of The Political Economy of the New Deal (Williston, VT: Edward Elgar Publishing, 1998).
6 See the policy platform of the Transportation Equity Network, a national coalition of more than 300 grassroots and partner organizations working to reform transportation and land use policies, http://transportationequity.org/index. php?option=com_content&task=view&id =15&Itemid=32. See also American Public Health Association, At the Intersection of Public Health and Transportation: Promoting Healthy Transportation Policy, 2009, http:// www.apha.org/NR/rdonlyres/43F10382- FB68-4112-8C75-49DCB10F8ECF/0/ TransportationBrief.pdf.
7 National Surface Transportation Policy and Revenue Commission, Transportation for Tomorrow, December 2007, http:// transportationfortomorrow.org/final_report/.
8 E. Burgess and A. Rood, Reinventing Transit: American Communities Finding Smarter, Cleaner, Faster Transportation Solutions (New York: Environmental Defense Fund, 2009), http://www.edf.org/documents/9522_ Reinventing_Transit_FINAL.pdf.
9 M. Turner, “Transit Oriented Development Revitalizes Chicago Neighborhood,” Race, Poverty, and the Environment (Winter 2005/2006), http://www.urbanhabitat.org/ files/24.Marcia.Turner.pdf.
10 See http://www.cleanandsafeports.org. Information and resources on the impacts that transporting goods have on health and community life are available from the Trade, Health, & Environment Impact Project, a community-academic partnership, http:// hydra.usc.edu/scehsc/web/Welcome/ Welcome.html.
11 For information on authorizations and allocations under SAFETEA-LU, the surface transportation bill that expires in September 2009, see http://www.fhwa.dot.gov/ safetealu/factsheets/step.htm.
12 Transportation for Tomorrow (see endnote 7).
13 Health impact assessments are a combination of procedures, methods, and tools to evaluate the potential health effects of a policy, program, or project as well as the distribution of those effects within a population. See http://www.cdc.gov/ healthyplaces/hia.htm.
14 Transit oriented development is a planning and design trend that seeks to create compact, mixed-use, pedestrian-friendly
Notes
The Transportation Prescription: A Summary of Findings and a Framework for Action
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 5 2
>
>
N o
te s
Notes
communities located around new or existing public transportation stations. For more information, see Todd Swanstrom’s chapter in this book. See also http://www.policylink. org/EDTK/TOD/default.html.
15 See http://www.fhwa.dot.gov/safetealu/ factsheets/stp.htm.
16 J. Pucher and J. L. Renne, “Socioeconomics of Urban Travel: Evidence from the 2001 NHTS,” Transport Quarterly 57 (2003): 49– 77, http://policy.rutgers.edu/faculty/pucher/ TQPuchRenne.pdf.
17 See http://www.fta.dot.gov/funding/grants/ grants_financing_3561.html.
18 Transit Riders for Public Transportation, “Ensuring Non-Discrimination in Transportation Investments,” http://www. publicadvocates.org/ourwork/transportation/ docs/TRPT-Ensuring_Non_Discrimination_in_ Transportation_Investments_04-08-09.pdf.
19 Swanstrom T. The Road to Good Jobs: Patterns of Employment in the Construction Industry, (September 30, 2008), http://www.umsl. edu/services/media/assets/pdf/study.pdf.
Chapter 1: Health Effects of Transportation Policy
1 National Surface Transportation Policy and Revenue Commission, Transportation for Tomorrow, December 2007, http:// transportationfortomorrow.org/final_report.
2 P. Latzin et al., “Air Pollution during Pregnancy and Lung Function in Newborns: A Birth Cohort Study,” European Respiratory Journal 33 (2009): 594–603.
3 W. J. Gauderman et al., “The Effect of Air Pollution on Lung Development from 10 to 18 Years of Age,” New England Journal of Medicine 351, no. 11 (2004): 1057–87.
4 C. A. Pope III et al., “Lung Cancer, Cardiopulmonary Mortality, and Long-Term Exposure to Fine Particulate Pollution,” Journal of the American Medical Association (JAMA) 287, no. 9 (2002): 1132–41.
5 American Lung Association, “Highlights of Recent Research on Particulate Air Pollution: Effects of Long-term Exposure,” 2008, http://www.lungusa.org/atf/cf/{7a8d42c2- fcca-4604-8ade-7f5d5e762256}/ANNUAL-
AVERAGE-PM-STUDIES-OCTOBER-2008.PDF.
6 M. Bell et al., “Ozone and Short-Term Mortality in 95 U.S. Urban Communities, 1987–2000,” Journal of the American Medical Association 292, no. 19 (2004): 2372–89, http://research.yale.edu/ environment/bell/research/files/bell_ mortality_jama.pdf.
7 American Lung Association, “Highlights of Recent Research” (see endnote 5).
8 See http://www.arb.ca.gov/research/health/ fs/pm_ozone-fs.pdf.
9 See http://www.lungusa.org/site/ pp.asp?c=dvLUK9O0E&b=44567.
10 Centers for Disease Control and Prevention (CDC), “America Breathing Easier,” http:// www.cdc.gov/asthma/pdfs/breathing_easier_ brochure.pdf.
11 S. Nicholas et al., “Addressing the Childhood Asthma Crisis in Harlem: The Harlem Children’s
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 5 3
<
<
N o
te s
Notes
Zone Asthma Initiative,” American Journal of Public Health 95, no. 2 (2005): 245–49.
12 W. J. Gauderman et al., “Childhood Asthma and Exposure to Traffic and Nitrogen Dioxide,” Epidemiology 16, no. 6 (2005): 737–43.
13 Meredith Minkler et al., “Promoting Healthy Public Policy Through Community-based Participatory Research,” PolicyLink, 2008, http://www.policylink.org/documents/CBPR_ final.pdf. See also http://www.weact.org.
14 K. L. Ebi et al., “U.S. Funding is Insufficient to Address the Human Health Impacts of and Public Health Responses to Climate Variability,” Environmental Health Perspectives Online, February 27, 2009, doi: 10.1289/ehp.0800088, http://dx.doi.org/.
15 See http://thomas.loc.gov/cgi-bin/query/ z?c111:H.R.2323.
16 T. Brikowski, Y. Lotan, and M. S. Pearle, “Climate-related Increase in the Prevalence of Urolithiasis in the United States,” Proceedings of the National Academy of Sciences 105, no. 28 (2008): 9841–46, http://www.pnas.org/ content/105/28/9841.full.pdf+html.
17 Ebi et al., “U.S. Funding” (see endnote 14).
18 CDC, “Preventing Obesity and Chronic Diseases Through Good Nutrition and Physical Activity,” 2008, http://cdc.gov/ nccdphp/publications/factsheets/Prevention/ pdf/obesity.pdf.
19 CDC, “Prevalence of Regular Physical Activity among Adults—United States, 2001 and 2005,” Morbidity and Mortality Weekly Report 56, no. 46 (November 23, 2007): 1209–12, http://www.cdc.gov/mmwr/ preview/mmwrhtml/mm5646a1.h2tm#tab.
20 Transportation Research Board and Institute of Medicine, “Does the Built Environment
Influence Physical Activity? Examining the Evidence,” Special Report 282 (Washington, DC: National Academy Press, 2005).
21 S. J. Olshansky et al., “A Potential Decline in Life Expectancy in the United States in the 21st Century,” New England Journal of Medicine 352, no. 11 (March 17, 2005): 1138–45, http://www.muni.org/iceimages/ healthchp/life%20expectancy1.pdf.
22 L. D. Frank, M. Andresen, and T. L. Schmid, “Obesity Relationships and Community Design, Physical Activity, and Time Spent in Cars,” American Journal of Preventive Medicine 27, no. 2 (2004): 87–96, http:// www.act-trans.ubc.ca/documents/ajpm- aug04.pdf.
23 U. LaChapelle and L. D. Frank, “Transit and Health: Mode of Transport, Employer- Sponsored Public Transit Pass Programs, and Physical Activity,” Journal of Public Health Policy 30 Supplement (2009): S73–S94, http://www.palgrave-journals.com/jphp/ journal/v30/nS1/pdf/jphp200852a.pdf.
24 L. Besser, M. Marcus, and H. Frumkin, “Commute Time and Social Capital in the U.S.,” American Journal of Preventive Medicine 34, no. 3 (2008): 207–11.
25 U.S. Department of Transportation, “Motor Vehicle Traffic Crashes as a Leading Cause of Death in the United States, 2005,” Research Note DOT HS 810 936 (Washington, DC: National Highway Traffic Safety Administration, 2008).
26 Lawrence J. Blincoe et al., “The Economic Impact of Motor Vehicle Crashes, 2000,” Report no. DOT HS-809-446 (Washington, DC: National Highway Traffic Safety Administration, 2002), http://www. nhtsa.dot.gov/staticfiles/DOT/NHTSA/ Communication%20&%20Consumer%20 Information/Articles/Associated%20Files/ EconomicImpact2000.pdf.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 5 4
>
>
N o
te s
Notes
27 CDC, “Web-based Injury Statistics Query and Reporting System (WISQARS),” http://www. cdc.gov/ncipc/WISQARS/.
28 CDC, “Pedestrian Fatalities—Cobb, DeKalb, Fulton, and Gwinnett Counties, Georgia, 1994–1998,” Morbidity and Mortality Weekly Report 48 (1999): 601–05, http://www.cdc. gov/mmwr/PDF/wk/mm4828.pdf.
29 J. Pucher and J. L. Renne, “Socioeconomics of Urban Travel: Evidence from the 2001 NHTS,” Transport Quarterly 57 (2003): 49– 77, http://policy.rutgers.edu/faculty/pucher/ TQPuchRenne.pdf.
30 David A. Morena et al., Older Drivers at a Crossroads (Washington, DC: Federal Highway Administration, 2007), http://www. tfhrc.gov/pubrds/07jan/02.htm.
31 U.S. Department of Transportation, “National Household Travel Survey,” Older Drivers: Safety Implications (Washington, DC: Federal Highway Administration, 2006).
32 Federal Highway Administration, “National Household Travel Survey,” 2001.
33 Fatality Analysis Reporting System Encyclopedia, http://www-fars.nhtsa.dot. gov/Main/index.aspx.
34 T. Litman and S. Fitzroy, “Safe Travels: Evaluating Mobility Management Traffic Safety Benefits,” Victoria Transport Policy Institute, 2006, http://www.vtpi.org/safetrav. pdf.
35 Peter L. Jacobsen, “Safety in Numbers: More Walkers and Bicyclists, Safer Walking and Bicycling,” Injury Prevention 9 (2003): 205– 09, http://www.tsc.berkeley.edu/newsletter/ Spring04/JacobsenPaper.pdf.
36 Steven Raphael and Alan Berube, “Socioeconomic Differences in Household Automobile Ownership Rates: Implications
for Evacuation Policy,” paper prepared for the Berkeley Symposium on “Real Estate, Catastrophic Risk, and Public Policy,” March 23, 2006, http://urbanpolicy.berkeley.edu/ pdf/raphael.pdf.
37 “Overcoming Obstacles to Health,” Robert Wood Johnson Foundation, 2008, http://www.commissiononhealth.org/PDF/ ObstaclesToHealth-Highlights.pdf. See also R. D. Wilkinson and K. E. Pickett, “Income Inequality and Population Health: A Review and Explanation of the Evidence,” Social Science & Medicine 62 (2006): 1768–84.
38 “Transportation Affordability: Strategies to Increase Transportation Affordability,” TDM Encyclopedia, updated July 2008, Victoria Transport Policy Institute, http://vtpi.org/ affordability.pdf.
39 Barbara Lipman, “A Heavy Load: The Combined Housing and Transportation Burdens of Working Families” (Washington, DC: Center for Housing Policy, October 2006), http://www.nhc.org/pdf/pub_heavy_ load_10_06.pdf.
40 “Realizing the Potential: Expanding Housing Opportunities near Transit,” Reconnecting America’s Center for Transit Oriented Development, 2007, http:// www.reconnectingamerica.org/public/ reports?page=2.
41 See http://www.bts.gov/publications/ issue_briefs/number_03/html/transportation_ difficulties_keep_over_half_a_million_ disabled_at_home.html.
42 L. Bailey, “Aging Americans: Stranded Without Options,” Surface Transportation Policy Project, 2004, http://www.apta.com/ research/info/online/documents/aging_ stranded.pdf.
43 Ibid.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 5 5
<
<
N o
te s
Notes
1 Federal Highway Administration, Safe Routes to Schools Program Factsheet, http://www. fhwa.dot.gov/safetealu/factsheets/saferoutes. htm.
2 American Public Health Association, At the Intersection of Public Health and Transportation: Promoting Healthy Transportation Policy, http://www.apha.org/ NR/rdonlyres/43F10382-FB68-4112-8C75- 49DCB10F8ECF/0/TransportationBrief.pdf .
3 Northeast-Midwest Institute, What is the Highway Trust Fund?, http://www.nemw.org/ HWtrustfund.htm.
4 Federal Highway Administration, Surface Transportation Program Factsheet, http:// www.fhwa.dot.gov/safetealu/factsheets/stp. htm.
5 Federal Transit Agency, Large Cities Program
(5307), http://www.fta.dot.gov/funding/ grants/grants_financing_3561.html.
6 Federal Highway Administration, Highway Safety Improvement Program Factsheet, http://www.fhwa.dot.gov/safetealu/ factsheets/hsip.htm.
7 Federal Transit Agency, Jobs and Reverse Commute Program, http://www.fta.dot.gov/ funding/grants/grants_financing_3550.html.
8 Federal Transit Administration, New Starts Factsheet, http://www.fta.dot.gov/planning/ newstarts/planning_environment_2607.html.
9 Surface Transportation Policy Partnership, From the Margins to the Mainstream, http:// www.transact.org/PDFs/margins2006/STPP_ guidebook_margins.pdf.
10 Ibid.
1 U.S. Census Bureau, 2007 American Community Survey 1-Year Estimates, 2007, http://www.census.gov.
2 Transportation Research Board, Does the Built Environment Influence Physical Activity? Examining the Evidence, Special Report 282, Committee on Physical Activity, Health, Transportation, and Land Use, 2005, http:// onlinepubs.trb.org/onlinepubs/sr/sr282.pdf.
3 ICF International, The Broader Connection between Public Transportation, Energy Conservation and Greenhouse Gas Reduction, American Public Transportation Association, 2008, http://www.apta.com/research/info/
online/documents/land_use.pdf; and Todd Litman, Evaluating Public Transit, Benefits and Costs, Victoria Transport Policy Institute (VTPI), 2008, http://www.vtpi.org/tranben. pdf.
4 John E. Evans and Richard H. Pratt, “Travel Response to Transportation System Changes,” in Transit Oriented Development, TCRP Report 95, Transportation Research Board, 2007, http://www.trb.org/TRBNet/ ProjectDisplay.asp?ProjectID=1034.
5 VTPI, “Multi-modal Level-of-service Indicators,” Online TDM Encyclopedia, 2008, http://www.vtpi.org/tdm/tdm129.htm.
Chapter 2: Transportation Authorization 101: A Backgrounder
Chapter 3: Public Transportation and Health
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 5 6
>
>
N o
te s
Notes
6 Federal Highway Administration (FHWA), Rural Transportation Planning, http://www. fhwa.dot.gov/planning/rural/index.html).
7 PATH, The PATH Guide: Planning Ideas, Tools, and Examples to Achieve Transportation Access and Equity in Rural California, prepared by Natural Resources Services, A Division of Redwood Community Action Agency, Eureka, CA, http://www.nrsrcaa.org/ path, 2006, with funding from The Caltrans Environmental Justice Program, http://www. nrsrcaa.org/path/pdfs/PATHGuide5_06.pdf.
8 European Commission, Energy and Transport in Figures, Directorate-General for Energy and Transport, European Commission, 2007, http://ec.europa.eu/dgs/energy_transport/ figures/pocketbook/doc/2007/pb_1_ general_2007.pdf; and FHWA, Highway Statistics, annual reports, http://www.fhwa. dot.gov/policy/ohpi/hss/index.htm.
9 Robert Puentes, The Road . . . Less Traveled: An Analysis of Vehicle Miles Traveled Trends in the U.S. (Washington, DC: Brookings Institution, 2008); and Todd Litman, “Changing Travel Demand: Implications for Transport Planning,” ITE Journal 76, no. 9 ( September 2006): 27–33, http://www.vtpi. org/future.pdf.
10 APTA, Transit Statistics, various years, http:// www.apta.com/research/stats/agency/index. cfm; and FHWA, Highway Statistics, annual reports, http://www.fhwa.dot.gov/policy/ ohpi/hss/index.htm.
11 Reconnecting America, Hidden in Plain Sight: Capturing the Demand for Housing near Transit, Center for Transit-Oriented Development, for the Federal Transit Administration, 2004, http://www. reconnectingamerica.org/public/download/ hipsi.
12 Belden Russonello & Stewart, 2004 American Community Survey: National Survey on
Communities, conducted for the National Association of Realtors and Smart Growth America, October 2004.
13 APTA, Transit Statistics, various years, http:// www.apta.com/research/stats/agency/ index.cfm; FHWA, Highway Statistics, annual reports, http://www.fhwa.dot.gov/policy/ ohpi/hss/index.htm (see endnote 10 for more details on both citations).
14 Todd Litman and Steven Fitzroy, Safe Travels: Evaluating Mobility Management Traffic Safety Benefits, VTPI, 2006, http://www.vtpi. org/safetrav.pdf.
15 Reid Ewing et al., “Relationship between Urban Sprawl and Physical Activity, Obesity, and Morbidity,” American Journal of Health Promotion 18, no. 1 (September/ October 2003): 47–57, http://www. healthpromotionjournal.com and http:// www.smartgrowth.umd.edu/research/ pdf/EwingSchmidKillingsworthEtAl_ SprawlObesity_DateNA.pdf.
16 William H. Lucy, “Mortality Risk Associated with Leaving Home: Recognizing the Relevance of the Built Environment,” American Journal of Public Health 93, no. 9 (September 2003): 1564–69, http://www. ajph.org/cgi/content/full/93/9/1564.
17 Todd Litman, Rail Transit in America: Comprehensive Evaluation of Benefits, VTPI, 2004, http://www.vtpi.org/railben.pdf.
18 Alison Cassady, Tony Dutzik and Emily Figdor, More Highways, More Pollution: Road- building and Air Pollution in America’s Cities, U.S. PIRG Education Fund, 2004, http://www. uspirg.org; and Anming Zhang et al., Towards Estimating the Social and Environmental Costs of Transportation in Canada, Centre for Transportation Studies, University of British Columbia, for Transport Canada, 2005, http://www.sauder.ubc.ca/cts/docs/Full-TC- report-Updated-November05.pdf.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 5 7
<
<
N o
te s
Notes
19 Christopher J. L. Murray et al., Global Burden of Disease and Injury, Center for Population and Development Studies, Harvard School of Public Health, 1996.
20 Peggy Edwards and Agis D. Tsouros, A Healthy City Is an Active City: A Physical Activity Planning Guide, World Health Organization Europe, 2008, http:// www.euro.who.int/InformationSources/ Publications/Catalogue/20081103_1; and U.S. Surgeon General, Physical Activity and Health, CDC, 1999, http://www.cdc.gov/ nccdphp/sgr/sgr.htm.
21 Oscar H. Franco et al., “Effects of Physical Activity on Life Expectancy with Cardiovascular Disease,” Archives of Internal Medicine 165, no. 20 (November 2005): 2355–60, http://archinte.ama-assn.org/cgi/ content/abstract/165/20/2355.
22 World Health Organization, A Physically Active Life Through Everyday Transport: With a Special Focus on Children and Older People and Examples and Approaches from Europe, Regional Office for Europe, 2003, http:// www.euro.who.int/document/e75662.pdf; and Richard Gilbert and Catherine O’Brien, Child- and Youth-Friendly Land-Use and Transport Planning Guidelines, Centre for Sustainable Transportation, 2005, http://cst. uwinnipeg.ca/documents/Guidelines_ON.pdf.
23 Ewing et al. (see endnote 15).
24 Lilah M. Besser and Andrew L. Dannenberg, “Walking to Public Transit: Steps to Help Meet Physical Activity Recommendations,” American Journal of Preventive Medicine 29, no. 4: 2005, http://www.cdc.gov/ healthyplaces/articles/besser_dannenberg. pdf.
25 Richard E. Wener and Gary W. Evans, “A Morning Stroll: Levels of Physical Activity in Car and Mass Transit Commuting,” Environment and Behavior 39, no. 1 (2007):
62–74, http://eab.sagepub.com/cgi/content/ abstract/39/1/62.
26 Ugo Lachapelle and Lawrence D. Frank, “Mode of Transport, Employer-Sponsored Public Transit Pass, and Physical Activity,” Journal of Public Health Policy 30, Suppl. no.1 (2009): S73–S94.
27 Ibid.
28 David Bassett et al., “Walking, Cycling, and Obesity Rates in Europe, North America, and Australia,” Journal of Physical Activity and Health 5, no. 6 (November 2008): 795–814, http://www.humankinetics.com/jpah/ journalAbout.cfm.
29 Ibid.
30 Roland Sturm, Urban Design, Lifestyle, and the Development of Chronic Conditions, the Built Environment and Childhood Obesity Plenary Session, National Institute of Environmental Health Sciences, 2005, http:// www-apps.niehs.nih.gov/conferences/drcpt/ oe2005/speakerdocs/strum-doc.pdf.
31 International Council for Local Environmental Initiatives, Health Benefits Economic Model, Cities for Climate Protection, 2003, http:// www3.iclei.org/ccp-au/tdm/index.html.
32 Todd Litman, Community Cohesion as a Transport Planning Objective, VTPI, 2007, http://www.vtpi.org/cohesion.pdf.
33 Jane Jacobs, Death and Life of the Great American Cities (New York: Random House, 2001).
34 Ontario College of Family Physicians, The Health Impacts of Urban Sprawl Information Series: Volume Four–Social & Mental Health, 2005, http://www.ocfp.on.ca/local/ files/Urban%20Sprawl/UrbanSpraw-Soc- MentalHlth.pdf.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 5 8
>
>
N o
te s
Notes
35 Donald Appleyard, Livable Streets (Berkeley, CA: University of California Press, 1981).
36 Heather Allen, Sit Next to Someone Different Every Day—How Public Transport Contributes to Inclusive Communities, Thredbo Conference, 2008, http://www.thredbo.itls. usyd.edu.au/downloads/thredbo10_papers/ thredbo10-plenary-Allen.pdf.
37 Ethan M. Berke et al., “Protective Association between Neighborhood Walkability and Depression in Older Men,” Journal of the American Geriatrics Society 55, no. 4 (2007): 526–33, http://www.blackwell-synergy.com.
38 Wener and Evans, “A Morning Stroll” (see endnote 25).
39 VTPI, “Basic Accessibility and Mobility,” TDM Encyclopedia, 2008, http://www.vtpi.org/ tdm/tdm103.htm.
40 APTA, The Route to Better Personal Health, 2003, http://spider.apta.com/lgwf/legtools/ better_health.pdf.
41 Irwin Redlener et al., The Growing Health Care Access Crisis for American Children: One in Four at Risk, The Children’s Health Fund, 2006, http://www.childrenshealthfund.org/ calendar/WhitePaper-May2007-FINAL.pdf.
42 Todd Litman, “Transportation Market Distortions” (issue theme “Sustainable Transport in the United States: From Rhetoric to Reality?”), Berkeley Planning Journal 19 (2006): 19–36, http://www-dcrp.ced. berkeley.edu/bpj, and http://www.vtpi.org/ distortions_BPJ.pdf.
43 Todd Litman, Comprehensive Transport Planning Framework: Best Practices for Evaluating All Options and Impacts, VTPI, 2007, http://www.vtpi.org/comprehensive. pdf.
44 Kjartan Sælensminde, “Cost-benefit Analysis of Walking and Cycling Track Networks Taking into Account Insecurity, Health Effects, and External Costs of Motor Vehicle Traffic,” Transportation Research A 38, no. 8 (October 2004): 593–606, http://www.elsevier.com/ locate/tra) at http://www.toi.no/toi_Data/ Attachments/887/sum_567_02.pdf.
45 Robert Puentes and Ryan Prince, Fueling Transportation Finance: A Primer on the Gas Tax (Washington, DC: Brookings Institution, Center on Urban and Metropolitan Policy, 2003), http://www.brookings.edu/reports/ 2003/03transportation_puentes.aspx.
46 Lawrence Frank, Sarah Kavage, and Todd Litman, Promoting Public Health Through Smart Growth: Building Healthier Communities Through Transportation and Land Use Policies, Smart Growth BC, 2006, http://www.vtpi.org/sgbc_health. pdf; and Richard Killingsworth, Audrey De Nazelle, and Richard Bell, “Building a New Paradigm: Improving Public Health Through Transportation,” ITE Journal 73, no. 6 (June 2003): 28–32, http://www.ite.org.
47 World Health Organization (WHO), World Report on Road Traffic Injury Prevention, WHO and World Bank, 2004, http://www. who.int/entity/world-health-day/2004/ infomaterials/world_report; and International City/County Management Association (ICMA), Creating a Regulatory Blueprint for Healthy Community Design: A Local Government Guide to Reforming Zoning and Land Development Codes, ICMA (http:// www.icma.org) and Active Living By Design (http://www.activelivingleadership.org), 2005.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 5 9
<
<
N o
te s
Notes
48 Dan Emerine and Eric Feldman, Active Living and Social Equity: Creating Healthy Communities for All Residents, International City/County Management Association, 2005, http://bookstore.icma.org.
49 Active Living Research (http://www. activelivingresearch.org).
50 VTPI, “Financing Options,” Online TDM
Encyclopedia, 2008, http://www.vtpi.org/ tdm/tdm119.htm.
51 Center for Neighborhood Technology, Housing + Transportation Affordability Index, 2008, http://htaindex.cnt.org.
52 Congress for the New Urbanism, Parking Requirements and Affordable Housing, 2008, http://www.cnu.org/node/2241.
1 J. Pucher and R. Buehler, “Making Cycling Irresistible: Lessons from the Netherlands, Denmark, and Germany,” Transport Reviews 28 (2008): 495–528.
2 David Bassett et al., “Walking, Cycling, and Obesity Rates in Europe, North America, and Australia,” Journal of Physical Activity and Health 5, no. 6 (November 2008): 795–814, http://www.humankinetics.com/jpah/ journalAbout.cfm.
3 Ibid.
4 J. Pucher, J. Dill, and S. Handy et al., “Infrastructure, Programs, and Policies to Increase Bicycling: An International Review,” Preventative Medicine. Vol 48, No. 2, February 2010.
5 U.S. Department of Health and Human Services, “2008 Physical Activity Guidelines for Americans,” http://www.health.gov/ PAGuidelines/pdf/paguide.pdf (accessed March 27, 2009).
6 Centers for Disease Control and Prevention (CDC), “Prevalence of Regular Physical Activity among Adults – United States, 2001 and 2005,” Morbidity and Mortality Weekly Report 56 (2007): 1209–12.
7 J. N. Morris and A. E. Hardman, “Walking to Health,” Sports Medicine 23 (1997): 306–32.
8 D. Ogilvie et al., “Interventions to Promote Walking: Systematic Review,” British Medical Journal 334 (June 2007): 1204.
9 National Highway Traffic Safety Administration (NHTSA), “Traffic Safety Facts 2007 Data: Pedestrians,” 2008, http:// www.nhtsa.dot.gov/portal/nhtsa_static_ file_downloader.jsp?file=/staticfiles/DOT/ NHTSA/NCSA/Content/TSF/2007/810994. pdf (accessed March 27, 2009); and NHTSA, “Traffic Safety Facts 2007 Data: Bicyclists and Other Cyclists,” 2008, www.nhtsa.dot. gov/portal/nhtsa_static_file_downloader. jsp?file=/staticfiles/DOT/NHTSA/NCSA/ Content/TSF/2007/810986.pdf (accessed March 27, 2009).
10 C. Gidelow et al., “A Systematic Review of the Relationship between Socio-economic Position and Physical Activity,” Health Education Journal 65 (2007): 338–67; and CDC, “Prevalence of Regular Physical Activity among Adults – United States, 2001 and 2005,” Morbidity and Mortality Weekly Report 56 (2007): 1209–12.
11 L. M. Besser and A. L. Dannenberg, “Walking
Chapter 4: Walking, Bicycling, and Health
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 6 0
>
>
N o
te s
Notes
to Transit: Steps to Help Meet Physical Activity Recommendations,” American Journal of Preventive Medicine 29 (2005): 273–80.
12 J. Pucher and J. L. Renne, “Socioeconomics of Urban Travel: Evidence from the 2001 NHTSA”; and Besser and Dannenberg, “Walking to Transit” (see endnote 11).
13 N. C. McDonald, “Exploratory Analysis of Children’s Travel Patterns,” Transportation Research Record 1977 (2006): 1–7.
14 NHTSA, “Traffic Safety Facts 2007 Data: Pedestrians” (see endnote 9, citation 1); and NHTSA, “Traffic Safety Facts 2007 Data: Bicyclists and Other Cyclists” (see endnote 9, citation 2).
15 L. Bailey, “Aging Americans: Stranded without Options,” 2004, http://www. transact.org/library/reports_html/seniors/ aging_exec_summ.pdf (accessed March 27, 2009).
16 NHTSA, “Traffic Safety Facts 2007 Data: Pedestrians” (see endnote 9, citation 1).
17 U.S. Department of Health and Human Services, “2008 Physical Activity Guidelines for Americans” (see endnote 5).
18 J. Pucher and R. Buehler, “Making Cycling Irresistible: Lessons from the Netherlands, Denmark, and Germany,” Transport Reviews 28 (2008): 495–528. J. Pucher and L. Dijkstra, “Promoting Safe Walking and Cycling to Improve Public Health: Lessons from the Netherlands and Germany,” American Journal of Public Health 93 (2003): 1509–16.
19 P. L. Jacobsen, “Safety in Numbers: More Walkers and Bicyclists, Safer Walking and Bicycling,” Injury Prevention 9 (2003): 205– 09.
20 Bureau of Transportation Statistics, “Transportation Statistics Annual Report,” 2007, http://www.bts.gov/publications/ transportation_statistics_annual_ report/2007/pdf/entire.pdf (accessed May 7, 2009).
21 S. Handy and K. Clifton, “Local Shopping as a Strategy for Reducing Automobile Travel,” Transportation 28 (2001): 317–46.
22 Pucher and Dijkstra, “Promoting Safe Walking and Cycling” (see endnote 18).
23 R. Buehler, “Transport Policies, Travel Behavior and Sustainability: A Comparison of Germany and the U.S.” 2008, unpublished dissertation.
24 N. C. McDonald, “Active Transportation to School: Trends among U.S. Schoolchildren, 1969–2001,” American Journal of Preventive Medicine 32 (2007): 509–16.
25 G. Tal and S. Handy, “Children’s Biking for Non-school Purposes: Getting to Soccer Games in Davis, CA,” Transportation Research Record 2074 (2008): 40–45.
26 E. Gaona, “Oxnard Plan Focuses on Bicycle Commuters,” Los Angeles Times, August 19, 2002, B-3.
27 R. L. Knoblauch et al., “The Pedestrian and Bicyclist Highway Safety Problem as It Relates to the Hispanic Population in the United States,” 2004, http://safety.fhwa.dot. gov/ped_bike/docs/03p00324/050329.pdf (accessed March 27, 2009).
28 S. Handy, “Regional Transportation Planning in the U.S.: An Examination of Changes in Technical Aspects of the Planning Process in Response to Changing Goals,” Transport Policy 15 (2008): 113–26.
29 K. Krizek et al., “Explaining Changes in Walking and Bicycling Behavior:
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 6 1
<
<
N o
te s
Notes
The Transportation Researcher’s Full Employment Act,” Environment and Planning (forthcoming).
30 NHTSA, “Traffic Safety Facts 2007 Data: Pedestrians” (see endnote 9); and NHTSA, “Traffic Safety Facts 2007 Data: Bicyclists and Other Cyclists” (see endnote 9).
31 S. Handy et al., “The Regional Response to Federal Funding for Bicycle and Pedestrian Projects,” 2009, Institute of Transportation Studies, University of California – Davis, working paper.
32 See, for example, http://www.walkinginfo.org, and http://www.bicycleinfo.org.
33 Handy et al., “The Regional Response to Federal Funding” (see endnote 31).
34 Ibid.
35 Ibid.
36 Land use planning powers have not explicitly been taken by the federal government and so are left to states, according to the reserved powers doctrine of the U.S. Constitution; most states have chosen to delegate this power to local governments, with some variation in the degree to which states have chosen to exert influence over local planning.
37 See http://www.activelivingbydesign.org.
38 S. Handy et al., “Is support for traditionally designed communities growing? Evidence from two national surveys,” Journal of the American Planning Association 74 (2008): 209–21.
1 Federal Highway Administration (FHWA), “National Household Travel Survey” (NHTS), Online Analysis Tool, 2001, http://nhts.ornl. gov/tables/ae/TableDesigner.aspx (accessed March 10, 2009).
2 Fatality Analysis Reporting System Encyclopedia, n.d., http://www-fars.nhtsa. dot.gov/Main/index.aspx (accessed October 7, 2008).
3 Centers for Disease Control and Prevention. Web-based Injury Statistics Query and Reporting System (WISQARS), http://www. cdc.gov/injury/wisqars/index.html (accessed June 16, 2009).
4 Task Force on Community Preventive Services, “Motor-Vehicle Occupant Injury: Strategies for Increasing Use of Child Safety Seats, Increasing Use of Safety Belts, and Reducing Alcohol-Impaired Driving,” Morbidity and
Mortality Weekly Report 50 (RR07) (2001): 1–13.
5 U.S. Department of Transportation (DOT), HS 811 017, “A Brief Statistical Summary August 2008 Traffic Safety Facts – Crash Stats: 2007 Traffic Safety Annual Assessment – Highlights.”
6 FHWA, “National Household Travel Survey,” Online Analysis Tool, 2001, http://nhts.ornl. gov/tables/ae/TableDesigner.aspx (accessed March 11, 2009).
7 FHWA, “Making the Case for Transportation Safety – Ideas for Decision Makers,” FHWA- HEP-08-017, 2008.
8 H. G. Garrison and C. E. Crump, “Commentary: Race, Ethnicity and Motor Vehicle Crashes,” Annals of Emergency Medicine 49, no. 2 (2007): 219–20.
Chapter 5: Roadways and Health: Making the Case for Collaboration
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 6 2
>
>
N o
te s
Notes
9 J. Pucher and J. Renne, “Socioeconomics of Urban Travel: Evidence from the 2001 NHTS,” Transportation Quarterly 57, no. 3 (Summer 2003).
10 L. J. Paulozzi, “Is It Safe to Walk in the Sunbelt? Geographic Variation among Pedestrian Fatalities in the United States, 1999–2003,” Journal of Safety Research 37, no. 5 (2006): 453–59.
11 F. Cheung et al., “An Analysis of Alcohol- related Motor Vehicle Fatalities by Ethnicity,” Annals of Emergency Medicine 34, no. 4, part 1 (1999): 550–53.
12 University of New South Wales, “A Virtuous Cycle: Safety in Numbers for Bicycle Riders,” Science Daily (September 7, 2008).
13 P. L. Jacobsen, “Safety in Numbers: More Walkers and Bicyclists, Safer Walking and Bicycling,” Injury Prevention 9, no. 3 (2003): 205–09.
14 FHWA, “National Household Travel Survey,” Online Analysis Tool, http://nhts.ornl.gov/ tables/ae/TableDesigner.aspx (accessed March 10, 2009).
15 Fatality Analysis Reporting System Encyclopedia, n.d., http://www-fars.nhtsa. dot.gov/Main/index.aspx (accessed October 7, 2008).
16 M. Zhu et al., “Urban and Rural Variation in Walking Patterns and Pedestrian Crashes,” Injury Prevention 14, no. 6 (2008): 377–80.
17 Todd Litman, Smart Transportation Investments: Reevaluating the Role of Highway Expansion for Improving Urban Transportation (Victoria, BC: Victoria Transport Policy Institute, 2006).
18 M. Abdel-Aty and J. Keller, “Exploring the Overall and Specific Crash Severity Levels at Signalized Intersections,” Accident Analysis & Prevention 37, no. 3 (2005): 417–25.
19 P. Liu, J. J. Lu, and H. Chen, “Safety Effects of the Separation Distances between Driveway Exits and Downstream U-turn Locations,” Accident Analysis & Prevention 40, no. 2 (2008): 760–67.
20 Pedestrian and Bicycle Information Center (PBIC), and Federal Highway Administration (FHWA). PEDSAFE Pedestrian Safety Guide and Countermeasure Selection System: Countermeasures. Federal Highway Administration, U.S. Department of Transportation. (2002); and J. Miller, Impact of Situational Factors on Survey Measured Fear of Crime. International Journal of Social Research Methodology, 11(4). (2008).
21 U. Shankar, Pedestrian Roadway Fatalities, No. DOT HS 809 456, Mathematical Analysis Division, National Center for Statistics and Analysis; National Highway Traffic Safety Administration, U.S. Department of Transportation, 2003.
22 SRTS Guide: Reduced Corner Radii, n.d., http://www.saferoutesinfo.org/guide/ engineering/reduced_corner_radii.cfm (accessed October 20, 2008); and PEDSAFE Pedestrian Safety Guide and Countermeasure Selection System: Curb Extensions, n.d., http://www.walkinginfo.org/pedsafe/ pedsafe_curb1.cfm?CM_NUM=19 (accessed October 20, 2008).
23 M. Ernst, Mean Streets 2004: How Far Have We Come?, Surface Transportation Policy Project, 2004.
24 J.-H. Mok, H. C. Landphair, and J. R. Nader, “Landscape Improvement Impacts on Roadside Safety in Texas,” Landscape and Urban Planning 78, no. 3 (2006): 263–74; K. Dixon and K. Wolf, “Benefits and Risks of Urban Roadside Landscape: Finding a Livable, Balanced Response,” Presentation at the 3rd Urban Street Symposium, Seattle, WA, 2007; and K. L. Wolf and N. Bratton, “Urban Trees and Traffic Safety: Considering U.S. Roadside Policy and Crash Data,” Arboriculture and Urban Forestry 32, no. 4 (2006): 170–79.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 6 3
<
<
N o
te s
Notes
25 J. M. Sullivan and M. J. Flanagan, “Determining the Potential Safety Benefit of Improved Lighting in Three Pedestrian Crash Scenarios,” Accident Analysis & Prevention 39, no. 3 (2007): 638–47.
26 Centers for Disease Control and Prevention (CDC), FastStats – Asthma, n.d., http://www. cdc.gov/nchs/fastats/asthma.htm (accessed March 11, 2009).
27 S. Pandian, S. Gokhale, and A. K. Ghoshal, “Evaluating Effects of Traffic and Vehicle Characteristics on Vehicular Emissions near Traffic Intersections,” Transportation Research Part D: Transport and Environment 14, no. 3 (2009): 180–96; and J. Lin and Y. E. Ge, “Impacts of Traffic Heterogeneity on Roadside Air Pollution Concentration,” Transportation Research Part D: Transport and Environment 11, no. 2 (2006): 166–70.
28 Environmental Protection Agency (EPA), The Plain English Guide to the Clean Air Act: Cleaning Up Commonly Found Air Pollutants, 2006, http://www.epa.gov/air/peg/cleanup. html.
29 W. J. Gauderman et al., “Association between Air Pollution and Lung Function Growth in Southern California Children,” American Journal of Respiratory and Critical Care Medicine 162, no. 4 (2000): 1383–90; and W. J. Gauderman et al., “The Effect of Air Pollution on Lung Development from 10 to 18 Years of Age,” New England Journal of Medicine 351, no. 11 (2004): 1057–67.
30 A. G. Barnett et al., “The Effects of Air Pollution on Hospitalizations for Cardiovascular Disease in Elderly People in Australian and New Zealand Cities,” Environmental Health Perspectives 114, no. 7 (2006); and T. F. Mar et al., “Fine Particulate Air Pollution and Cardiorespiratory Effects in the Elderly,” Epidemiology 16, no. 5 (2005): 681–87.
31 L. K. Baxter et al., “Predicting Residential Indoor Concentrations of Nitrogen
Dioxide, Fine Particulate Matter, and Elemental Carbon Using Questionnaire and Geographic Information System Based Data,” Atmospheric Environment 41, no. 31 (2007): 6561–71.
32 J. McCreanor et al., “Respiratory Effects of Exposure to Diesel Traffic in Persons with Asthma,” New England Journal of Medicine 357, no. 23 (2007): 2348–58.
33 National Center for Chronic Disease Prevention and Health Promotion. Obesity: Halting the Epidemic by Making Health Easier - At A Glance 2009. Centers for Disease Control and Prevention (CDC). (2009).
34 D. Chenoweth and J. Leutzinger, “The Economic Cost of Physical Inactivity and Excess Weight in American Adults,” Journal of Physical Activity & Health 3, no. 2 (2006): 148–63.
35 R. D. Putnam, Bowling Alone: The Collapse and Revival of American Community (New York: Simon & Schuster, 2000).
36 J. Gehl, Life Between Buildings (New York: Van Nostrand Reinhold, 1987).
37 L. E. Jackson, “The Relationship of Urban Design to Human Health and Condition,” Landscape and Urban Planning 64, no. 4 (2003): 191–200.
38 Surface Transportation Policy Project and Center for Neighborhood Technology, Driven to Spend: The Impact of Sprawl on Household Transportation Expenses, 2000.
39 P. Gordon-Larsen et al., “Inequality in the Built Environment Underlies Key Health Disparities in Physical Activity,” Pediatrics 117 (2006): 417–24; S. L. Huston et al., “Neighborhood Environment, Access to Places for Activity, and Leisure-time Physical Activity in a Diverse North Carolina Population,” American Journal of Health Promotion 18, no. 1 (2003): 58–69; S. Parks, R. Houseman, and R. Brownson,
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 6 4
>
>
N o
te s
Notes
“Differential Correlates of Physical Activity in Urban and Rural Adults of Various Socioeconomic Backgrounds in the United States," Journal of Epidemiology and Community Health 57 (2003): 29–35; W. Taylor et al., “Environmental Justice: Obesity, Physical Activity, and Healthy Eating,” Journal of Physical Activity and Health 3, Suppl. 1 (2006): S30–S54; and D. Wilson et al., “Socioeconomic Status and Perceptions of Access and Safety for Physical Activity,” Annals of Behavioral Medicine 28 (2004): 20–28.
40 Surface Transportation Policy Project, “Transportation and Social Equity,” n.d., http://www.transact.org/library/factsheets/ equity.asp.
41 L. D. Frank, M. A. Andresen, and T. L. Schmid, “Obesity Relationships with Community Design, Physical Activity, and Time Spent in Cars,” American Journal of Preventive Medicine 27, no. 2 (2004): 87–96.
42 A. V. Moudon, Effects of Site Design on Pedestrian Travel in Mixed-Use, Medium- Density Environments (Seattle, WA: Washington State Transportation Center (TRAC, 1997); and H. C. Borst et al., “Relationships between Street Characteristics and Perceived Attractiveness for Walking Reported by Elderly People,” Journal of Environmental Psychology, in press, corrected proof.
43 C. L. Addy, “Associations of Perceived Social and Physical Environmental Supports with Physical Activity and Walking Behavior,” American Journal of Public Health 94, no. 3 (2004): 440–43.
44 H. M. Badland, G. M. Schofield, and N. Garrett, “Travel Behavior and Objectively Measured Urban Design Variables: Associations for Adults Traveling to Work,” Health & Place 14, no. 1 (2008): 85–95; and A. W. Agrawal, M. Schlossberg, and K. Irvin, “How Far, by Which Route and Why? A Spatial Analysis of Pedestrian Preference,” Journal of Urban Design 13, no. 1 (2008): 81–98.
45 National Surface Transportation Policy and Revenue Study Commission, Final Report: Transportation for Tomorrow, http://www. transportationfortomorrow.org/final_report/.
46 R. Puentes, A Bridge to Somewhere: Rethinking American Transportation for the 21st Century. Metropolitan Infrastructure Initiative, Number 3: The Brookings Institution. (2006).
47 National Surface Transportation Policy and Revenue Study Commission, Final Report: Transportation for Tomorrow, http://www. transportationfortomorrow.org/final_report/.
48 Ibid.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 6 5
<
<
N o
te s
Notes
1 Amartya Sen, Inequality Reexamined (Cambridge, MA: Harvard University Press, 1992), 39.
2 Research has shown, for example, that lacking control over one’s work is associated, after controlling for a range of variables, with cardiovascular symptoms and other health problems. This research is summarized in Richard G. Wilkinson, Unhealthy Societies: The Afflictions of Inequality (New York: Routledge, 1996), 181. In general, living in concentrated-poverty neighborhoods is associated with a sense of fatalism, the belief that nothing can be done to improve the situation, which leads to the internalization of stress that has shown to be highly correlated with poor health outcomes. For the connection between concentrated poverty and fatalism, see James E. Rosenbaum, Lisa Reynolds, and Stefanie DeLuca, “How Do Places Matter? The Geography of Opportunity, Self-Efficacy, and a Look inside the Black Box of Residential Mobility,” Housing Studies 17, no. 1 (2002): 71–82. For a nontechnical synthesis of the research on the connection between stress and disease, see Grace Budrys, Unequal Health: How Inequality Contributes to Health of Illness (Lanham, MD: Rowman & Littlefield, 2003), ch. 9.
3 Transportation needs also vary with age, gender, and disabilities. A full treatment of transportation equity, which is beyond the scope of this essay, would need to take into account these conditions as well. See “Equity Evaluation: Perspectives and Methods for Evaluating the Equity Impacts of Transportation,” TDM Encyclopedia (updated July 23, 2008), http://www.vtpi.org/tdm/ tdm13.htm.
4 The literature on the health effects of urban sprawl is voluminous. For a short overview, see Robert Burchell et al, Sprawl Costs: Economic Impacts of Unchecked Development (Washington, DC: Island Press, 2005). See also Howard Frumkin, Lawrence Frank, and Richard Jackson, Urban Sprawl and Public Health: Designing, Planning, and Building for Healthier Communities (Washington, DC: Island Press, 2004).
5 For evidence of the negative impact of inequality in general and geographical inequalities in particular on health, see Ichiro Kawachi, Bruce P. Kennedy, and Richard G. Wilkinson, eds., The Society and Population Health Reader: Income Inequality and Health (New York: New Press, 1999).
6 The following account of the health effects of concentrated poverty is based on Peter Dreier, John Mollenkopf, and Todd Swanstrom, Place Matters: Metropolitics for the Twenty-First Century, rev. ed. (Lawrence, KS: University Press of Kansas, 2004), 76–82.
7 For an insightful discussion of policy monopolies (also called subgovernments, iron triangles, or policy silos), see Frank R. Baumgartner and Bryan D. Jones, Agendas and Instability in American Politics (Chicago: University of Chicago Press, 1993), 6–9. For analysis of the highway policy silo in its heyday, see Alan Altschuler, The City Planning Process: A Political Analysis (Ithaca, NY: Cornell University Press, 1965) and John Mollenkopf, The Contested City (Princeton, NJ: Princeton University Press, 1983).
8 According to a survey of urban scholars, the 41,000-mile federal interstate highway program was the most important influence
Chapter 6: Breaking Down Policy Silos: Transportation, Economic Development, and Health
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 6 6
>
>
N o
te s
Notes
shaping America’s metropolitan areas in the past half-century. Reported in Robert Fishman, “The American Metropolis at Century’s End: Past and Future Influences,” Housing Policy Debate 11, no. 1 (2000): 199–213.
9 Over time, however, the urbanization of the suburbs and the almost complete reliance on the automobile has generated serious health costs, including higher levels of air pollution and low activity levels related to obesity. The increased stress from long commutes and traffic congestion negatively affect health. According to a study of eight metropolitan areas, even though rates of homicide by strangers are higher in inner urban areas than in outlying suburbs, the higher traffic fatality rates in outlying areas swamp this effect, making outlying areas less safe than central cities and inner suburbs. See William H. Lucy, “Mortality Risk Associated with Leaving Home: Recognize the Significance of the Built Environment,” American Journal of Public Health 93, no. 9 (September 2003): 1564–69.
10 Elizabeth Kneebone, “Job Sprawl Revisited: The Changing Geography of Metropolitan Employment,” The Brookings Institution, Center on Urban and Metropolitan Policy, May 2009.
11 According to Ingrid Gould Ellen and Margery Austin Turner, of six literature reviews on the spatial mismatch, three find substantial support for it, two find moderate support, and one finds the evidence too mixed to reach a conclusion. See “Do Neighborhoods Matter and Why?,” in Choosing a Better Life: Evaluating the Moving to Opportunity Social Experiment, eds. John Goering and Judith D. Feins (Washington, DC: Urban Institute Press, 2003), 328.
12 Between 1950 and 2000 the average size of a new home increased by more than 50 percent (from 1,470 square feet to 2,265 square feet). In 2000 the average new house
was almost two-thirds more expensive than in 1960 (in constant dollars), and the share of new housing purchased by the top 20 percent of the income range increased dramatically. Rachel Dwyer, “Expanding Homes and Increasing Inequalities: U.S. Housing Development and the Residential Segregation of the Affluent,” Social Problems 54, no. 1 (2007): 23–46.
13 Paul Jargowsky, “Sprawl, Concentration of Poverty, and Urban Inequality,” in Urban Sprawl: Causes, Consequences, and Policy Responses, ed. Gregory D. Squires (Washington, DC: Urban Institute Press, 2002), 57.
14 Between 1956 and 1972, highway building and urban renewal displaced an estimated 3.8 million persons, overwhelmingly poor and minorities, from their homes. Susan Fainstein and Norman Fainstein, eds., Restructuring the City: The Political Economy of Urban Development (New York: Longman, 1986), 49.
15 In Bowling Alone Robert Putnam reports that joining your first group will “cut your risk of dying over the first year in half.” Bowling Alone: The Collapse and Revival of American Community (New York: Simon and Schuster, 2000), 331. For a comprehensive analysis of the costs of displacement on African American communities by urban renewal (and highway building), see Mindy Thompson Fullilove, Root Shock: How Tearing Up City Neighborhoods Hurts America (New York: Ballantine, One World, 2004).
16 To this day, African Americans are underrepresented in the construction workforce relative to their participation in the overall workforce. See Todd Swanstrom, The Road to Good Jobs: Patterns of Employment in the Construction Industry in the Top Twenty-five Metropolitan Areas (St. Louis, MO: Transportation Equity Network, Public Policy Research Center, University of Missouri – St. Louis, 2008).
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 6 7
<
<
N o
te s
Notes
17 “Transportation Affordability: Strategies to Increase Transportation Affordability,” TDM Encyclopedia, updated July 2008, Victoria Transport Policy Institute, http://www.vtpi. org/tdm/tdm106/htm.
18 Transportation expenditures are not regressive with respect to family expenditures because many low-income households, such as older adults, live on savings and therefore spend more than they earn. However, for all vehicle-owning households, transportation expenditures are regressive as a proportion of household expenditures. See Todd Litman, “Transportation Affordability: Evaluation and Improvement Strategies,” Victoria Transport Policy Institute, November 10, 2008, http:// www.vtpi.org/affordability.pdf.
19 Barbara Lipman, A Heavy Load: The Combined Housing and Transportation Burdens of Working Families (Washington, DC: Center for Housing Policy, October 2006), http://www.nhc.org/pdf/pub_heavy_ load_10_06.pdf.
20 American Automobile Association, Your Driving Costs (Heathrow, FL: AAA, 2007). The estimate is based on gasoline costing $2.256 a gallon. Low-income persons can own a car for less by purchasing a cheap used car, but then they are subject to repairs, and unreliable transportation can cost them their job. Also, insurance costs tend to be higher in poor communities.
21 Reauthorizations of ISTEA in 1998 and 2005 strengthened the law, for example, creating incentives to link transportation and land use (Transportation and Community and System Preservation [TCSP] Pilot Program) and funding reverse commuting programs to transport inner-city workers to suburban jobs (Job Access and Reverse Commute Program [JARC]).
22 For a comprehensive and largely critical review of federal transportation policy, see Bruce Katz, Robert Puentes, and Scott Bernstein, “Getting Transportation Right
for Metropolitan America,” in Taking the High Road: A Metropolitan Agenda for Transportation Reform, eds. Bruce Katz and Robert Puentes (Washington, DC: Brookings Institution Press, 2005), 15–42.
23 John Pucher, “Public Transportation,” in The Geography of Urban Transportation, 3rd ed., eds. Susan Hanson and Genevieve Giuliano (New York: Guilford Press, 2004), 207.
24 http://www.apta.com/media/ releases/081208_ridership_surges.cfm.
25 Margaret Weir, Jane Rongerude, and Christopher K. Ansell, “Collaboration is Not Enough: Virtuous Cycles of Reform in Transportation Policy,” Urban Affairs Review 44, no. 4 (March 2009): 455–89.
26 Katz, Puentes, and Bernstein (see endnote 22).
27 For example, to demonstrate that the public had been consulted, the Chicago MPO (CATS) produced a 15-pound compilation of public comments that had never been analyzed. Weir, Rongerude, and Ansell, “Collaboration is Not Enough,” 476 (see endnote 25).
28 See Robert Cervero, “Effects of Light Rail and Commuter Rail Transit on Land Prices: Experiences in San Diego County,” Journal of the Transportation Research Forum 43, no. 1 (2004): 121–38.
29 Shelley Poticha, “Building Housing Near Transit: A Long-Lasting Affordability Strategy,” Congressional Testimony before the Appropriations Subcommittee on Transportation, Housing and Urban Development, and Related Agencies, U.S. House of Representatives, March 8, 2007. Poticha is President and CEO of Reconnecting America, Oakland, CA.
30 Center for Transit-Oriented Development and Center for Neighborhood Technology, The Affordability Index: A New Tool for Measuring the True Affordability of Housing
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 6 8
>
>
N o
te s
Notes
Choice, Brookings Institution Urban Markets Initiative, Innovation Brief (January 2006), 10. Recognizing this saving, Fannie Mae created a Location Efficient Mortgage that enables borrowers to qualify for a larger loan if they are buying in areas that have lower average transportation costs.
31 John Pucher, “Public Transportation,” 212 (see endnote 23).
32 Dena Belzer et al., The Case for Mixed- Income, Transit-Oriented Development in the Denver Region (Oakland, CA: Center for Transit-Oriented Development, February 2007), 42.
33 One way to address this problem would be Pay-As-You-Drive car insurance. See “Transportation Affordability” (endnote 17).
34 Robert Cervero and Yu-Hsin Tsai, “San Francisco City CarShare: Travel-Demand Trends and Second-Year Impacts,” Institute of Urban & Regional Development, IURD Working Paper Series, Paper WP-2003-05 (August 1, 2003), http://repositories.cdlib. org/iurd/wps/WP-2003-05.
35 Directed by Congress, U.S. DOT and HUD have begun to collaborate on policies to promote affordable housing near transit. See Better Coordination of Transportation and Housing Programs to Promote Affordable Housing Near Transit, a Report to Congress from the U.S. Department of Transportation, Federal Transit Administration, and the U.S. Department of Housing and Urban Development, 2008. The HUD-FTA Interagency Working Group should continue to operate to identify legislation and administrative actions to better coordinate housing and transportation policies.
36 For example, the House Appropriations Subcommittee on Housing, Transportation and Urban Development, chaired by Rep. John Olver (D-MA), held a hearing with
a joint appearance by DOT Secretary Ray LaHood and HUD Secretary Shaun Donovan. In a joint press release, LaHood and Donovan announced a new partnership to coordinate housing and transportation to cut costs for working families, http://www.hud.gov/ news/release.cfm?content=pr09-023.cfm.
37 Present federal regulations do permit limited funds to be used for this purpose. The Metro system in Portland, OR, has used Congestion Mitigation and Air Quality (CMAQ) funds to acquire and sell land around transit stations for TOD, usually with an affordable housing component. PolicyLink, Equitable Development Toolkit: Transit Oriented Development, available at http://www. policylink.org.
38 Sarah Grady with Greg Leroy, Making the Connection: Transit-Oriented Development and Jobs (Washington, DC: Good Jobs First, March 2006).
39 Workforce housing is usually defined as housing that costs no more than 35 percent of the median wage in the area.
40 For a penetrating account of what happens “when work disappears” from communities, see William Julius Wilson, When Work Disappears: The World of the New Urban Poor (New York: Alfred A. Knopf, 1996).
41 Center to Protect Workers’ Rights, The Construction Chart Book: The U.S. Construction Industry and Its Workers, 4th ed. (Silver Spring, MD: Center to Protect Workers’ Rights, Center for Construction Research and Training, December 2007).
42 A recent study of 25 metropolitan areas found that hourly wages in construction (2004–2007) varied from $15.65 in the Dallas metropolitan area to $27.70 in the Chicago region. Todd Swanstrom, The Road to Good Jobs: Patterns of Employment in the Construction Industry (St. Louis, MO:
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 6 9
<
<
N o
te s
Notes
Transportation Equity Network and Public Policy Research Center, U. of Missouri – St. Louis, 2008).
43 Center to Protect Workers’ Rights, The Construction Chart Book (see endnote 41).
44 Daniel Hecker, “Occupational Employment Projections to 2014,” Monthly Labor Review (November 2005): 70–101.
45 Calculation of the number of jobs produced is based on Thomas P. Keane, “The Economic Importance of the National Highway System,” Public Roads 59, no. 4 (1996): 16–21.
46 The act is named after its Republican sponsors, James J. Davis, a Senator from Pennsylvania who was Secretary of Labor under three presidents, and Representative Robert L. Bacon of Long Island, NY. For Davis- Bacon wage rates state by state: see http:// www.gpo.gov/davisbacon/allstates.html.
47 Lisa Ranghelli, Replicating Success: The Alameda Corridor Job Training & Employment Program (Washington, DC: Center for Community Change, 2002).
48 For more examples of TEN’s successes: see http://www.transportationequity.org.
49 For best practices in pre-apprenticeship programs, see Kathleen Mulligan-Hansel, Making Development Work for Local Residents: Local Hire Programs and Implementation Strategies That Serve Low- Income Communities (Milwaukee, WI: Partnership for Working Families, 2008), http://www.communitybenefits.org/ downloads/Making%20Development%20 Work%20for%20Local%20Residents.pdf.
50 The 30 percent standard has been shown to be achievable in a number of projects around the country, such as in the St. Louis I-64 partnering agreement. A copy of that agreement is available on the Transportation Equity Network website, http://www.
transportationequity.org.
51 Presently, the federal law permits one-half of one percent of surface transportation funds to be used for local workforce development. One percent would do a better job of meeting the need while still representing a small cost to the overall project.
52 See http://www.uspirg.org/home/ reports/report-archives/transportation/ transportation2/a-better-way-to-go. Similarly, research has shown smart growth transportation policies, such as “fix-it-first” highway projects or public transportation, create more jobs than new highways that fuel more sprawl. Phillip Mattera with Greg Leroy, The Jobs are Back in Town: Urban Smart Growth and Construction Employment (Washington, DC: Good Jobs First, 2003).
53 An example of the political problems this can cause is the lawsuit filed by the Los Angeles Bus Riders’ Union against massive expenditures on a light-rail system at the same time that bus service was being cut. In March 1999, the Bus Riders’ Union won a court ruling for 532 new buses and 1,500 new union jobs for drivers and mechanics. For a discussion of the tensions between environmentalists and advocates of the poor in the transportation arena, see Joel Rast, “Environmental Justice and the New Regionalism,” Journal of Planning Education and Research 25 (2006): 249–63.
54 Research has demonstrated that car ownership increases employment and wages for low-income persons. See Steven Raphael and Michael Stoll, “Can Boosting Minority Car-Ownership Rates Narrow Inter-Racial Employment Gaps?,” Working Paper W00- 002, Program on Housing and Urban Policy, University of California – Berkeley, http:// urbanpolicy.berkeley.edu, and Paul Ong, “Car Ownership and Welfare-to-Work,” School of Public Policy and Social Research, University of California – Los Angeles, February 26, 2001, http://www.uctc.net/papers/540.pdf.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 7 0
>
>
N o
te s
Notes
1 U.S. Department of Agriculture, Economic Research Service, “Global Food Markets: Global Industry Structure,” 2008, http://www.ers.usda.gov/Briefing/ GlobalFoodMarkets/Industry.htm (accessed January 31, 2009).
2 Food insecurity is said to exist whenever “the availability of nutritionally adequate and safe food, or the ability to acquire acceptable foods in socially acceptable ways, is limited or uncertain.” S. A. Anderson, ed., “Core Indicators of Nutritional Status for Difficult-to-sample Populations,” Journal of Nutrition 120 (1990): 1559–1600, 1560. Food insecurity ranges from a painful sensation of hunger, at its most severe, to families being relegated to a few inexpensive staple foods—such as macaroni and cheese—that do not alone make up a nutritious and varied diet. Inconsistent availability of food, lack of transportation to grocery stores, and skipping meals to keep food costs down all are indicators of food insecurity. Conversely, food security refers to access by all people at all times to a sufficient quantity of safe, nutritious, affordable, and culturally appropriate food for an active, healthy life, obtained through conventional sources.
3 In this paper, “access” is used to signify spatial proximity or convenient and affordable transportation to destinations. Proximity is central because low-income urban households display lower rates of automobile ownership and may need to rely for grocery shopping on walking, taking the bus, or rides from acquaintances. Social, cultural, and economic categories of access of food are also key to this paper; they are defined, however, by the term “food security” (see endnote 2).
4 For example, the top five grocery retail chains captured 48 percent of the market in 2007, double that in 1997, http://www.nfu.org/wp-
content/2007-heffernanreport.pdf (accessed January 19, 2009).
5 Brookings Institution, From Poverty, Opportunity: Putting the Market to Work for Lower-income Families (Washington, DC: Brookings Institution, 2006), http://www. brookings.edu/reports/2006/07poverty_ fellowes.aspx (accessed January 19, 2009); K. Pothukuchi, “Attracting Supermarkets to Inner-city Neighborhoods: Economic Development Outside the Box,” Economic Development Quarterly 19 (2005): 232–44; E. Eisenhauer, “In Poor Health: Supermarket Redlining and Urban Nutrition,” GeoJournal 53 (2004): 125; and R. W. Cotterill and A. W. Franklin, “The Urban Grocery Store Gap,” Food Marketing Policy Issue Paper 8 (Storrs, CT: Food Marketing Policy Center, University of Connecticut, April 1995).
6 Today, the top food retailers control their own supply chains and manage their own fleets of trucks, warehouses, and buying offices. For example, Kroger has roughly 30 distribution centers to serve its 2,500 supermarkets, and other leading chains do the same to fully integrate their supply chains as a key strategy for remaining profitable. See Oakland Institute Report, “Food Chain Consolidation in U.S., 2007,” http://www. foodpolicy.in/html/archive/2007/rep/ oakland1.htm (accessed January 19, 2009). See also M. Hendrickson et al., “The Global Food System and Nodes of Power,” Report prepared for Oxfam America, August 2008 (accessed March 24, 2009), paper can be downloaded by clicking on SSRN at http:// papers.ssrn.com/sol3/papers.cfm?abstract_ id=1337273; and Competition Commission, Groceries Market Roundtable Meeting (Amended Notes) (London, UK: October 9, 2006).
7 See, for example, a Canadian study: K. Larsen and J. Gilliland, “Mapping the
Chapter 7: Sustainable Food Systems: Perspectives on Transportation Policy
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 7 1
<
<
N o
te s
Notes
Evolution of Food Deserts in a Canadian City: Supermarket Accessibility in London, Ontario, 1961–2005,” International Journal of Health Geographics 7 (2008), http://www. ij-healthgeographics.com/content/pdf/1476- 072X-7-16.pdf (accessed January 31, 2009). The study showed that in 1961, more than 75 percent of London’s downtown population lived within convenient access to grocery stores (i.e., a 10-minute bus ride combined with a 500-meter walk at the beginning or end of a bus trip). Because Canadian cities saw similar patterns of urban sprawl but at a lower intensity or scale than did most U.S. cities, it is safe to apply this study to U.S. cities as a general pattern.
8 M. A. Delucchi and J. Murphy, “How Large Are Tax Subsidies to Motor-vehicle Users in the U.S.?,” Journal of Transport Policy 15 (2008): 196–208.
9 For elaborations on this theme, see Brookings Institution, From Poverty, Opportunity: Putting the Market to Work for Lower- income Families (see endnote 5); D. Hendrickson, C. Smith, and N. Eikenberry, “Fruit and Vegetable Access in 4 Low-income Food Desert Communities in Minnesota,” Agriculture and Human Values 23 (2006): 371–83; M. Gallagher, Examining the Impact of Food Deserts on Public Health in Detroit (Chicago: Mari Gallagher Research and Consulting Group, 2007); M. Gallagher, Examining the Impact of Food Deserts on Public Health in Chicago (Chicago: Mari Gallagher Research and Consulting Group, 2006); S. N. Zenk et al., “Neighborhood Racial Composition, Neighborhood Poverty, and Spatial Accessibility of Supermarkets in Metropolitan Detroit,” American Journal of Public Health 95 (2005): 660–67; E. Bolen and K. Hecht, Neighborhood Groceries New Access to Healthy Food in Low-income Communities (San Francisco: California Food Policy Advocates, January 2003); and many others. The Brookings Institution study found, for example, that the average grocery
store in its sample of 2,384 low-income neighborhoods is 2.5 times smaller than the average grocery store in a high-income neighborhood. Also, there is about one mid- or large-sized grocer for every 69,055 residents in low-income neighborhoods, half the availability found in other neighborhoods. Access to only small grocery stores results in higher food prices for low-income shoppers. In particular, more than 67 percent of the same food products in its sample of 132 different products are more expensive in small grocery stores than in larger grocery stores.
10 T. C. Blanchard and T. A. Lyson, “Retail Concentration, Food Deserts, and Food Disadvantaged Communities in Rural America,” in Remaking the North American Food System, eds. C. C. Hinrichs and T. A. Lyson (Lincoln, NE: University of Nebraska Press, 2007); C. Wirth, R. Strochlic, and C. Getz, Hunger in the Fields: Food Insecurity among Farmworkers in Fresno County (CA: California Institute for Rural Studies, 2007); A. D. Liese et al., “Food Store Types, Availability, and Cost of Foods in a Rural Environment,” Journal of the American Dietetic Association 107 (November 2007): 1916–23; T. Blanchard and T. Lyson, “Food Availability & Food Deserts in the Nonmetropolitan South,” Southern Rural Development Center, 2006, http://srdc.msstate.edu/focusareas/health/ fa/fa_12_blanchard.pdf (accessed January 19, 2009); L. W. Morton et al., “Solving the Problems of Iowa Food Deserts: Food Insecurity and Civic Structure,” Rural Sociology 70 (2005): 94–112; L. W. Morton and T. C. Blanchard, “Starved for Access: Life in Rural America’s Food Deserts,” Rural Realities 1 (2007): 10; E. A. Bitto et al., “Grocery Store Access Patterns in Rural Food Deserts,” Journal for the Study of Food and Society 6 (2003): 35–48; C. Getz, “Perceived High Cost Deters Farmworkers from Eating Produce, According to UC Study,” University of California, News and Information Outreach, 2006, http://news.ucanr.org/
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 7 2
>
>
N o
te s
Notes
newsstorymain.cfm?story=899 (accessed January 19, 2009); and P. Kaufman and S. M. Lutz, “Competing Forces Affect Food Prices for Low-income Households,” Food Review 20 (May–August, 1997): 8–12.
11 K. Morland and S. Filomena, “Disparities in the Availability of Fruits and Vegetables between Racially Segregated Urban Neighbourhoods,” Cambridge Journals Online 10 (2007): 1481–89; L. V. Moore, A. V. Diez-Roux, “Associations of Neighborhood Characteristics with the Location and Type of Food Stores,” American Journal of Public Health 96 (2006): 325–31; D. Block and J. Kouba, “A Comparison of the Availability and Affordability of a Market Basket in Two Communities in the Chicago Area,” Public Health Nutrition 9 (2007): 837–45; M. Gallagher, 2007 and 2006 (see endnote 9); J. Block, R. A. Scribner, and K. B. De Salvo, “Fast Food, Race/Ethnicity, and Income: A Geographic Analysis,” American Journal of Preventive Medicine 27 (2004): 211–17; K. Morland et al., “Neighborhood Characteristics Associated with the Location of Food Stores and Food Service Places,” American Journal of Preventive Medicine 22 (2002): 23–29, and many others.
12 For a documentation of the decline and rise of farmers’ markets over the 20th century, see H. Tangires, Public Markets and Civic Culture in Nineteenth Century America (Baltimore: Johns Hopkins University Press, 2003); A. Brown, “Farmers’ Market Research 1940– 2000: An Inventory and Review,” American Journal of Alternative Agriculture 17 (2002): 167–76.
13 Since 1994, when the USDA started to track growth in farmers’ markets, more than 3,000 farmers’ markets have opened nationally, reaching a total of 4,685 markets in August 2008. USDA, Economic Research Service, “Global Food Markets: Global Industry Structure,” 2008, http://www.ers.usda.gov/ Briefing/GlobalFoodMarkets/Industry.htm
(accessed January 31, 2009).
14 American Farmland Trust, “Farming on the Edge Report,” http://www.farmland.org/ resources/fote/default.asp (accessed January 19, 2009).
15 Surface Transportation Policy Project, “Surface Transportation and Poverty Alleviation,” n.d., http://www.transact.org/ library/factsheets/poverty.asp (accessed January 19, 2009); R. D. Bullard and G. S. Johnson, eds., Just Transportation: Dismantling Race and Class Barriers to Mobility (Gabriola Island, BC: New Society Publishers, 1997); and A. D. Gardenshire, “Economic and Sociodemographic Influences on Autolessness: Are Missing Variables Skewing Results?,” Transportation Research Record 1670 (1999): 13–16.
16 S. H. Babey et al., Designed for Disease: The Link Between Local Food Environments and Obesity and Diabetes (Los Angeles: California Center for Public Health Advocacy, PolicyLink, and UCLA Center for Health Policy Research, 2008), http://www.healthpolicy. ucla.edu/pubs/publication.asp?pubID=250 (accessed January 19, 2009); L. Mikkelsen, S. Chehimi, and L. Cohen, Healthy Eating & Physical Activity: Addressing Inequities in Urban Environments (Oakland, CA: Prevention Institute, 2007); M. Gallagher, 2007 and 2006 (see endnote 9); M. C. Wang et al., “Changes in Neighbourhood Food Store Environment, Food Behaviour and Body Mass Index, 1981–1990,” Cambridge Journals Online 11 (2007): 963–70; K. Morland, S. Wing, and A. V. Diez-Roux, “The Contextual Effect of the Local Environment on Residents’ Diets: The Atherosclerosis Risk in Communities Study,” American Journal of Public Health 11 (2002): 1761–67; and The Food Trust, “Food Geography: How Food Access Affects Diet and Health,” http://www.thefoodtrust. org/catalog/download.php?product_id=120 (accessed January 19, 2009).
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 7 3
<
<
N o
te s
Notes
17 D. Rose and R. Richards, “Food Store Access and Household Fruit and Vegetable Use among Participants in the U.S. Food Stamp Program,” Public Health Nutrition 7 (2007): 1081–88; and K. Morland et al., “The Contextual Effect of the Local Environment on Residents’ Diets” (see endnote 16).
18 See, for example, Zenk et al., “Neighborhood Racial Composition” (endnote 9).
19 S. L. Handy, “Understanding the Link between Urban Form and Nontravel Behavior,” Journal of Planning Education and Research 15 (1996): 183–98; and J. Boivin and P. Matharu, “Bus Transit and Grocery Shopping in Detroit, Wayne State University Department of Geography and Urban Planning, unpublished paper, 2008.
20 S. L. Handy and K. Clifton, “Local Shopping as a Strategy for Reducing Automobile Travel,” Transportation 28 (2001): 317–46.
21 See, for example, K. Clifton, “Mobility Strategies and Food Shopping for Low- Income Families: A Case Study,” Journal of Planning Education and Research 23 (2004): 402–13.
22 This is an especially significant problem for the rural elderly. Food Security, Insecurity, and Hunger: Rural Food Access Patterns: Elderly Open-Country and In-Town Residents (Ames, IA: Iowa State University Extension, 2004); and K. Clifton, “Mobility Strategies” (see endnote 21).
23 N. Wrigley, “Understanding Store Development Programmes in Post-property- crisis UK Food Retailing,” Environment and Planning A 30 (1998): 15–35.
24 For recent statistics on national food insecurity, see, for example, M. Nord, M. Andrews, and S. Carlson, “Household Food Security in the United States, 2007,” USDA, Economic Research Service, Report #66,
2008, http://ers.usda.gov/Publications/ERR66/ ERR66.pdf (accessed January 19, 2009).
25 R. D. Bullard, G. S. Johnson, and A. O. Torres, eds., Sprawl City (Washington, DC: Island Press, 2000); R. D. Bullard and G. S. Johnson, eds., Just Transportation: Dismantling Race and Class Barriers to Mobility (Gabriola Island, BC: New Society Publishers, 1997); and Surface Transportation Policy Project, “Surface Transportation and Poverty Alleviation,” http://www.transact.org/library/factsheets/ poverty.asp (accessed January 19, 2009).
26 Food Research and Action Center, “Food Stamp Participation in May 2008 Sets Another Record High,” http://www.frac.org/ html/news/fsp/2008.05_FSP.htm (accessed January 19, 2009).
27 Food Research and Action Center, “A Guide to Food Stamp Outreach,” http://www.frac. org/html/federal_food_programs/programs/ fsoutreachprg.html#anchor826588 (accessed March 22, 2009).
28 C. Hefflin, “Who Exits the Food Stamp Program after Welfare Reform?,” http://www.ers.usda.gov/Briefing/ FoodNutritionAssistance/Funding/ RIDGEprojectSummary.asp?Summary_ID=46. (accessed January 19, 2009).
29 A Fresno County, CA, study found that nearly half of all farm worker households were food insecure compared to 36 percent of all county households. The same study also found that just over half and only about 36 percent of those eligible used food stamps in the winter and summer, respectively. C. Getz, “Perceived High Cost Deters Farmworkers from Eating Produce, According to UC Study,” University of California, News and Information Outreach, 2006, http://news. ucanr.org/newsstorymain.cfm?story=899 (accessed January 19, 2009).
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 7 4
>
>
N o
te s
Notes
30 T. Litman, “Transportation Affordability: Evaluation and Improvement Strategies,” Victoria Transport Policy Institute, 2007, http://www.vtpi.org/affordability.pdf. (accessed February 13, 2009).
31 E. Roberto, “Commuting to Opportunity: The Working Poor and Commuting in the United States,” Brookings Institution, 2008, http://www.brookings.edu/~/media/Files/ rc/reports/2008/0314_transportation_ puentes/0314_transportation_puentes.pdf (accessed February 13, 2009).
32 Surface Transportation Policy Project, “Surface Transportation and Poverty Alleviation,” http://www.transact.org/library/ factsheets/poverty.asp (accessed January 19, 2009).
33 M. Mauch and B. D. Taylor, “Gender, Race, and Travel Behavior: An Analysis of Household Serving Travel and Commuting in the San Francisco-Bay Area,” Transportation Research Record 1607 (1997): 147–53; and M. L. DeVault, Feeding the Family: The Social Organization of Caring as Gendered Work (Chicago: University of Chicago Press, 1991).
34 S. L. Handy and K. Clifton, “Local Shopping as a Strategy,” 331 (see endnote 20).
35 E. A. Bitto et al., “Grocery Store Access Patterns in Rural Food Deserts,” Journal for the Study of Food and Society 6 (2003): 35–48.
36 L. F. Alwitt and T. D. Donley, “Retail Stores in Poor Urban Neighborhoods,” Journal of Consumer Affairs 31 (1997): 139–64; C. Chung and S. L. Myers, “Do the Poor Pay More for Food? An Analysis of Grocery Store Availability and Price Disparities,” The Journal of Consumer Affairs 33 (1999): 276–96; and P. Kaufman and S. M. Lutz, “Competing Forces Affect Food Prices” (see endnote 10).
37 However, WIC rules do allow states leeway in deciding whether or how to
address transportation issues in making healthcare appointments. USDA, Food and Nutrition Service, Federal Register: “Special Supplemental Nutrition Program for Women, Infants, and Children (WIC): Miscellaneous Provisions,” Proposed Rule, 2002, http://www.fns.usda.gov/ cga/Federal-Register/2002/120202.pdf (accessed January 19, 2009). For example, in Michigan, WIC participants are allowed to seek transportation assistance for both healthcare as well as nutrition counseling appointments, whereas in West Virginia, only healthcare appointments are funded for transportation assistance. Other programs such as the Summer Food Service Program and senior nutrition programs are more sensitive to the transportation needs of their younger and older clients, respectively, and provide community grants for transportation assistance, http://www.summerfood.usda. gov/Community/transportation-grants.html (accessed March 24, 2009). Additionally, a small pot of USDA funding exists for farmers to bring product to market. Few studies exist on who benefits from this funding and how it is used.
38 Food Research and Action Center, “Rural Transportation Grants Successfully Increase Summer Food Participation,” 2006, http:// www.frac.org/afterschool/pdf/rural_ transportation_grants_report_2006.pdf (accessed February 12, 2009).
39 V. James Rhodes, The Agricultural Marketing System, 4th Ed. (Scottsdale, AZ: Gorsuch, Scarisbrick Publishers, 1993), cited in R. Pirog et al., Food, Fuel, and Freeways: An Iowa Perspective on How Far Food Travels, Fuel Usage, and Greenhouse Gas Emissions (Ames, IA: Iowa State University, Leopold Center for Sustainable Agriculture, 2001).
40 In March 2008, for example, wholesale food prices, an indicator of retail prices, rose the previous month at the fastest rate since 2003, with egg prices jumping 60
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 7 5
<
<
N o
te s
Notes
percent from a year ago, pasta products 30 percent, and fruits and vegetables 20 percent, according to the Labor Department. R. Gavin, “Surging Costs of Groceries Hits Home,” Boston Globe, March 9, 2008, http:// www.boston.com/business/personalfinance/ articles/2008/03/09/surging_costs_of_ groceries_hit_home/ (accessed January 19, 2009).
41 R. D. Bullard et al., Sprawl City; R. D. Bullard and G. S. Johnson, Just Transportation: Dismantling Race and Class Barriers to Mobility (see endnote 25, first two citations); E. Roberto, “Commuting to Opportunity” (see endnote 31); and T. Litman, “Transportation Affordability” (see endnote 30).
42 B. Taylor and P. Ong, “Spatial Mismatch or Automobile Mismatch? An Examination of Race, Residence and Commuting in U.S. Metropolitan Areas,” Urban Studies 32 (1995): 1453–73.
43 P. Ong and E. Blumenberg, “Job Access, Commute and Travel Burden among Welfare Recipients,” Urban Studies 35 (1998): 77–94.
44 R. Strochlic et al., “An Assessment of the Demand for Farm Worker Housing and Transportation in Mendicino County” (California Institute for Rural Studies, August 2008). See also Rural Assistance Center (RAC), http://www.raconline.org/info_ guides/transportation/ (accessed January 30, 2009). According to the RAC, 40 percent of all rural residents live in areas with no public transportation, and another 28 percent live in areas with limited levels of service.
45 R. Strochlic et al., “An Assessment of the Demand for Farm Worker Housing and Transportation in Mendicino County” (see endnote 44).
46 For one 2007 example, see CBC News, “Van Packed with Farm Workers Crashes in B.C.,
Killing 3,” March 7, 2007, http://www.cbc.ca/ canada/british-columbia/story/2007/03/07/ bc-van-crash.html?ref=rss (accessed February 12, 2009). The article reports an accident in which three people were killed and several others injured after a van designed for 10 people but carrying 17 workers rolled over Highway 1 in B.C.’s Fraser Valley.
47 See, for example, M. C. Heller and G. A. Keoleian, Life Cycle-Based Sustainability Indicators for Assessment of the U.S. Food System (CSS00-04) (Ann Arbor, MI: University of Michigan, Center for Sustainable Systems, 2000). From a personal communication with Ken Dahlberg, Professor Emeritus at Western Michigan University (September 27, 2006), of the energy used in the U.S. food system, roughly 21 percent is used for agricultural production, 14 percent for transportation, 16 percent for processing, 7 percent for packaging, 7 percent for eating establishments, 4 percent for food retailing, and 31 percent for home food refrigeration and cooking.
48 In Michigan, the nation’s second most agriculturally diverse state (California is first), only about 10 percent of the $25.7 billion spent on groceries at home and for eating out went to the state’s producers. P. Cantrell, The New Entrepreneurial Agriculture (Benzie, MI: Michigan Land Use Institute, 2003), http://mlui.org/downloads/newag.pdf (accessed January 19, 2009). Similar trends exist in Iowa and other agricultural states. For example, see Pirog et al., Food, Fuel, and Freeways (see endnote 39).
49 M. Hora and J. Tick, From Farm to Table: Making the Connection in the Mid-Atlantic Food System (Washington, DC: Capital Area Food Bank of Washington, DC, 2001) (citation derived from R. Pirog et al., 2001; see endnote 39).
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 7 6
>
>
N o
te s
Notes
50 This figure varies between 11 percent (Pirog et al., 2001) and 14 percent (Dahlberg, personal communication), suggesting that more new research is needed on this topic.
51 R. Pirog et al., Food, Fuel, and Freeways (see endnote 39).
52 For example, see D. Houston, J. Wu, and P. Ong, “Structural Disparities of Urban Traffic in Southern California: Implications for Vehicle-related Air Pollution Exposure in Minority and High-poverty Neighborhoods,” Journal of Urban Affairs 26 (2008): 565–92, for research related to the traffic that is generated by the Los Angeles (CA) port and its impacts on low-income and minority communities that are located nearby. Container traffic at the Ports of Los Angeles and Long Beach, CA, has tripled in the past 15 years, resulting in massive port-related heavy-duty diesel truck (HDDT) traffic on surface streets in the low-income and minority communities of Wilmington and western Long Beach adjacent to the ports. The volumes of HDDTs often reached 400 to 600/hour for several hours immediately upwind of sensitive land uses, such as schools, open-field parks, and residences. The documented health and environmental consequences of HDDT emissions raise serious public health concerns for the inhabitants who reside, work, attend school, or recreate in close proximity to roadways with HDDT traffic.
53 R. Pirog et al., Food, Fuel, and Freeways (see endnote 39).
54 Ibid.
55 R. Pirog and T. Van Pelt, “How Far Do Your Fruits and Vegetables Travel?,” Iowa State University, Leopold Center for Sustainable Agriculture, 2002, http://www.leopold. iastate.edu/pubs/staff/ppp/food_chart0402. pdf (accessed January 19, 2009).
56 R. Pirog et al., Food, Fuel, and Freeways, 33 (see endnote 39).
57 Pew Center on Global Climate Change, 2004, http://www.pewclimate.org/global-warming- basics/facts_and_figures/us_emissions/ usghgemsector.cfm (accessed March 24, 2009).
58 Roy Darwin, USDA, Economic Research Service, “Climate Change and Food Security,” 2001, http://www.ers.usda.gov/ publications/aib765/aib765-8.pdf (accessed March 23, 2009); and Food and Agriculture Organization, “Climate Change and Food Security,” 2007, http://www.un.org/ climatechange/pdfs/bali/fao-bali07-6.pdf (accessed March 23, 2009).
59 USDA, Economic Research Service, “Vegetables and Melons,” 2008, http:// www.ers.usda.gov/Briefing/Vegetables/ tomatoes.htm (accessed March 23, 2009).
60 USDA, “Foreign Agriculture Trade of the United States, Value of U.S. trade— Agricultural, Nonagricultural, and Total—and Trade Balance, by Fiscal Year,” updated January 13, 2009, http://www.ers.usda. gov/data/FATUS/index.htm#value (accessed January 19, 2009). According to the Census of Agriculture, in 2007, U.S. farms sold $297 billion in agricultural products while incurring $241 billion in production expenses, http:// www.agcensus.usda.gov/Publications/2007/ Online_Highlights/Fact_Sheets/economics. pdf (accessed March 22, 2009).
61 For example, see B. Meertens, “Agricultural Performance in Tanzania under Structural Adjustment Programs: Is It Really So Positive?,” Agriculture and Human Values 17 (2000): 333–46.
62 Indeed, both the history of supermarket development—seen, for example, in the rise of the first supermarkets of the Great Atlantic and Pacific Tea Company—and the
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 7 7
<
<
N o
te s
Notes
current dominance of Wal-mart underscore the preeminent contribution of logistics and transportation to their retail dominance.
63 A. Smith et al., “Validity of Food Miles as an Indicator of Sustainable Development,” AEA Technology Environment, commissioned by Defra, 2005, https://statistics.defra.gov. uk/esg/reports/foodmiles/execsumm.pdf (accessed January 19, 2009); A. C. McKinnon and A. Woodburn, “Logistical Restructuring and Road Freight Traffic Growth: An Empirical Assessment,” Transportation 23 (1996): 141–61; M. D. Boehlje, S. L. Hofing, and R. C. Shroeder, “Value Chains in the Agricultural Industries,” Staff Paper # 99-10, August 31, 1999, http://www.centrec.com/ Articles/value_chain_ag_industry/value_ chains_in_ag_industry.pdf (accessed March 24, 2009); and A. Potter and B. Gardner, “Management of Transport Resources: Investigating the External Cost Impact of Integrated Inbound Logistics,” Logistics and Operations Management Section, Cardiff Business School, December 2006, http:// www.cardiff.ac.uk/carbs/research/working_ papers/logistics/Investigating%20the%20 external%20cost%20impact%20of%20 integrated%20inbound%20logistics.pdf (accessed March 24, 2009).
64 M. H. Sonstegaard, “Competitive Access to North American Rail,” Transportation Quarterly 57 (2003): 61–67.
65 See, for example, D. Pimental, “The Ecological and Energy Integrity of Corn Ethanol Production,” in Reconciling Human Existence with Ecological Integrity eds. L. Westra, K. Bossellmann, and R. Westra (London: Earthscan, 2008); and J. P. W. Scharlemann and W. F. Laurance, “How Green Are Biofuels?,” Science 319 (2008): 43–44. Relative to petroleum, nearly all biofuels diminish greenhouse gas emissions, although crops such as switchgrass easily outperform soy and corn. Scharlemann and Laurance argue, however, that the process
for selecting one biofuel over another needs to consider its full environmental effects. When deforestation by palm oil producers or nitrogen use by corn producers is considered and their energy and emissions taken into account, they conclude that corn or canola biofuels may be worse for global warming than simply burning fossil fuels.
66 Pimental, “The Ecological and Energy Integrity of Corn Ethanol Production,” 252– 53 (see endnote 67).
67 See, for example, C. F. Runge and B. Senauer, “How Biofuels Could Starve the Poor,” Foreign Affairs (May/June 2007); and USDA, Agricultural Marketing Service, Ethanol Transportation Backgrounder: “Expansion of U.S. Corn-based Ethanol from the Agricultural Transportation Perspective,” 2007, http://www.ams.usda.gov/AMSv1.0/ge tfile?dDocName=STELPRDC5063605&acct= atpub (accessed January 19, 2009). Increased demand for ethanol has raised prices, which has resulted in increased production but also the diversion of corn from food-related uses to fuel.
68 D. Cronin, Inter Press Service, “Development: ‘Food Miles’ Hard to Digest,” 2008, http:// ipsnews.net/news.asp?idnews=41183 (accessed January 19, 2009).
69 Corn-based bioethanol has higher burden on environment and human health. A. Jha, “Biofuels More Harmful to Humans than Petrol and Diesel, Warn Scientists,” Guardian, February 2, 2009, http://www. guardian.co.uk/environment/2009/feb/02/ biofuels-health (accessed January 30, 2009). Researchers found the total environmental and health costs of gasoline are about 71 cents per gallon, while an equivalent amount of corn-ethanol fuel has associated costs of 72 cents to $1.45, depending on the use of chemicals in its production. However, there are high hopes for the next generation of biofuels, which can be made from organic
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 7 8
>
>
N o
te s
Notes
waste or plants grown on marginal land that is not used to grow foods. These have less than half the combined health and environmental costs of standard gasoline and one-third of current biofuels.
70 USDA, Agricultural Marketing Service, Ethanol Transportation Backgrounder: “Expansion of U.S. Corn-based Ethanol” (see endnote 67).
71 Prevention Institute, “Setting the Record Straight: Nutritionists Define Healthful Food,” 2009, http://www.preventioninstitute.org/ sa/documents/SettingtheRecordStraight_ final_031309_000.pdf (accessed March 24, 2009). See also policy guides by American Planning Association, http://www.planning. org/policy/guides/adopted/food.htm; and American Public Health Association, http:// www.apha.org/advocacy/policy/policysearch/ default.htm?id=1361 (both accessed March 23, 2009); and Catholic Healthcare West Food and Nutrition Vision Statement, n.d.
72 The recommendations also have other benefits, such as household savings and job creation. For example, see T. Litman, “Smart Transportation Economic Stimulation: Infrastructure Investments That Support Strategic Planning,” 2009, http://www. vtpi.org/econ_stim.pdf (accessed: February 6, 2009). Litman argues that a reasonable scenario of aggressive fuel economy targets, investments in alternative modes, and supportive land use policies can reduce U.S. fuel consumption 20 percent–40 percent, saving future consumers $150–$350 billion annually in fuel and vehicle expenses; providing economic benefits from reduced fuel import costs of similar magnitude; producing additional economic, social, and environmental benefits; and generating one to two million additional annual domestic jobs.
73 For example, Numero Uno Market in Los Angeles, CA, capitalized on the population density and high transit-dependence in the inner city to establish a van shuttle
service that takes shoppers who spend at least $30 to their door. Coordinated with two Metropolitan Transportation Authority bus routes as a means for people to get to the store, Numero Uno’s nine-van shuttle service made it one of the top-five grossing supermarkets in Los Angeles. M. Vallianatos, A. Shaffer, and R. Gottlieb, Transportation and Food: The Importance of Access (Los Angeles: Occidental College, Center for Food and Justice, Urban and Environmental Policy Institute, 2002). See also a feasibility study, for example, which makes a business case for such shuttles when provided by supermarkets. D. Cassady and V. Mohan, “Doing Well by Doing Good? A Supermarket Shuttle Feasibility Study,” Journal of Nutrition Education Behavior 36 (2004): 67–70.
74 Dedicated or special bus routes to connect low-income consumers have been provided by Austin, TX, and Hartford, CT. See Vallianatos et al., Transportation and Food (see endnote 73).
75 For example, Belo Horizonte’s (Brazil) municipal food programs include vans that act as mobile grocery stores. Together, these programs—along with special stores that sell foods in bulk, farm stands, and popular restaurants in low-income neighborhoods— cost less than one percent of the city’s budget. C. Rocha, “Urban Food Security Policy: The Case of Belo Horizonte, Brazil,” Journal for the Study of Food and Society 5 (2001): 36–47.
76 Research on nonemergency medical transportation shows cost savings as well as increased welfare as a result of transportation subsidies. R. Wallace et al., “Access to Health Care and Non-emergency Medical Transportation: Two Missing Links,” Transportation Research Record 1924 (2005): 76–84; and P. Hughes-Cromwick and R. Wallace, Executive Summary: Cost- benefit Analysis of Providing Nonemergency Medical Transportation (Washington, DC:
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 7 9
<
<
N o
te s
Notes
Transit Cooperative Research Program, Transportation Research Board, January 2006). Such a preventive health approach should be adopted in integrating transportation into federal nutrition programs.
77 An innovative example of farm worker transportation is demonstrated by Agricultural Industries Transportation Services, which provides vanpools to qualified farm workers in Kings, Tulare, and Fresno counties (CA), http://www.kartaits.com/ aitshome.htm (accessed January 19, 2009).
78 In addition to connecting rural food production with urban consumers, some cities are linking transportation and food production within the urban setting. In Tennessee, ISTEA funded a program that constructs community gardens along recreational corridors such as bike and walking trails. In Madison, WI, low-income gardeners working with the Community Action Coalition set up food gardens in highway rights of way, within cloverleaf intersections, and by the side of roads. See Vallianatos et al., Transportation and Food (endnote 73).
79 USDA, Cooperative State Research, Education and Extension Services, http://www.csrees. usda.gov/fo/communityfoodprojects.cfm (accessed March 24, 2009).
80 USDA, Summer Food Service Program, http:// www.summerfood.usda.gov/Community/ transportation-grants.html (accessed March 24, 2009).
81 R. Wallace et al., “Access to Health Care and Non-emergency Medical Transportation,” and P. Hughes-Cromwick and R. Wallace, Executive Summary: Cost-benefit Analysis (see endnote 76 for both citations).
82 For more information about these coalitions and organizations see: http://www.
transportationequity.org/; http://www. transportationforamerica.org/; http://www.transact.org/; http://www. completestreets.org/; and http://www. smartgrowthamerica.org/transportation.html.
83 For more information about these coalitions and organizations see: http://www.foodsecurity.org; http:// sustainableagriculture.net/; http://www.frac. org/; http://www.nffc.net/; and http://www. farmland.org/.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 8 0
>
>
N o
te s
Notes
1 “Fatally Hurt by Automobile,” The New York Times, September 14, 1899.
2 U.S. Department of Transportation, National Highway Traffic Safety Administration (NHTSA), Motor Vehicle Traffic Crashes as a Leading Cause of Death in the United States, 2005, Research Note DOT HS 810 936 (Washington, DC: National Highway Traffic Safety Administration, 2008).
3 Margie Peden et al., eds., World Report on Road Traffic Injury Prevention (Geneva, Switzerland: The World Health Organization, 2004).
4 Lawrence J. Blincoe et al., The Economic Impact of Motor Vehicle Crashes 2000, Report no. DOT HS-809-446 (Washington, DC: National Highway Traffic Safety Administration, 2002).
5 Ibid.
6 David Sleet, T. Bella Dinh-Zarr, and Ann Dellinger, “Traffic Safety in the Context of Public Health and Medicine,” in Improving Traffic Safety Culture in the United States: The Journey Forward, ed. AAA Foundation for Traffic Safety (Washington, DC: AAA Foundation, 2007).
7 National Center for Injury Prevention and Control, CDC Injury Fact Book (Atlanta: Centers for Disease Control and Prevention, 2006).
8 Larry Cohen and Susan Swift, “The Spectrum of Prevention: Developing a Comprehensive Approach to Injury Prevention,” Injury Prevention 5 (1999): 203–07.
9 Robert A. Caro, The Power Broker: Robert Moses and the Fall of New York (New York: Random House, Inc., 1975).
10 Reid Ewing et al., “Relationship between Urban Sprawl and Physical Activity, Obesity, and Morbidity,” American Journal of Health Promotion 18 (2003): 47–57.
11 Peter L. Jacobsen, “Safety in Numbers: More Walkers and Bicyclists, Safer Walking and Bicycling,” Injury Prevention 9 (2003): 205– 09.
12 Todd Litman, “Safe Travels: Evaluating Mobility Management Traffic Safety Benefits,” Victoria Transport Policy Institute, 2006, http://www.vtpi.org/safetr.pdf.
13 John Pucher and Lewis Dijkstra, “Promoting Safe Walking and Cycling to Improve Public Health: Lessons from the Netherlands and Germany,” American Journal of Public Health 93 (2003): 1509–16.
14 Fatality Analysis Reporting System, “National Statistics,” http://www-fars.nhtsa.dot.gov/ Main/index.aspx; and Centers for Disease Control and Prevention (CDC), “Web-based Injury Statistics Query and Reporting System (WISQARS),” http://www.cdc.gov/injury/ wisqars/index.html.
15 U.S. Department of Transportation, National Highway Traffic Safety Administration, 2007 Traffic Safety Annual Assessment, a brief statistical summary, DOT HS 811 017 (Washington, DC: National Highway Traffic Safety Administration, 2008).
16 NHTSA, Motor Vehicle Traffic Crashes (see endnote 2).
17 NHTSA, 2007 Traffic Safety Annual Assessment (see endnote 16).
18 Michelle Ernst, Mean Streets 2004: How Far Have We Come (Washington, DC: Surface Transportation Policy Partnership, 2004).
Chapter 8: Traffic Injury Prevention: A 21st-Century Approach
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 8 1
<
<
N o
te s
Notes
19 John Pucher and Lewis Dijkstra, “Promoting Safe Walking and Cycling” (see endnote 13).
20 Fatality Analysis Reporting System, “National Statistics” (see endnote 14).
21 John Pucher and John L. Renne, “Socioeconomics of Urban Travel: Evidence from the 2001 NHTS,” Transport Quarterly 57 (2003): 49–77.
22 Fatality Analysis Reporting System, “National Statistics” (see endnote 14); and U.S. Department of Transportation, NHTSA, Traffic Safety Facts: Pedestrians, Data DOT- HS-810-994, (Washington, DC: National Highway Traffic Safety Administration, 2007).
23 CDC, “Web-based Injury Statistics Query and Reporting System” (see endnote 14).
24 Kenneth G. Keppel et al., Trends in Racial and Ethnic-specific Rates for the Health Status Indicators: United States, 1990–98 (Hyattsville, MD: National Center for Health Statistics, Centers for Disease Control and Prevention, 2002).
25 Satomi Imai and Christopher Mansfield, “Disparities in Motor Vehicle Crash Fatalities of Young Drivers in North Carolina,” North Carolina Medical Journal 69 (2008): 182–87.
26 John Pucher and John L. Renne, “Socioeconomics of urban travel” (see endnote 21).
27 CDC, “Pedestrian Fatalities—Cobb, DeKalb, Fulton, and Gwinnett Counties, Georgia, 1994–1998,” Morbidity and Mortality Weekly Report 48 (1999): 601–05.
28 Richard Marosi, “Pedestrian Deaths Reveal O.C.’s Car Culture Clash,” Los Angeles Times Orange County Edition, November 28, 1999.
29 Robert B. Voas, A. Scott Tippetts, and Deborah A. Fisher, Ethnicity and Alcohol-
Related Fatalities: 1990 to 1994 (Washington, DC: National Highway Traffic Safety Administration, 2000).
30 David Shinar, “Demographic and Socioeconomic Correlates of Safety Belt Use,” Accident Analysis and Prevention 25 (1993): 745–55; Joann K. Wells, Allan F. Williams, and Charles M. Farmer, Seat Belt Use among African Americans, Hispanics, and Whites (Arlington, VA: Insurance Institute for Highway Safety, 2001); Robert B. Voas, A. Scott Tippetts, and Deborah A. Fisher, Ethnicity and Alcohol-Related Fatalities (see endnote 29); and U.S. Department of Transportation, NHTSA, N.C.f.S.a.A., Traffic Safety Facts 2001: Pedestrians (Washington, DC: U.S. Dept. of Transportation, 2001).
31 U.S. Department of Transportation, NHTSA, N.C.f.S.a.A., Traffic Safety Facts 2001: Overview (Washington, DC: U.S. Department of Transportation, 2001).
32 U.S. Department of Transportation, Federal Highway Administration (FHWA), National Household Travel Survey. Older Drivers: Safety Implications (Washington, DC: Federal Highway Administration, 2006).
33 Ibid.
34 Ibid.
35 Catherine E. Staunton, Howard Frumkin, and Andrew L. Dannenberg, “Changing the Built Environment to Prevent Injury,” in Handbook of Injury and Violence Prevention, eds. L. Bonzo Doll et al. (New York: Springer, 2007)
36 The Americans with Disabilities Act (ADA) of 1990 is a civil rights law that protects against discrimination based on disability. Title III delineates how public accommodations, including transportation, should be accessible to disabled persons.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 8 2
>
>
N o
te s
Notes
37 Insurance Institute for Highway Safety, “Q&As: Speed and speed limits,” http:// www.iihs.org/research/qanda/speed_limits. html.
38 The National Committee for Injury Prevention and Control, Injury Prevention Meeting the Challenge (New York: Oxford University Press, 1989).
39 Ibid.
40 U.S. Department of Transportation, NHTSA, Traffic Safety Facts: Alcohol-Impaired Driving (Washington, DC: National Highway Traffic Safety Administration, 2008).
41 The National Committee for Injury Prevention and Control, Injury Prevention Meeting the Challenge (see endnote 38).
42 A blood alcohol content (BAC) of 0.08 percent means 0.08 grams of pure alcohol per 100 milliliters of a person’s blood. BAC laws with a 0.05 percent threshold are believed to be ideal and recommended by the American Medical Association because driving skills begin to deteriorate markedly at 0.05 percent BAC. Current political will, however, makes 0.08 percent BAC laws more politically feasible. Laws with a 0.08 BAC exist in all states and the District of Columbia; and Ruth A. Shults, Randy W. Elder, and David A. Sleet, “Reviews of Evidence Regarding Interventions to Reduce Alcohol-impaired Driving,” American Journal of Preventive Medicine 21 (2001): 66–88.
43 Ruth A. Shults et al., “Reviews of Evidence” (see endnote 42); and CDC, “Impaired Driving,” http://www.cdc.gov/ncipc/ factsheets/drving.htm.
44 Randy W. Elder et al., “Effectiveness of Sobriety Checkpoints for Reducing Alcohol- involved Crashes,” Traffic Injury Prevention 3 (2002): 266–74.
45 Vehicles with breathalyzers installed cannot start if the alcohol ignition interlock equipment determines that the driver is intoxicated. These mechanisms can be required in the automobile manufacturing process, but there is some controversy with regards to infringing on personal privacy; and U.S. Department of Transportation, NHTSA, Reducing Impaired-driving Recidivism Using Advanced Vehicle-based Alcohol Detection Systems, a report to Congress (Washington, DC: National Highway Traffic Safety Administration, 2007).
46 Ralph W. Hingson, Monica H. Swahn, and David A. Sleet, “Interventions to Prevent Alcohol-related Injuries,” in Handbook of Injury and Violence Prevention, eds. L. S. Doll et al. (New York: Springer, 2007).
47 David A. Sleet et al., “Interventions to Reduce Impaired Driving and Traffic Injury,” in Drugs, Driving and Traffic Safety, eds. J. C. Verster et al. (Basel, Switzerland: Birkhauser, 2009).
48 Maria L. Alaniz, “Alcohol Availability and Targeted Advertising in Racial/Ethnic Minority Communities,” Alcohol Health and Research World 22 (1998): 286–89; and Thomas A. LaVeist and John M. Wallace, Jr., “Health Risk and Inequitable Distribution of Liquor Stores in African American Neighborhood,” Social Science and Medicine 51 (2000): 613–17.
49 U.S. Department of Transportation, NHTSA, Overview Traffic Safety Facts 1996 (Washington, DC: National Highway Traffic Safety Administration 1996).
50 CDC, “Bicycle-related Injuries: Data from the National Electronic Injury Surveillance System,” Morbidity and Mortality Weekly Report 36 (1987): 269–71.
51 U.S. Consumer Product Safety Commission, “NEWS from CPSC: CPSC Issues New Safety Standard for Bike Helmets,” http://www.cpsc. gov/cpscpub/prerel/prhtml98/98062.html.
H e
a lt
h y,
E q
u it
a b
le T
ra n
s p
o rt
a ti
o n
P o
li c
y
p g
. 1 8 3
<
<
N o
te s
Notes
52 Ibid.
53 Insurance Institute for Highway Safety, “Helmet Use Laws: Current U.S. Motorcycle and Bicycle Helmet Laws, http://www.iihs. org/laws/HelmetUseOverview.aspx.
54 U.S. Department of Transportation, NHTSA, Seat Belt Use in 2008—Overall Results (Washington, DC: National Highway Traffic Safety Administration, 2008).
55 Advocates for Highway and Auto Safety, “2009 Roadmap to State Highway Safety Laws,” http://www.saferoads.org/2009- roadmap-state-highway-safety-laws
57 U.S. Department of Transportation, NHTSA, States with Primary Enforcement Laws Have Lower Fatality Rates, Research Note DOT HS 810 557 (Washington, DC: National Highway Traffic Safety Administration, 2006).
58 U.S. Department of Transportation, NHTSA, Strengthening Safety Belt Use Laws— Increase Belt Use, Decrease Crash Fatalities and Injuries (Washington, DC: National Highway Traffic Safety Administration, 2004).
59 U.S. Department of Transportation, NHTSA, Traffic Safety Facts 2006: Motorcycles. (Washington, DC: National Highway Traffic Safety Administration, 2006).
60 NHTSA, Traffic Safety Facts 2006: Children (Washington, DC: National Highway Traffic Safety Administration, 2008).
61 Stephanie Zaza et al., Task Force on Community Preventive Services, “Reviews of Evidence Regarding Interventions to Increase the Use of Child Safety Seats,” American Journal of Preventive Medicine 21 (2001): 31–47.
62 U.S. Department of Transportation, 2009. Identifying Strategies to Reduce the Percentage of Unrestrained Young Children (Washington, DC: National Highway Traffic Safety Administration, 2009).
63 U.S. Department of Transportation, NHTSA, Announcement for Section 2003(b): Child Passenger Protection Education Grants (Washington, DC: National Highway Traffic Safety Administration, 2003).
64 Ibid.
65 National Safety Council, Special Issue: “Novice Teen Driving GDL and Beyond— Research Foundations for Policy and Practice Symposium,” Journal of Safety Research 38 (2007): 129–266.
66 Complete streets legislation was introduced in Congress by Rep. Doris Matsui (H.R. 1443) and Sen. Tom Harkin (S. 584) in 2008, but neither bill passed. Several states have passed their own complete streets bills, but broad federal recognition of the value and priority of complete streets in the next authorization of the surface transportation bill could encompass the versions in the Senate and the House of Representatives and signal a commitment to safe mobility among all travelers.
p
g . 1 8 4
>
>
Photo Credits: p.9 (left to right): Frances Twitty, www.pedbikeimages.org / Dan Burden p.21 (left to right): www.pedbikeimages.org / Dan Burden, www.pedbikeimages.org / Dan Burden p.22: Tim McCaig p.24: Jacom Stephens p.27 (left to right): D. Samuel Marsh Photography, Dave Huss p.33: D. Samuel Marsh Photography p.37: Todd Litman, Todd Litman p.57: Tomaz Levstek p.60: John Zellmer p.61: ©iStockphoto.com (ddoorly) p.63 (left to right): www.pedbikeimages.org / Dan Burden, David H. Lewis p.75: www.pedbikeimages.org / Dan Burden p.77: www.pedbikeimages.org / Dan Burden p.79 (left to right): www.pedbikeimages.org / Dan Burden, Atlanta Regional Commission p.82: PEDS.org p.83: Clevelandskyscrapers.com p.85: www.pedbikeimages.org / Dan Burden p.86: David Kolb (www.dkolb.org) p.88: www.pedbikeimages.org / Dan Burden p.89: www.pedbikeimages.org / Dan Burden p.99 (left to right): D. Samuel Marsh Photography, Jeffrey Zavitski p.103: ©iStockphoto.com (Rhoberazzi) p.105: Mario Savoia p.110: Lisa F. Young p.113 (left to right): Anna Bryukhanova, Claudia Dewald p.116: Lorie Slater p.122: Britta Kasholm-Tengve p.128: Bruce Block p.131 (left to right): Duncan Walker, www.pedbikeimages.org / Austin Brown p.140: Izabela Habur p.142: ©iStockphoto.com (tillsonburg)
221 Oak Street Oakland, CA 94607 t 510-444-7738 f 510-663-1280 www.preventioninstitute.org
Headquarters:
1438 Webster Street, Suite 303 Oakland, CA 94612 t 510 663-2333 f 510 663-9684 [email protected]
CommuniCations offiCe:
55 West 39th Street, 11th Floor New York, NY 10018 t 212 629-9570 f 212 629-7328 www.policylink.org
Commissioned by:
www.convergencepartnership.org
- book_cover_final4_web
- POL_book_interior_final7
- POL_notes_final3_web.pdf
memo3/HEI_SpR17_Exec_Sum.pdf
Special RepoRt 17
January 2010
Traffic-Related Air Pollution: A Critical Review of the Literature on
Emissions, Exposure, and Health Effects
A Special Report of the HEI Panel on the Health Effects of Traffic-Related Air Pollution
ExECuTIvE SummARy
ABOuT HEI
The Health Effects Institute is a nonprofit corporation chartered in 1980 as an independent research organization to provide high-quality, impartial, and relevant science on the effects of air pollution on health. To accomplish its mission, the institute
• Identifies the highest-priority areas for health effects research;
• Competitively funds and oversees research projects;
• Provides intensive independent review of HEI-supported studies and related research;
• Integrates HEI’s research results with those of other institutions into broader evaluations; and
• Communicates the results of HEI research and analyses to public and private decision makers.
HEI receives half of its core funds from the u.S. Environmental Protection Agency and half from the worldwide motor vehicle industry. Frequently, other public and private organizations in the united States and around the world also support major projects or certain research programs. Additional work for this report was funded by the u. S. Federal Highway Administration.
HEI has funded more than 280 research projects in North America, Europe, Asia, and Latin America, the results of which have informed decisions regarding carbon monoxide, air toxics, nitrogen oxides, diesel exhaust, ozone, particulate matter, and other pollutants. These results have appeared in the peer-reviewed literature and in more than 200 comprehensive reports published by HEI.
HEI’s independent Board of Directors consists of leaders in science and policy who are committed to fostering the public–private partnership that is central to the organization. The Health Research Committee solicits input from HEI sponsors and other stakeholders and works with scientific staff to develop a Five-year Strategic Plan, select research projects for funding, and oversee their conduct. The Health Review Committee, which has no role in selecting or overseeing studies, works with staff to evaluate and interpret the results of funded studies and related research.
All project results and accompanying comments by the Health Review Committee are widely disseminated through HEI’s Web site (www.healtheffects.org), printed reports, newsletters, and other publications, annual conferences, and presentations to legislative bodies and public agencies.
special report 17
INTRODUCTION
Motor vehicles are a significant source of urban air pollu- tion and are increasingly important contributors of anthropo- genic carbon dioxide and other greenhouse gases. As awareness of the potential health effects of air pollutants has grown, many countries have implemented more stringent emissions controls and made steady progress in reducing the emissions from motor vehicles and improving air quality. However, the rapid growth of the world’s motor-vehicle fleet due to population growth and eco- nomic improvement, the expansion of metropolitan areas, and the increasing dependence on motor vehicles because of changes in land use has resulted in an increase in the fraction of the popu- lation living and working in close proximity to busy highways and roads — counteracting to some extent the expected benefits of pollution-control regulations and technologies.
This Special Report, developed by the Health Effects Institute (HEI) Panel on the Health Effects of Traffic-Related Air Pollu- tion, summarizes and synthesizes information linking emis- sions from, exposures to, and health effects of traffic sources (i.e., motor vehicles). The term traffic-related exposure is used in this report to refer to exposure to primary emissions from motor vehicles, not to the more broadly dispersed secondary pollutants such as ozone (O3) that are derived from these emis- sions. The report focuses on specific scenarios with a high aggregation of motor vehicles and people — that is, urban set- tings and residences in proximity to busy roadways.
EMISSIONS FROM MOTOR VEHICLES
Motor vehicles emit large quantities of carbon dioxide (CO2), carbon monoxide (CO), hydrocarbons (HC), nitrogen oxides (NOX), particulate matter (PM), and substances known as mobile- source air toxics (MSATs), such as benzene, formaldehyde, acet- aldehyde, 1,3-butadiene, and lead (where leaded gasoline is still in use). Each of these, along with secondary by-products, such as ozone and secondary aerosols (e.g., nitrates and inor- ganic and organic acids), can cause adverse effects on health and the environment. Pollutants from vehicle emissions are
related to vehicle type (e.g., light- or heavy-duty vehicles) and age, operating and maintenance conditions, exhaust treatment, type and quality of fuel, wear of parts (e.g., tires and brakes), and engine lubricants used. Concerns about the health effects of motor-vehicle combustion emissions have led to the introduc- tion of regulations and innovative pollution-control approaches throughout the world that have resulted in a considerable reduc- tion of exhaust emissions, particularly in developed countries. These reductions have been achieved through a comprehensive strategy that typically involves emissions standards, cleaner fuels, and vehicle-inspection programs. Recognizing the likely continued growth in the vehicle fleet and the remaining prob- lems in traffic-related air quality, the United States, European countries, Japan, and other countries are continuing to push for even stricter emissions controls in coming years.
Resuspended road dust, tire wear, and brake wear are sources of noncombustion PM emissions from motor vehicles. As emissions controls for exhaust PM become more widespread, emissions from noncombustion sources will make up a larger proportion of vehi- cle emissions. Noncombustion emissions contain chemical com- pounds, such as trace metals and organics, that might contribute to human health effects. However, current estimates of these emis- sions are highly uncertain. Thus, although they are not regulated in the way exhaust emissions are, noncombustion emissions will need to be considered more closely in future assessments of the impact of motor vehicles on human health.
The quantification of motor-vehicle emissions is critical in estimating their impact on local air quality and traffic-related exposures and requires the collection of travel-activity data over space and time and the development of emissions inven- tories. Emissions inventories are developed based on complex emissions models (of which the U.S. Environmental Protection Agency’s MOBILE6 has been the most widely used) that pro- vide exhaust and evaporative emissions rates for total HC, CO, NOx, PM, sulfur dioxide (SO2), ammonia (NH3), selected air tox- ics, and green house gases (GHGs) for specific vehicle types and fuels. The quality of the travel-activity data (such as vehicle- miles traveled, number of trips, and types of vehicles) and the complex algorithms used to derive the emissions factors sug- gest the presence of substantial uncertainties and limitations in the resulting emissions estimates (NARSTO 2005). It should be noted that estimates of PM emissions have had very limited field valuation and verification.
The actual measurement of motor-vehicle emissions is critically important for validating the emissions models. Studies that have sampled the exhaust of moving vehicles in real-world situations (specifically, in tunnels or on roadways) have contributed very useful information about the emissions rates of the current motor- vehicle fleet and also have allowed the evaluation of the impact of new emission-control technologies and fuels on emissions.
EXECUTIVE SUMMARY Traffic-Related Air Pollution: A Critical Review of the Literature on Emissions, Exposure, and Health Effects
This Executive Summary is excerpted from HEI Special Report 17, Traffic-Related Air Pollution: A Critical Review of the Literature on Emissions, Exposure, and Health Effects, by the Institute’s Panel on the Health Effects of Traffic-Related Air Pollution. The entire report is available at www.healtheffects.org or from HEI.
This document was produced with partial funding by the United States Environ- mental Protection Agency under Assistance Award CR–83234701 to the Health Effects Institute; however, it has not been subjected to the Agency’s peer and admin- istrative review and therefore may not necessarily reflect the views of the Agency, and no official endorsement by it should be inferred. The contents of this docu- ment also have not been reviewed by private party institutions, including those that support the Health Effects Institute; therefore, it may not reflect the views or policies of these parties, and no endorsement by them should be inferred.
3
ExECuTIvE SummARy
special report 17
Receptor models have been used to estimate the contributions of various types of sources, including motor vehicles, to ambi- ent air pollution. Some of the models (those defined as chemi- cal mass balance models) require the knowledge of the chemi- cal profile of both the emissions of all the area sources and the air at the receptor (that is, the impacted location). Other models (referred to as principal components and factors analyses) do not require a priori knowledge of the source profiles. The application of these models has yielded a wide range of results on the contri- bution of motor vehicles to ambient pollution, depending on the model, the location of the monitoring sites, and the other sources present. In U.S. cities, the results show that motor-vehicle con- tributions range from 5% in Pittsburgh, Pa., under conditions with very high secondary aerosol, to 49% in Phoenix, Ariz., and 55% in Los Angeles, Calif. Outside the United States, estimates of the motor-vehicle contribution to PM2.5 (PM # 2.5 µm in aero- dynamic diameter) range from 6% in Beijing, China, to 53% in Barcelona, Spain.
Ultimately, an important goal of emissions-characterization studies is to improve our ability to quantify human exposure to emissions from motor vehicles, especially in locations with high concentrations of vehicles and people. Such characterization requires improving emissions inventories and a more complete understanding of the chemical and physical transformations on and near roadways that can produce toxic gaseous, semivolatile, and particle-phase chemical constituents.
ASSESSMENT OF EXPOSURE TO TRAFFIC-RELATED AIR POLLUTION
Traffic-related emissions contribute to primary and second- ary local, urban, and regional (background) pollutant concen- trations against a background of similar contaminants emitted from other sources. Traffic emissions are the principal source of intra-urban variation in the concentrations of air pollutants in many cities; thus, population-oriented central monitors cannot by themselves capture this spatial variability. Studies that have examined gradients in pollutants as a function of distance from busy roadways have indicated exposure zones for traffic-related air pollution in the range of 50 to 1500 m from highways and major roads, depending on the pollutant and the meteorologic conditions.
Because it is not practical or feasible to measure all the com- ponents of the traffic-pollutant mix, surrogates of traffic-related pollution have been used as a reasonable compromise for assess- ing the contribution of traffic emissions to ambient air pollution and for estimating traffic exposure. Surrogates can also help in the assessment of spatial and temporal distributions of ambi- ent pollution related to motor vehicles and of traffic-mitigation control strategies.
Two broad categories of surrogates have been used in epide- miology studies to estimate traffic exposure: (1) measured or modeled concentrations of pollutant surrogates and (2) direct
measures of traffic itself (such as proximity, or distance, of the residence to the nearest road and traffic volume within buffers). The most commonly used traffic-pollutant surrogates include CO, NO2, elemental carbon (EC; or black carbon [BC] or black smoke [BS]), PM, benzene, and ultrafine particles (UFP). Exposure mod- els include geostatistical interpolation, land-use regression, dis- persion, and hybrid models (the latter combine time–activity data, personal measurements, and models). They incorporate numer- ous parameters (such as meteorologic variables, data on land use, traffic data, and monitoring data or emissions rates depending on the model) and can improve the spatial representation of the local impact of traffic against a background of regional and urban concentrations. However, the accuracy of the inputs is critical to the usefulness of any given model.
None of the pollutant surrogates considered in the report met all the criteria for an ideal surrogate. Data are not available to assess the ratios of the surrogates to emissions from all sources over time. CO, benzene, and NOx (in this case NO2), found in on-road vehicle emissions, are components of emissions from all sources, making it difficult to disentangle the contributions from motor vehicles from other sources (including some in indoor environments). Primary, on-road vehicle emissions of PM (PM2.5 or PM10 [PM # 10 µm in aerodynamic diameter]) represent only a small contribution to emissions from all sources, typically around 3%. EC has been used as a surrogate, primarily for diesel exhaust, although it is not a specific marker, unless other sources are ruled out. UFP concentrations are very high in vehicle-exhaust plumes but decrease rapidly with distance from the source, which poses a significant challenge for characterization of the spatial and tem- poral concentration gradients of UFP from roadway traffic.
With regard to exposure models, the Panel noted that, although proximity models (direct measures of traffic) are the easiest to implement, they are error prone because they ignore the parameters that affect the dispersion and physicochemical activity of the pollutants. Moreover, estimates based on proxim- ity can be confounded by factors such as socioeconomic status and noise. Geostatistical interpolation models are best imple- mented in conjunction with dense, well-distributed monitoring networks; their chief limitations are the size of the network and the number of measurements needed over time to estimate the spatial distribution of pollution surrogates accurately. Land-use regression is appealing in that it can account for the diversity of sources that contribute to a surrogate; however, the true contri- bution (in terms of associated variance) of traffic to the regres- sion is not always known or reported. Dispersion models utilize motor-vehicle–emissions and air-quality data and incorporate meteorologic data, but must be calibrated correctly to realize their advantages. These models are very data- and computation- intensive and depend on the validity of the model assumptions. Hybrid models that combine measurements of personal expo- sure to traffic surrogates or time–activity data with exposure models come closest to a logistically feasible “best” estimate of human exposure.
4
special report 17
ExECuTIvE SummARy
Factors influencing ambient concentrations of a traffic-pollut- ant surrogate are related to time–activity patterns, meteorologic conditions, vehicle volume and type, driving patterns, land-use patterns, the rate at which chemical transformations take place, and the degree to which the temporal and spatial distribution of the surrogate reflects the traffic source.
To improve assessment of exposure to traffic-related pollution, a potential solution is the deployment of a large number of moni- tors in places where concentrations of air pollutants are expected to be highly variable and the population density is high. The use of models that incorporate numerous spatial factors in order to estimate exposures that are more relevant to the individual’s exposure situation can also be helpful.
The Panel concluded that the impact of vehicle emissions extends beyond the local scale to the urban and regional scales. What people are exposed to is influenced by their proximity to the sources, the presence of other ambient or microenvironmen- tal sources, and time–activity patterns. If, as the evidence sug- gests, groups of lower socioeconomic status experience higher exposures than groups of higher socioeconomic status, this mer- its consideration in the interpretation of epidemiologic findings and in future regulatory actions.
Based on a synthesis of the best available evidence, the Panel identified an exposure zone within a range of up to 300 to 500 m from a highway or a major road as the area most highly affected by traffic emissions (the range reflects the variable influence of background pollution concentrations, meteorologic conditions, and season) and estimated that 30% to 45% of people living in large North American cities live within such zones.
HEALTH EFFECTS OF TRAFFIC-RELATED AIR POLLUTION: EPIDEMIOLOGY AND TOXICOLOGY
In reviewing the epidemiologic literature on the association between exposure to traffic-related air pollution and health out- comes, the Panel developed criteria for the inclusion of stud- ies based on the characterization of traffic exposure. The Panel decided to include only studies that investigated associations between primary emissions from traffic and human health and that provided specific documentation of a traffic source and estimates of exposure on a local scale. Thus, studies that relied exclusively on measurements from a central monitoring site were not included unless the site was in proximity to traffic. The Panel also developed criteria for inferring whether associations between exposure and health outcome were causal by adapting the criteria used by the U.S. Surgeon General in the report The Health Consequences of Smoking: A Report of the Surgeon Gen- eral (U.S. Department of Health and Human Services 2004). In order to deem the evidence sufficient to conclude that association between a metric of traffic exposure and an outcome was causal, it was necessary for the magnitude and direction of the effect esti- mates to be consistent across different populations and times and to rule out with reasonable confidence chance, bias in subject
selection, and confounding (in particular, socioeconomic sta- tus). The four inference criteria applied to this review are listed in Table 1. To these criteria the Panel added a traffic-specific coherence criterion (also included in Table 1) to account for the degree of validity of the traffic-specific exposure metrics. As noted earlier, the Panel concluded that not all traffic-exposure measures have equivalent validity and considered simple mea- sures of proximity to roads or road length and of pollutant sur- rogates without specific traffic data to be the least specific. The proximity measures are also likely to introduce confounding.
Modeled estimates of exposure to traffic pollution were thought to be, a priori, more valid than traffic density estimates alone because they account for other factors that affect the exposure, such as geography, land use, and meteorology, when making estimates for particular locations. In addition, the validity of estimates can be enhanced by modeling strategies that sepa- rately estimate the contribution of traffic and background pollu- tion to personal exposure.
The Panel developed qualitative and quantitative summa- ries (in tables and figures) for the estimates of the associations between traffic-related exposure and various health outcomes for the studies reviewed, but did not derive meta-analytic summaries by pooling associations estimates because of the lack of equiva- lence among the exposure measures and populations studied.
The Panel also reviewed the literature on the toxicology of traffic-related pollution. This included studies of direct expo- sures to traffic emissions (though there were very few in this cat- egory), studies that utilized laboratory atmospheres that replicate aspects of the traffic mix (such as concentrated ambient particles, or gasoline or diesel exhaust), and studies of specific components of emissions from motor vehicles. The aim was to identify pos- sible mechanisms by which exposure to traffic pollutants may cause effects and provide an understanding of the role of traffic emissions in the effects being observed in epidemiology studies. While toxicology studies are limited in their ability to capture the full complexity of human exposure — because of the small num- ber of subjects and, in animal studies, the relevance of the results to humans — they offer the opportunity to explore hypotheses on specific pathophysiologic mechanisms of action.
The Panel evaluated whether oxidative stress might be the underlying mechanism of action by which exposure to pollutants from traffic may lead to adverse health effects. Oxidative stress results from events occurring in any tissue in the body when the prooxidant–antioxidant balance is disturbed. This imbal- ance can happen when the generation of reactive oxygen species, or free radicals, exceeds the available antioxidant defenses and is characterized by the presence of increased cellular concen- trations of oxidized lipids, proteins, and DNA. Oxidative stress can trigger inflammatory reactions, which lead to an increased production of oxidants by activated phagocytes recruited to the airways, perpetuating the cycle of oxidative injury.
The Panel concluded that, although the evidence supported the hypothesis that oxidative stress is an important determinant of health effects associated with ambient air pollution in general,
5
ExECuTIvE SummARy
special report 17 6
special report 17
ExECuTIvE SummARy
the extent to which primary traffic-related pollutants contribute to the burden of reactive oxygen species experienced by humans near roadways remains undefined.
The Panel’s main conclusions regarding the epidemiologic associations between exposure to traffic-related air pollution and health outcomes and the toxicologic evidence (when avail- able) are presented below for each health outcome. A discussion of the extent to which toxicology studies do or do not provide general mechanistic support for the observations and inferences contributed by epidemiology studies is also provided.
all-caUSe aND caRDioVaScUlaR MoRtalitY
epidemiology
Very few studies of all-cause mortality or cardiovascular mor- tality and long-term exposure met the criteria for inclusion in the report. Mostly because of the small number of studies, the evidence for an association of all-cause mortality with long- term exposure was classified as “suggestive but not sufficient” to infer a causal association. Additional factors that led to this classification were the substantial differences among popula- tions, time periods, and confounders across studies.
Only four time-series studies of all-cause mortality associ- ated with short-term exposure met the Panel’s criteria; these, too, were classified as “suggestive but not sufficient,” largely on the strength of one well-done study (Maynard et al. 2007). Two time-series studies based on source-apportionment models were found to have a number of limitations that prevented a stronger statement about inferred causality.
Many of the issues that applied to studies of all-cause mortality applied as well to studies of cardiovascular mortality associated with long-term exposure and led, similarly, to a classification of “suggestive but not sufficient.” Only two time-series studies of cardiovascular mortality met the inclusion criteria, and although they both show positive associations, the Panel concluded that, given the overall paucity of studies, the evidence for effects of short-term exposure was “inadequate and insufficient.”
caRDioVaScUlaR MoRBiDitY
epidemiology
Studies that documented changes in cardiac physiology (such as heart-rate variability) after short-term exposure to traffic- related pollution (which was assessed using surrogates, source apportionment, or pseudo-personal monitoring) provided strong evidence for a causal association with the exposure. However, the failure of some studies to consider stress and noise as poten- tial confounders led the Panel to classify them as “suggestive but not sufficient” to infer a casual association. Among the stud- ies that evaluated cardiovascular morbidity, two well-executed studies on hospitalization for acute myocardial infarction were identified (Rosenlund et al. 2006; Tonne et al. 2007). In addition, a prospective study in a German cohort reported an association
between living near a major road and coronary-artery calcifi- cation as well as higher prevalence of coronary heart disease (Hoffmann et al. 2006, 2007). Collectively, these studies made a very strong case for an association between exposure to traffic- related pollutants and atherosclerosis. However, because of the small number of studies, the Panel classified them as “sugges- tive but not sufficient” to infer a causal association.
toxicology
There have been a few toxicology studies that examined the cardiovascular effects of traffic emissions specifically. However, the Panel concluded that the recent toxicology literature pro- vides suggestive evidence that exposure to pollutants that are components of traffic emissions, including ambient and labora- tory-generated PM and exhaust from diesel and gasoline-fueled engines, alters cardiovascular function. There is also evidence, albeit inconsistent, for acute effects on vascular homeostasis and suggestive evidence in animal models that repeated expo- sures to ambient PM in general enhance the development of ath- erosclerosis. Some studies support the involvement of oxidative stress. Although the evidence from toxicology studies in isola- tion is not sufficient in terms of a causal association between traffic emissions and the incidence or progression of cardiovas- cular disease, when viewed together with the epidemiologic evidence, a stronger case could be made for a potential causal role for traffic-related pollutants in cardiovascular-disease mor- bidity and mortality. The extent to which these associations apply to individuals without underlying cardiovascular disease cannot be determined from the evidence available at this time.
aStHMa aND ReSpiRatoRY SYMptoMS
Asthma is an inflammatory disease of the lung airways char- acterized by episodic obstruction of the airways, which can lead to chronic obstructive lung disease. The most prevalent form of asthma in children and young adults is allergic asthma, which develops as an immune response to inhaled allergens. Indi- viduals with asthma and other allergic conditions who have an increased tendency to develop immediate and localized reac- tions to allergens (such as pollens) that are mediated by immu- noglobulin E (IgE) are referred to as “atopic.”
epidemiology
In epidemiology studies, asthma is most frequently identi- fied by means of responses to questionnaires that do not make use of a single, universally accepted set of questions, alone or in combination with other criteria. This is further complicated by the challenges of distinguishing factors that affect its onset from those (often the same factors) that lead to its episodic worsening. A history of asthma symptoms (such a wheezing) often is used in epidemiology studies as part of the definition both of asthma’s onset (incidence) and of its prevalence and exacerbation.
Respiratory Health Problems in Children: Asthma Incidence and Prevalence Seven studies conducted in four separate
7
ExECuTIvE SummARy
special report 17
cohorts and one case-control study qualified as studies of asthma incidence in children. Eleven studies qualified as studies of asthma prevalence in children. From these studies, the Panel concluded that living close to busy roads appears to be an inde- pendent risk factor for the onset of childhood asthma. The Panel considered the evidence for a causal relation to be in a gray zone between “sufficient” and “suggestive but not sufficient.” The results found across the studies followed a pattern that would be expected under the plausible assumption that the pollutants really are causally associated with asthma development, if only among a subset of children with some accompanying pattern of endogenous or exogenous susceptibility factors. The conditions that underlie an increased risk for asthma development among children exposed to traffic-related pollutants are not known.
Exacerbation of Symptoms in Children with and without Asthma and Health-Care Utilization for Respiratory Problems Among the more than 20 cohort and cross-sectional studies reviewed that examined the association between exposure to traffic-related pollution and wheezing (an important symptom in the expression and diagnosis of asthma) in children, there was a high degree of consistency in finding positive associa- tions, many of which reached statistical significance (i.e., had reasonably precise point estimates of associations). This was true particularly for the large majority of studies that used mod- els to assign estimates of local concentrations of pollutants, such as NO2 or soot (the carbonaceous component of PM), to the place of residence of the study participants. Studies based on proximity or traffic density also indicated an association between exposure and wheezing. In addition, exacerbation of other asthma-related symptoms, such as cough or dry cough, was consistently associated with exposure across a variety of exposure measures. Although most studies were not restricted to children with asthma, all these symptoms were more prevalent among those with asthma, and it is very likely that the observed associations were driven by exacerbations of asthma in mixed groups of participants. The Panel concluded that the evidence is “sufficient” to infer a causal association between traffic expo- sure and exacerbations of asthma but that it is “inadequate and insufficient” to infer a causal association between exposure and respiratory symptoms in children without asthma.
Nine studies assessed the association between exposure to traf- fic-related pollution and the use of health-care services to treat respiratory problems in children. Most of the studies reported positive associations between exposure and hospital-admission rates, but the majority had methodologic problems that hampered their interpretation. The panel concluded that there is “inade- quate and insufficient” evidence to infer a causal association.
Respiratory Health Problems in Adults: Asthma Onset and Respi- ratory Symptoms The Panel noted that the evidence between exposure to traffic-related pollution and new adult asthma was “inadequate and insufficient” as this was investigated in only one study (Modig et al. 2006). The Panel reviewed 17 studies on respi- ratory symptoms, of which all but one relied on proximity to roads or traffic-density measures, and concluded that the evidence for a causal association is “suggestive but not sufficient.”
toxicology
The few human studies in which subjects were exposed to realistic traffic conditions (a road tunnel or busy street) are sup- portive of the possibility that persons with asthma may be more susceptible to adverse health effects (such as decrements in lung function and enhanced responses to allergens) related to such exposure. The Panel’s evaluation of the toxicologic data on the respiratory system regarding the effects of components of traffic- related air pollution was that such exposures result in mild acute inflammatory responses in healthy individuals and enhanced allergic responses in allergic asthmatics and animal models.
When the epidemiologic and toxicologic data were viewed together, the Panel noted that a case could be made that there are likely to be causal associations related to exposure to traffic- related air pollution and asthma exacerbation and some other respiratory symptoms. However, given the lack of a large body of toxicologic data based on human and animal exposures to real-world traffic scenarios, the Panel noted that it was hazard- ous to conclude that causality has been established at this time for all respiratory symptoms at all ages.
lUNG FUNctioN aND cHRoNic oBStRUctiVe pUlMoNaRY DiSeaSe
Changes in lung function are considered reliable markers of health that reflect the effects of endogenous and cumulative exposure to exogenous factors that might have adverse health consequences. Reduced lung function is strongly associated with future morbidity from a variety of causes and is a predic- tor of life expectancy (Hole et al. 1996); however, the relevance to health of small, short-term changes has not been assessed. The Panel considered lung function and chronic obstructive pulmonary disease (COPD) together in this review, because the principal criterion for the diagnosis of COPD is based on lung- function measures.
epidemiology
Lung Function in Children and Adults The studies reviewed were heterogeneous in their design, approach to exposure assessment, and lung-function measures. Given their limited comparability, the Panel concluded that the evidence is “sug- gestive but not sufficient” to infer a causal association between short- and long-term exposure to traffic-related pollution and decrements in lung function. However, in the case of long- term exposure, there was some coherence in the data, suggest- ing that (1) long-term exposure is associated with changes in lung function in adolescents and young adults; (2) lung-func- tion measures are lower in people who live in more polluted areas; and (3) changing residence to a less-polluted area in one study is associated with improvements in lung function (Burr et al. 2004). The first and second points are consistent with lon- ger-lasting effects on lung structure and/or function. The third point can be interpreted to indicate that some component of the apparent effects on lung function is reversible or is more the result of short-term exposure.
8
special report 17
ExECuTIvE SummARy
Chronic Obstructive Pulmonary Disease Because only two of the COPD studies fulfilled the criteria for inclusion in the review and their results were not consistent, the Panel concluded that there is “inadequate and insufficient” evidence for causal asso- ciations between exposure to traffic pollution and COPD.
toxicology
A very limited database of controlled human exposure has shown short-term reductions in forced expiratory volume in 1 second (FEV1) and increases in inflammation with exposure to traffic-related air pollution. However, the two end points have not been associated with each other. Virtually no data are avail- able from animal models. There are no studies of traffic-related air pollution and COPD.
While the epidemiology studies do provide suggestive evi- dence for chronic exposure effects on lung function in adoles- cents and young adults, there are too few toxicologic data to indicate what mechanisms underlie these observations. The aggregate epidemiologic and toxicologic evidence on chronic exposure to traffic-related air pollution and altered lung func- tion in older adults and the occurrence of COPD is too sparse to permit any inference with respect to causal association.
alleRGY
epidemiology
The 16 epidemiology studies on this outcome included in the review not only had to meet criteria for the quality of their exposure data but also had to report at least one of the follow- ing: (1) positive skin-prick testing for common aeroallergens; (2) serum-specific IgE to common aeroallergens; (3) a physician’s diagnosis of eczema or allergic rhinitis; or (4) use of question- naires on the history of symptoms of hay fever, seasonal runny nose, rhinitis or conjunctivitis, or itchy eyes. With a few incon- sistent exceptions, results based on the skin-prick test reactivity or allergen-specific IgE failed to show associations with any of the traffic-exposure surrogates. Inconsistent results with self- reported symptoms were also noted. The Panel concluded that there is “inadequate and insufficient” evidence to infer a causal association, or even a noncausal association, between exposure to traffic-related pollution and IgE-mediated allergies. Overall, the lack of consistency across epidemiology studies might have reflected a failure to identify susceptible subgroups.
toxicology
The Panel noted that the toxicology data provide strong mech- anistic evidence with respect to the diesel particle component of traffic-generated pollution and IgE-mediated allergic reactions and some evidence for NO2 and late-phase response to allergen. However, the epidemiology studies were inconsistent. The rel- evance of the toxicology studies (often by nasal instillation with diesel exhaust particles) to the actual manifestations of non- asthmatic allergic phenotypes (e.g., allergic rhinitis or conjunc- tivitis, eczema, serum-specific IgE, and evidence of sensitization to aeroallergens) could not be determined.
BiRtH oUtcoMeS
epidemiology
Although a considerable body of data from around the world has identified consistent associations between exposure to ambi- ent air pollution in general and various birth-outcome measures (low birth weight, small for gestational age, and perinatal mor- tality), only four studies of exposure to traffic-related pollution met the criteria for inclusion in this review. The small number of studies and their limited geographic coverage led the Panel to conclude that there is “inadequate and insufficient” evidence to infer causality.
toxicology
The toxicology studies reported effects on reproductive organs and sperm functionality in animals, but these outcomes were not evaluated in the epidemiology studies. Among the challenges in interpreting these results are the data limitations and the almost-universal use of very high exposure concentrations that have questionable relevance to actual ambient concentrations. Due to their lack of overlap, the epidemiology and toxicology studies on reproductive health and birth outcomes do not lend themselves to any overall synthesis.
caNceR
epidemiology
The Panel focused on general-population exposure studies and did not review the extensive epidemiologic literature on cancer from occupational exposure to traffic emission constituents (e.g., benzene and diesel exhaust). Among the studies reviewed, five were of childhood cancers (mainly leukemias, lymphomas, and cancers of the central nervous system), and four of adult can- cers (two of lung cancer, one of female breast cancer, and one of several cancers combined). Data on childhood cancers were inconclusive in terms of overall consistency and of specific can- cers. Too few data were available in adults. Overall the Panel concluded that the evidence was “inadequate and insufficient” to make inferences for causality between exposure to traffic pol- lution and cancer.
toxicology
The toxicologic research summarized included in vitro muta- genicity studies of exposure of cells to PM from traffic pollution, diesel or biodiesel exhaust, and organic components of some of these mixtures, as well as animal carcinogenicity studies after exposure to exhaust from diesel and gasoline-fueled engines. Although studies in cells demonstrating the capacity of DEP to induce DNA-strand breaks, base oxidation, and mutagenicity provide a possible mechanism for the induction of carcinogenic- ity by traffic-related pollution, the applicability of in vitro muta- genicity studies to human risk assessment has been questioned.
9
ExECuTIvE SummARy
special report 17
Animal studies have demonstrated the ability of high concentra- tions of exhaust components in both diesel and gasoline-fueled engines to cause tumors in animals. However, caution must be exercised in extrapolating these data to people exposed to much lower concentrations of pollutants, as seen in the epidemiology studies. Therefore, the Panel concluded that any statement that tries to relate the toxicologic to the epidemiologic data is prema- ture at this time.
OVERALL CONCLUSIONS
Studies have shown that traffic-related emissions affect ambient air quality on a wide range of spatial scales, from local roadsides and urban scales to broadly regional background scales. Based on a synthesis of the best available evidence, the Panel identi- fied an exposure zone within a range of up to 300 to 500 m from a major road as the area most highly affected by traffic emissions (the range reflects the variable influence of background pollution concentrations, meteorologic conditions, and season).
Surrogates for traffic-related exposure have played, and are likely to continue to play, a preeminent role in exposure assess- ments in epidemiology studies. The optimal selection of rel- evant surrogates (especially surrogates that are single chemicals) depends on accurate knowledge of the degree to which they rep- resent the chemical and physical properties of the actual primary traffic-pollution mixtures to which humans are exposed, which, in turn, depends on accurate knowledge of motor-vehicle–emissions composition and near-source transformation and dispersion. The Panel concluded that none of the pollutant surrogates (CO, NO2, UFP, EC, and benzene) is unique to emissions from motor vehi- cles. Among the surrogates based on traffic-exposure models, the question remains as to the extent to which the proximity model (i.e., the simple distance-to-road measures) should be employed in future epidemiology studies because it is particularly prone to yielding measures potentially containing extraneous information that can lead to the confounding of associations between health effects and exposure. In the Panel’s view, the hybrid model is the current optimal method of assigning exposures to primary traffic- related pollution.
Many aspects of the epidemiologic and toxicologic evidence relating adverse human health effects to exposure to primary traffic-generated air pollution remain incomplete. However, the Panel concluded that the evidence is sufficient to support a causal relationship between exposure to traffic-related air pol- lution and exacerbation of asthma. It also found suggestive evi- dence of a causal relationship with onset of childhood asthma, nonasthma respiratory symptoms, impaired lung function, total and cardiovascular mortality, and cardiovascular morbidity, although the data are not sufficient to fully support causality. For a number of other health outcomes, there was limited evidence of associations, but the data were either inadequate or insuffi- cient to draw firmer conclusions. The Panel’s conclusions have to be considered in the context of the progress made to reduce emissions from motor vehicles. Since the epidemiology studies
are based on past estimates of exposure from older vehicles, they may not provide an accurate guide to estimating health associa- tions in the future.
In light of the large number of people residing within 300 to 500 m of major roads, the Panel concludes that the sufficient and suggestive evidence for these health outcomes indicates that exposures to traffic-related pollution are likely to be of public health concern and deserve public attention. Although policy recommendations based on these conclusions are beyond the scope of this report, the Panel has tried to organize, summarize, and discuss the primary evidence in ways that will facilitate its usefulness to policy makers in the years ahead.
REFERENCES
Burr ML, Karani G, Davies B, Holmes BA, Williams KL. 2004. Effects on respiratory health of a reduction in air pollution from vehicle exhaust emissions. Occup Environ Med 61:212–218.
Hoffmann B, Moebus S, Möhlenkamp S, Stang A, Lehmann N, Dragano N, Schmermund A, Memmesheimer M, Mann K, Erbel R, Jöckel K-H. 2007. Residential exposure to traffic is associated with coronary atherosclerosis. Circulation 116:489–496.
Hoffmann B, Moebus S, Stang A, Beck EM, Dragano N, Möhlen- kamp S, Schmermund A, Memmesheimer M, Mann K, Erbel R, Jöckel KH. 2006. Residence close to high traffic and prevalence of coronary heart disease. Eur Heart J 27:2696–2702.
Hole D, Watt G, Davey-Smith G, Hart C, Gillis C, Hawthorne V. 1996. Impaired lung function and mortality risk in men and women: Findings from the Renfrew and Paisley prospective population. Br Med J (Clin Res Ed) 313:711–715.
Maynard D, Coull BA, Gryparis A, Schwartz J. 2007. Mortality risk associated with short-term exposure to traffic particles and sulfates. Environ Health Perspect 115:751–755.
Modig L, Järvholm B, Rönnmark E, Nyström L, Lundbäck B, Anders- son C, Forsberg B. 2006. Vehicle exhaust exposure in an incident case-control study of adult asthma. Eur Respir J 28:75–81.
NARSTO. 2005. Improving emission inventories for effective air quality management across North America. NARSTO 05-001. Pasco, Washington. Available from ftp://narsto.esd.ornl.gov/ pub/EI_Assessment/Improving_Emission_Index.pdf.
Rosenlund M, Berglind N, Pershagen G, Hallqvist J, Jonson T, Bellander T. 2006. Long-term exposure to urban air pollution and myocardial infarction. Epidemiology 17:383–390.
Tonne C, Melly S, Mittleman M, Coull BA, Goldberg R, Schwartz J. 2007. A case-control analysis of exposure to traffic and acute myocardial infarcion. Environ Health Perspect 115:53–57.
U.S. Department of Health and Human Services. 2004. The Health Consequences of Smoking: A Report of the Surgeon Gen- eral. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health. Atlanta, GA.
10
special report 17
EXECUTIVE SUMMARY
Ira Tager, Chair, Professor of Epidemiology, School of Public Health, University of California–Berkeley; member, HEI Research Committee
Kenneth Demerjian, Director, Atmospheric Sciences Research Center; Professor, Department of Earth and Atmospheric Sci- ences, State University of New York; member, HEI Research Committee
Mark Frampton, Professor of Medicine and Environmental Med- icine, University of Rochester School of Medicine and Dentistry
Michael Jerrett, Associate Professor, Division of Environmental Health Sciences, School of Public Health, University of Califor- nia–Berkeley
Frank Kelly, Professor of Environmental Health and Director of Environmental Research Group, King’s College, London, U.K.
Lester Kobzik, Professor of Environmental Health, Harvard School of Public Health
Nino Künzli, Professor of Social and Preventive Medicine, Pub- lic Health University of Basel, Institute of Social and Preventive Medicine at the Swiss Tropical Institute, Basel, Switzerland
Brian Leaderer, Susan Dwight Bliss Professor of Public Health, Division of Environmental Health Science, Yale University School of Public Health; former member, HEI Review Committee
Thomas Lumley, Associate Professor, Department of Biostatis- tics, University of Washington School of Public Health and Com- munity Medicine
Frederick W. Lurmann, Manager of Exposure Assessment and President Emeritus, Sonoma Technology, Inc.
Sylvia Richardson, Professor of Biostatistics, Department of Epidemiology and Public Health, Imperial College School of Medicine, London, U.K.; member, HEI Research Committee
Jonathan Samet, Director, Professor, and Flora L. Thornton Chair, Department of Preventive Medicine, Keck School of Medi- cine, University of Southern California
Michael Walsh, Consultant on vehicle emissions and fuels issues worldwide; former head of the U.S. EPA Office of Mobile Source Air Pollution Control
Maria Costantini, Health Effects Institute Principal Scientist, Project Coordinator
HEI Panel on the Health Effects of Traffic-Related Air Pollution. 2010. Traffic-Related Air Pollution: A Critical Review of the Literature on Emissions, Exposure, and Health Effects. HEI Special Report 17. © by the Health Effects Institute, Boston, Mass. The entire report is available at www.healtheffects.org or from HEI.
Health Effects Institute 101 Federal Street, Suite 500, Boston, MA 02110, USA
Phone: +1-617-488-2300. Fax: +1-617-488-2335. www.healtheffects.org.
Traffic Review Panel
11
memo3/Koslowsky_1997.pdf
APPLIED PSYCHOLOGY A N INTERNATIONAL REVIEW. 1997.46 (2). 153-173
Commuting Stress: Problems of Definition and Variable Identification
Meni Koslowsky Bar-llan University, Ramat Gan, Israel
Bien que les chercheurs reconnaisssent I’experience du changement comme source potentielle de stress, les rksultats concernant son impact sur des conduites variees individuelles ou organisationnelles ont ete peu concluants. Dans le present article, une revue des etudes qui examinkrent I’influence potentielle du stress de changement sur des mesures de contraintes psychologiques, physiologiques et comportementales, est proposke et relkve des contradictions. Un modde est dkcrit, liant des mesures objectives et subjectives de l’agent de stress, directement ou indirectement aux conduites. II est suggCrt qu’en incluant plusieurs moderateurs critiques, par exemple le contrBle, la predictibilite et I’urgence dans le temps, des predictions Cmanant de stimuli de stress de changement pourraient &tre mises en avant.
Although researchers recognise the commuting experience as a potential source of stress, findings regarding its impact on various individual and organisational outcomes have been inconclusive. In the present paper, a review of relevant studies that examined the potential influence of commuting stress on psychological, physiological, and behavioural strain measures is reported and inconsistencies noted. A model linking objective and subjective measures of the stressor directly and indirectly to outcome is described. I t is suggested that by including several critical moderators, such as control, predictability, and time urgency, predictions emanating from commuting stress stimuli can be enhanced.
INTRODUCTION
Research over the years has shown that the commuting experience may indeed be another source of work-related stressors (Novaco, Stokols, & Milanesi, 1990; Schaeffer, Street, Singer, & Baum, 1988). Typically, according to these investigators, a commuter confronted with any one of a wide range of stimuli associated with the trip between home and work (e.g.
Requests for reprints should be sent to Dr. Meni Koslowsky. Department of Psychology. Bar-Ilan University. Ramat Gan 52900. Israel.
0 1997 Jnternational Association of Applied Psychology
154 KOSLOWSKY
traffic congestion. noise. crowding. traffic signals, etc.) can be expccted to respond with increased anxiety, lowcr productivity, greater absenteeism, and even a greater likelihood for certain physical ailments and diseases (Abdel-Salam, Eyres, & Cleary, 1991; Novaco, Kliewer, & Broquet, 1991). Although. over the years, bivariate, and sometimes even multivariate relationships between commuting variables and outcome measures have been reported. the need for integrating some of this research has only recently been proposed (Koslowsky. Kluger, & Reich, 199.5). The purpose of the present paper is to summarise the literature and redefine some of the major components involved in the commuting process.
In other stress settings, both in the laboratory and the field, social psychologists have found environmental variables to be significant predictors of negative outcomes (Evans, 1982; Evans & Cohen, 1987; Stokols, 1992). Attempts to apply some of these concepts for understanding the effects of commuting produced mixed results (Koslowsky et al., 199.5: Novaco, Stokols, Campbell, & Stokols, 1979). For example, simple use of objective measures of the commuting experience, time or distance, individually or in combination, does not appear to provide adequate explanation or prediction of individual and organisational outcomes. Building on some of the concepts presented previously by Koslowsky et al. (199.5), the presentation here will focus on reviewing the literature and identifying some of the critical independent variables and moderators that can be posited as influencing the obscrved relationships. Formulating the commuting stress experience more precisely may have both theoretical and practical implications. It can help us understand the stressors that are at play and, in some cases, provide better coping mechanisms for expected strain reactions.
Some Previous Investigations in the Field
In addition to describing general attitudinal and emotional reactions to the trip between home and work (Baldassare & Katz, 1988), investigators in the field have attempted to identify the proper independent variable and link it with various outcome measures. In a study that examined various psychological outcomes of commuting stress. Novaco et al. (1990) found a low to moderate relationship between their definition of the commuting experience (a combination of time and distance) and certain indicators of residential satisfaction. Nevertheless, closer examination of the data showed that the picture was not clear. Although the objective measures of commuting was predictive of job satisfaction but not mood after arrival a t home, a subjective measure. by itself. was correlated with several other outcome measures but not with job satisfaction or residential satisfaction.
COMMUTING STRESS 155
Schaeffer et al. (1988) did not find any relationship between average travelling speed and measures of anxiety or mood. Interestingly, single- drivers, as compared to car-pool drivers, did have significantly higher scores on hostility and anxiety measures. In a study by Novaco et al. (1979), the authors tested the effects of travel time and distance on four measures of mood. They reported that subjects were more tense and nervous after long, slow trips but no differences be,tween groups were observed on sub-tests measuring irritability and impatience.
Physiological responses have also been examined in the literature. Evans and Carrere (1 991) reported a high degree of association between exposure to peak traffic conditions and elevation of certain urinary catecholamines. In a study comparing the impact of driving a car with that of being a passenger, Novaco and Sandeen (1992) reported that the blood presure of the latter was consistently lower than the former. Nevertheless, other hypotheses concerning the effects of ridesharing and gender, individually or as an interaction term, did not yield clear results.
Among the behavioural variables frequently studied in the same context as commuting are lateness and absence. It is expected that negative commuting experiences will probably result in increased levels of these withdrawal measures. Just as with findings from attitude studies, some researchers have confirmed this hypothesis; others have not. Using the survey responses of about 1500 workers, Leigh and Lust (1988) found that commuting long distances to work, among other variables, was negatively associated with lateness. Nicholson and Goodge (1976) also reported that lateness was negatively correlated with commuting distance but only for women. Women who lived closer to the plant were more likely to arrive late at work. The authors explained these results by suggesting that a short commute on a public transportation system, which is often the case among women workers, affords more frequent opportunities for minor lateness as compared with infrequent schedules over longer distances.
Findings relating the commuting experience with absenteeism were also inconsistent. When a group of frequently absent subjects was compared to a control group, more than twice as many of the former were found to live an hour or more from work as compared to the latter (Knox, 1961). Martin (1971) reported that for male workers the relationship between absenteeism and distance travelled was significant, but for women the association was not significant. Martin suggested that gender may be a moderator variable, with women more likely to find jobs closer to home and men willing to live further away and travel greater distances to work. In contrast, Pop and Belohlav (1982) reported no significant relationship between commuting and absence for a male sample. Finally, Novaco et al. (1990) found that subjects with long trips over a long period of time had the largest number of absence-illness days. However, when the analysis was controlled for covariates such as age,
156 KOSLOWSKY
smoking, weight, and alcohol consumption the relationship was no longer linear.
From the various studies. i t appears that the commuting experience may very well have an impact on psychological, physiological, and behavioural outcomes, but empirical findings. so far, have been inconsistent. In an attempt to better understand the linkages involved here and to develop more accurate predictions of outcome. this paper will describe some new w a y of looking at the critical variables in the commuting process.
A MODEL OF COMMUTING STRESS
Based on inferences from various studies, a model for describing the impact of commuting on outcome is suggested. I t is adapted from Koslowsky et al. (1995) and Novaco et al. (1990) and contains several critical variables of the process. As can be seen in Fig. 1, at least three major stages can be identified in the commuting stress paradigm. Objective impedance in stage 1. subjective impedance in stage 2, and outcomes (physiological, psychological. and behavioural) in stage 3. Moderators. which will be described later. appear between stages 1 and 2 and between stages 2 and 3. In each case. the small circles represent the interaction term between the antecedent variable (objective impedance or subjective impedance) and the indicated moderator. One mediator. subjective impedance. linking objective impedance and outcomes, is postulated.
The model contains several moderators placed at various points and should be considered only as a suggested formulation. The placement of the moderators is based on empirical findings or. in some cases, their position is derived from my understanding of the stress literature and the commuting process, Until now. investigators have not included all these variables in one model and it is possible that moderators may change position after more data are collected. Only by testing the model in the field can the correct links and positions be determined.
The directions (depicted by arrows) are expressed as linkages in a latent factor model with the third stage consisting of outcomes or a weighted combination of the three strain measures depicted in the figure (see Loehlin, 1992. for a discussion of this and related modelling issues). For simplicity purposes, we have denoted the objective and subjective impedance measures as unitary indices. However, each of these independent variables can be viewed as latent factors consisting of a weighted combination of their respective components. In the following few pages, I will present some of the arguments in support of the model.
COMMUTING STRESS 157
FIG. 1. A three-stage model of commuting stress.
Stressor and Moderator Variables in the Commuting Experience
One possible approach for explaining some of the contradictory results is to posit other variables than those commonly studied in the past. It may be necessary in this regard to change the stressor or independent variable as well as to suggest moderators that distinguish between groups where the effect is observed and where it is not. If the commuting experience is seen as the beginning of a process that leads to negative consequences, then such intervening variables would serve as critical links between the stressor and outcome measure.
Except for citing the strain or outcome measures used in different commuting studies, this paper will not deal with defining outcome variables or the associations among them. For example, a decrease in performance or an increased likelihood of organisational withdrawal (e.g. lateness, absence, and turnover) can be considered as typical behavioural reactions to the commuting experience. Similarly, blood pressure or job satisfaction as measures of physiological and psychological responses, respectively, have also been examined in the commuting literature. The exact interrelation- ships, however, are not clear (see, for example, Campion, 1991; Hanisch & Hulin, 1990) and, as the presentation here focuses more on the causal, rather
158 KOSLOWSKY
than effect. variables. this aspect of the model will not be expanded. Nevertheless. it is evident that any formulation proposed here is not complete and further work on identifying all relevant associations is required before the entire network of links is known.
Defining the independent Variable
The time i t takes to go from home to work or the distance traversed during this trip was the first operational definition of the commuting experience used in applied research. It should be noted that the time from work to home was rarely considered by researchers. The latter time may indeed be different from the former one, as going to work takes place in a smaller timeframe than the journey back home. People tend to leave work during a wider time interval ranging from the late afternoon to early evening period (Koslowsky, in prep).
Martin (1971) cited data to support the practical equivalence of the two terms (time and distance), particularly in studies of public transportation near the worker's home. According to the author, the negative impact of commuting could be traced to these independent variables. Novaco et al. (1979) reported a correlation of .93 between the two measures, indicating that the two may very well be interchangeable. Nevertheless. there is probably information available from each of these variables that is unique; for instance. i t is entirely plausible that time varies as a result of other factors such as traffic congestion. A comprehensive understanding of the stressor, therefore, must take into consideration both measures. This reconceptuali- sation of the independent variable is one of the primary contributions of Nevaco and his colleagues.
Novaco et al. (1979) suggested the concept of "commute impedance". Impedance was defined as a behavioural restraint on movement or goal attainment (Novaco et a]., 1991). The authors contended that impedance includes anything that frustrates the goal of arriving at a given time at a particular destination-for example, distance, slow speed, or traffic congestion. It is at a maximum when long distances are covered slowly and at a minimum when short distances are covered quickly. Determining the best way to measure impedance and how it is linked with outcome measures are among the major current issues in the field.
At first glance, i t would appear that average speed should be the ideal measure. If it takes longer to get to work (or to return home), impedance would be higher with strain as a potential outcome. One of the first groups of researchers to use average speed as a measure to predict outcomes from the commuting experience was Schaeffer et al. (1988). Although some of the findings regarding psychological effects (e.g. mood) were ambiguous, clearer results were reported with blood pressure and cognitive
COMMUTING STRESS 159
performance. The former increased and the latter decreased as a function of impedance. Overall, the work of Schaeffer et al. (1988) provided impetus for researchers to continue with the strategy of defining impedance.
In particular, after a close examination of the data reported by Schaeffer et al., Novaco and colleagues felt that a different approach was required. Average speed, they argued, was probably confounded with many other extraneous factors and its application and proper testing in the field quite difficult. For example, a commuter driving at 35mph is not necessarily experiencing more impedance than another driver moving at an average speed of 50mph. The former may be using only local streets, travelling at the maximum permissible speed, and experiencing no “obstacles” along the way, whereas the latter may be driving on a superhighway (on a German autobahn, this driver is probably only crawling), and may report many “negative stimuli” on the trip to work. Interestingly, Novaco et al. (1990, p.234) found that average speed increases as an artifact of commuting distance, a further indication that speed is not a function of impedance but of other extraneous variables. A need to identify some other component of the stressor variable was necessary.
Objective and Subjective Impedance The work stress literature provided the additional stimulus needed to understand the commuting process. In their discussion of correlates and causes of strain in the workplace, Frese (1985; Frese & Zapf, 1988) made a distinction between objective and subjective indicators of the stressful stimulus. The former refer to stimuli that are not related to one’s perceptions, whereas the latter involve cognitive and emotional processing by the target person. Using both types of variables in a model, two major sets of linkages can be proposed: (1) both components of the stressful stimulus directly affect outcome (without mediating effects), or (2) the objective components affect the subjective ones which, in turn, influence outcome variables. In the latter case, a direct impact of the objective components may still be said to exist but the main influence is indirect and through the subjective components, which are acting as mediators.
The need to include objective and subjective characteristics in stress models has been cited by several investigators. Kasl(l987) and Fox, Dwyer, and Ganster (1991) have reported findings that support the notion that only after including the subjective measures is the error in strain prediction reduced. For example, Fox et al., in a study of nurses, found that the objective stressors were related to physiological responses and the subjective indicators were associated with affective outcomes. The authors concluded that both variables are required because the objective measures did not predict affect and the subjective measures did not predict physiological responses.
160 KOSLOWSKY
Recently investigators in the commuting field have also begun to distinguish among types of stressors. Specifically, some creative applications by the Novaco group to the commuting field resulted in the distinction between objective and subjective stressors. Novaco et al. (1979, 1990) distinguished between objective impedance, a combination of time and distance between home and work, and the subjective components, obtained from self-report data requiring respondents to describe how various stimuli (traffic lights, stop signs, etc.) affected their trip to work.
The theoretical contribution by these researchers is a reformulation of the commuting experience. As with other areas of stress, perceptions, as well as the objective dimension. must be considered in understanding the effects of the process. By dividing it into separate dimensions which may overlap to some extent, but remain essentially independent. Novaco et al. (1990) offered a new definition for the independent variable.
In operationalising the concept of subjective impedance, Novaco et al. (1990) identified four components: evening commute congestion (e.g. self-report of times brakes were deemed necessary, etc.), aversiveness of travel (e.g. ratings of the commute in terms of slow-fast, stop-and-go, etc.), morning commute congestion (e.g. necessity to apply brakes on the way to work, etc.), and surface street constraints (travel speed reduced by stop signs, etc.). Novaco et al.‘s observations basically involved automobile commuters and the perceived stimuli were viewed as potentially significant stressors when measured in terms of degree of commute impedance. These characteristics can be modified so as to include users of public transportation; in such a case, commuting stress can be perceived as a function of the number of stages in the commute (Taylor & Pocock, 1972), the crowded conditions of the commute (Aiello, Derisi, Epstein, & Karlin, 1977), and the complexity of the journey to work (Knox, 1961).
Recent work by Koslowsky (in prep) has expanded on this concept by examining 20 items that describe various aspects of the commute. These items were derived mostly from the literature, particularly the Novaco et al. (1990) scale and from pretesting several new items on a sample of subjects. Four categories of subjective commuting stressors or indicators were derived from a factor analysis of a 20-item scale followed by a varimax rotation. The factors were labelled as “bus conditions”, “driver behaviour”, “road conditions”, and “commuting environment“. Construct validity was also examined by correlating the subjective factors as well as objective measures of the commute with a self-report measure of strain. Findings showed that Factor 3 (road conditions) from the subjective scale was the best predictor of the outcome measure, and considerably superior to a rate measure (distance/time).
In all likelihood, identifying the “proper” independent variable in the commuting process can reduce some of the error variance associated with
COMMUTING STRESS 161
outcome prediction. Using techniques such as structural equations allows for examination of direct and indirect effects. Also, as will be expanded on later, researchers must try to differentiate between the objective and subjective data. If possible, there is some merit in obtaining the values of the independent variable from a source outside the subject. For example, a stopwatch, or other timing device attached to the commuter, can be used for the measure. Traditional methods of obtaining time or distance measures may actually yield only another set of subjective data, biasing the results and not permitting a proper examination of the actual effects of objective impedance.
For the subjective scale this problem does not exist. Here, we are actually interested in perceptions. Thus, one can argue that it isn’t the absolute number of traffic lights or the number of times the brake is applied that is important but how the commuter perceived their impact on his or her driving experience. A commuting experience with few actual traffic lights but perceived as having reduced the driver’s speed is what we are really looking for in a subjective scale, i.e. the perception, not necessarily the objective conditions. At this stage in commuting research, I would recommend collecting data from different sources including self-report as well as other sources for the objective and subjective components of the independent variable in commuting stress, and testing predictions from each separately and in combination.
Although this paper has emphasised the fact that, generally, subjective impedance mediates between the objective characteristics of the commute and outcome, the model, as seen in Fig. 1, does include a direct impact between the latter pair of variables. Such a hypothesised link is consistent with the findings by Novaco et al. (1991) who reported that physical impedance had a direct impact on certain dependent measures such as residential satisfaction. In most of the other analyses, however, the relationship between objective characteristics and outcome disappeared when subjective characteristics were entered into the equation. Therefore, I suggest that both direct and indirect links exist but the latter ones are considerably more important in modelling the commuting experience. Using structural equation techniques, it is possible to test this formulation and provide specific weights associated with each of the links.
MODERATORS In understanding the new formulation, the role of moderators is critical. As described by Baron and Kenny (1986), these so-called third variables interact with an independent variable to influence an outcome that would not be predicted by each one of the variables individually. Discovery of a significant moderator in an analysis will, generally, result in new inferences
162 KOSLOWSKY
and conclusions. In this section, I will review some moderators that have been discussed previously in other stress contexts and present some new ideas for commuter stress research.
Gender One of the variables that appears to act as a moderator in commuting is gender. In Fig. 1, i t appears between objective impedance and outcome. Several researchers have supported this placement for gender. For example, Nicholson and Goodge (1976) found that commuting distance and lateness are positively correlated for women but not for men. Similarly, according to Novaco et al. (1990), the interaction between gender and physical impedance was found to have a significant relationship with five of six measures of residential satisfaction. In particular. women with high levels of physical impedance on their journey to work were more likely to feel discontent at home. Although this finding of negative affect for women was consistent with expectations, the authors did not provide a satisfying interpretation for the association. I t is likely that gender acts as a surrogate for other constructs such as “activities en route”. Women are possibly doing more on their trip to work (shopping, picking up the children, etc.) than men. To test this explanation, further work in this area seems warranted.
Control
Although the study of commuting stress can be considered part of the general topic of environmenal stress, it contains many features that are unique to this area of research. Many of the inferences on the impact of control in work stress and its role as a moderator are derived from laboratory work where control was manipulated in some fashion by the experimenter (Jackson. 1989; Thomson, 1981). In the Figure, control has been placed as a moderator between objective impedance and outcomes.
In the traditional social/environmental context the origin of control is outside the individual; however, with commuting stress the origin, is, at least, partially internal. Thus, the length of the trip and the time it takes to complete the journey are potentially controlled by the driver or passenger. He or she may decide to switch jobs if the trip to work takes too long, or an alternative highway may be used if the usual one is blocked. To the extent that the journey to work is no: in the corn-liter’s control, the potential for strain response increases.
Similarly, mode of transportation, i.e. whether one drives to work or uses public transportation, is another potential moderator that may be considered as an aspect of control (in the Figure, it is considered as part of control). For the car driver, control (or the lack of i t ) is a major issue and potential contributor to strain as a result of the commute, whereas for a bus
COMMUTING STRESS 163
or train passenger, control is relinquished when boarding and does not play a very important role in producing strain. Findings by Koslowsky and Krausz (1993) have indicated that the car driver is more likely to react to the stressful stimuli. It appears that on a crowded road the lack of control experienced by a driver who expects to have control about many aspects of the journey to work is a potential source of stress. Moreover, real control does not need to be exercised; if the person feels that there is a possibility of control, that perception alone may be sufficient to lower strain responses (Landy, 1992).
Evans & Cohen (1987) and Schaeffer et al. (1988) observed that control acted as a moderator between the independent variable and outcome. Thus, a stressor such as the length of the commute may be said to interact with control so that under conditions of a short trip, the commuter is not even aware of the extent to which he or she controls the environment. However, when the trip gets to be very long, the commuter may suddenly become aware of a lack of control and the negative consequences are heightened.
From a somewhat different perspective, an issue raised by Murphy (1988) may be particularly relevant here. He argues (p.324) that from the available data it seems that it is less stressful never to have had control over a situation than to have had it and then to have lost it. This fact can be used to explain why bus or train passengers may not be under the same stressor stimuli. They have never had, nor even assumed that they would have, much control over the ride to work. Outside of deciding when and where to board, the journey is under control of the bus driver or conductor, not the commuter. Control is not expected and, therefore, obstacles encountered along the way are not necessarily stressful.
Interestingly, in a study of the effects of traffic congestion on bus drivers, Evans and Carrere (1991) reported that perceived job control acted as a mediator or intervening construct. A recent work by Thomas and Ganster (1995; p.13) supported this contention; specifically, perception of control served as a mediator between supportive organisational policies in response to work-family conflict and strain measures consisting of physiological responses and job satisfaction. As the bulk of the research considers control as a moderator, this was the approach taken in the model. Nevertheless, field research on this controversy is needed to decide this issue definitively.
Locus of Conrrof and Commuting Strcss. Unlike the control variable which describes an actual or perceived relationship between person and environment, locus of control is a personafiry measure which reveals whether an individual believes that what happens to him or her is primarily determined by external or internal factors. The role of locus of control in studies of the stress-strain process has a somewhat chequered past with inconsistent findings (Gulian et al., 1989; Montag & Comrey, 1987; Novaco
164 KOSLOWSKY
et al., 1979). For example, Novaco et al. (1979) examined locus of control as a potential moderator of the relationship between impedance and outcomes. Although significant interactions in the hypothesised direction were obtained in predicting some dependent measures (e.g. task performance), for other outcomes measures (e.g. mood indices) significant findings opposite to those expected were found. According to the present model, it is expected that those with a greater sense of internal control will seek out alternative paths to avoid traffic or make the trip along the most advantageous route and reduce strain responses. As such an investigation has yet to be carried out, the potentiai role of locus of control in the commuting process remains ambiguous and has been left out of the Figure.
Sense of Motion and Conrrol. Another aspect of control that may also serve as a moderator on the journey to work is the degree to which the commuter experiences a sense of motion during the trip (in the Figure it should be considered as another control variable). The driver of a vehicle that is moving at an acceptable pace generally perceives some feeling of control over the commuting experience. When stuck in traffic, the reaction to stressors may be moderated by the extent to which the commuter feels that control has been lost. The feeling of motion or lack of it when stuck in a traffic jam probably plagues the car driver more than the user of mass transportation. The driver who is stuck in one place and cannot easily extricate himself or herself may respond with one of several different strain reactions. If the driver moves towards some goal even if it is not in the most optimal fashion, then stress stimuli are reduced. This phenomenon manifests itself in the case where a driver takes back or side streets instead of crowded main roads to work. With the former, motion is maintained at the cost of travelling longer distances.
A particularly good illustration of the application of motion to relax people is the use of winding waiting lines at an amusement or theme park such as Disneyworld. A visitor who is told that the expected wait for a specific attraction is one hour, for example, is likely to experience some type of psychological strain reaction such as frustration. However, when the line moves, even if it is mostly sideways rather than straight ahead towards the entrance to the attraction. strain reactions may be reduced.
The perception of motion towards a goal generates an impression to a participant or an observer that the goal will eventually be reached even if somewhat delayed. However, when the driver does not experience motion, there is a fear (perhaps unreasonable or irrational) that the goal (the workplace or home) will not be attained. This latter situation is likely to cause negative outcomes.
A bus passenger. especially in cities where special bus lanes exist, or a train commuter, is generally not subject to these feelings of frustration. The
COMMUTING STRESS 165
movement is at a pace known to the commuter and even if elapsed time is relatively great (such as the commutes on the railroads near large metropolitan areas such as New York City), they are preferred by the commuter. When the motion on public transportation sysems becomes problematic such that time of arrival becomes a variable, then the commuter may very well abandon this mode, too.
Predictability
Seligman and Miller (1979) suggested that when circumstances do not permit control of the environment then other variables may replace it in determining the existence of the stressful stimulus. Thus, people who cannot control their environment may be satisfied with being able to predict it. Schulz (1976) showed that predictability of a positive event as opposed to its random appearance can be as useful as real control in having a positive impact on health outcomes of elderly people. In Fig. 1, predictability has been placed in two locations: (1) between objective impedance and outcomes and (2) between objective and subjective impedance. Consistent with this contention, G.W. Evans (personal communication, 26 October, 1995) has argued that an interaction between predictability and some objective characteristic of the commute may very well exist.
This aspect of predictability (or lack of it) has ramifications in many actual driving situations and is well illustrated in the commute facing the typical urban dweller. His or her commute into the city is often unpredictable and what may be a short commute on a particular day can turn into a nightmare the next. A difference of an hour or more between a good and bad day is not unheard of in many an individual’s commuting experience. Such variance prevents an individual from planning his or her arrival time at home. Even worse, perhaps, it does not permit the commuter to relax during the trip. The potential commuting delays encountered may raise his or her awareness and sensitivity to all sorts of stressors in the environment. Many of these stimuli will be perceived as contributing to the potential delayed arrival time. Although the individual may control various aspects of the commute including the decision to take a train or car, what time to start out from home, even what highways or roads to use, unpredictability of arrival time may still exist.
Thus, the feature of commuting unpredictability may very well influence the subjective impedance reported by the commuter. Kluger (1 992) reports that an unpredictable commute may contribute to perceptions of fear or lack of enjoyment with the commuting experience. In the present model, this would be consistent with the placement of a moderator between objective and subjective impedance. In other words, the effects of a commute that has some of the objective characteristics of being stressful (i.e. long distance and
166 KOSLOWSKY
excessive time) and is unpredictable will lead to much greater subjective impedance than would be predicted from either variable alone.
In addition, predictability also seems to moderate the relationship between objective impedance and outcome. In this regard, Kluger (1992) has argued that i t is possible that the actual independent variable in commuting research is unpredictability. Although his data did not refute this contention. i t also did not support it. A measure of unpredictability has recently been found by Koslowsky (in prep) to correlate significantly with outcome and interact with commuting time but was inferior to a measure of subjective impedance in explaining outcome variance.
As the data have supported both roles for unpredictability, the present model places this variable in two positions. Investigations to clarify whether this is a correct assumption and the determination of the relative importance of each interaction are needed.
Finally. several ways of measuring unpredictability can be suggested. By asking the commuter how long it takes to get to work on a bad day, a good day. and on an average day, several computations can lead to the desired measures. These questions can also be repeated for the drive from home to work. The first index is based on work reported by Kluger (1992). For each respondent, a mean and standard deviation for the response for each question was calculated. The within standard deviation of time was one index of predictability: the higher the standard deviation, the lower the predictability. A simpler method involves determining the difference between a bad and good day, separately, for each part of the trip and, also. the mean for both directions. In addition, a ratio measure using the average day as the denominator can be used to calculate fraction of time above a standard indicator. Finally, a questionnaire item asking the commuter to describe in a particular month how often the trip to work takes less than a half hour, between a half hour and an hour, one to two hours, and more than two hours can be recorded. The appropriate standard deviation formula that includes internal frequencies can then be applied. Adjustments of the intervals in specific cases can be made, if necessary.
Time Urgency and Time Management
Recently, investigators have isolated a new aspect of personality that relates an individual’s perspective about time as a potentially meaningful predictor of a range of dependent variables in organisational settings (Schriber & Gutek, 1987). Of particular concern here is the element of strain responses resulting from time concerns. As Landy, Rastegary. Thayer, and Colvin state (1991. p.655), it is clear that for organisational researchers “the role of time in studies of stress in the workplace is axiomatic”. Its influence can be viewed as double-edged. On the one hand, an individual’s response to this demand is
COMMUTING STRESS 167
a function of disposition or personality. On the other hand, the environment or job demands often require an individual to consider the element of time or use it more efficiently, even if this is not the natural tendency of the employee. For the present purposes, the first aspect has been labelled time urgency and the second one, time management. Both are expected to play a role in an individual’s reaction to the commuting experience (see Fig. 1).
It may very well be that time urgency and time management are the most critical moderator variables in a commuting model. In the model as depicted in Fig. 1, time urgency is posited as interacting with the objective impedance scale and influencing the quality of the subjective response. Thus, only for time-urgent individuals is subjective stress experienced. Awareness of time, and its importance, is transparent in an individual who is constantly looking at his or her watch during the trip tn work, who fiddles with the radio to pick up the news at the top of the hour, and who frequently asks a fellow passenger on the bus or train what is the exact time-such an individual can be described as time-aware. This commuter, if delayed beyond a reasonable time (which, itself, may be subject-dependent), becomes susceptible to negative perceptions about the commute. For the high time-urgent types, the whole commuting experience with its countless time deadlines or pressures such as arriving at the train station in time, making the right connection at a specified minute and hour, is magnified and experienced as more stressful than those who are low time-urgent types.
The recent work by Landy et al. and Wright, McGurdy, and Rogoll(l992) have provided us with two measures of time urgency. Although they overlap to some extent theoretically, they tap somewhat different aspects of the variable. The items are also presented in different formats. Research by the author is presently under way to determine which scale best enhances prediction in a commuting stress paradigm.
Furthermore, as can be seen in Fig. 1, it is expected that subjective stress and time management interact and affect the outcome responses. The contiguity between time management and outcome is consistent with some of the previous research in the literature. For example, in a study by Schriber and Gutek (1987), the time variable was considered as part of the culture of an organisation. The authors argued that organisational outcomes such as satisfaction, withdrawal, and productivity may all be a function of the match between an individual’s use of time and the organisation’s demands as perceived by the individual.
Several controlled studies have shown that time management can be taught, is amenable to intensive training programmes, and may have positive impact. For example, King, Winett, & Lovett (1986) analysed responses of subjects who had undergone time-management instruction and found that they were better able to cope with stressful stimuli. Such subjects spent more time in stress-reducing and fun type behaviours with a concomitant
168 KOSLOWSKY
reduction in negative outcomes. In a second study, Macan, Shahani, Dipboye, and Phillips (1990) reported performance, psychological, and physiological benefits for subjects who had undergone time-management training. Similar studies may need to be done on commuters. especially for subjects who report subjective impedance, so as to ascertain what benefits can be expected.
Morning and Evening Commutes
As indicated earlier, there are few work-related activities that are more dependent on time and schedules than the trip to and from work. I would like to suggest that the journey home from work may be fraught with greater sense of stressful stimuli than the commute to work. The trip in the other direction, i.e. from home to work, may be considered by many workers as part of the workday and if one gets stuck in a traffic jam or a train breakdown, i t can be rationalised as "company" time. However. on the way home, a person may view the delay as a loss to him or her personally. Exccpt for the measure of evening congestion discussed by Novaco et al. (1991) in their concept of subjective impedance. researchers have not compared the qualitative and quantitative differences in strain between the evening and morning commute. As empirical data supporting (or negating) the conjecture of the impact of commute direction on outcomes has not been collected, the variable was not included in the model. Presently, a study, under the author's supervision. comparing the effects of the trip to work with the trip home from w x k is under way, and hopefully, some more clearly defined hypotheses will be available in the near future.
CONCLUSIONS
The present paper presents a model for understanding the relationship between commuting stress. moderators, mediators. and outcomes. It seems that only by positing a formulation that contains these various elements can we expect to explain the impact of stress on our working and private lives. Nevertheless, the model as described here should be viewed as a tentative formulation based on previous research and educated guesses. For example, empirical data may show that moderators exist at different points than depicted here or that they do not have any role at all in commuting stress. Research designs that allow for significance testing and the examination of various paths are needed to determine the actual links and the positions of the variables in the Figure.
Several measurement issues must be considered before the true relationships in the field can be known. Probably the most critical one relates to the definition of each of the variables. As was shown previously, there is no consensus as to what is the proper measure ot the commuting stressor. I
COMMUTING STRESS 169
have suggested one approach for its examination but empirical data in support of this formulation does not yet exist.
Another issue that may be appropriate here is the time lag between measures and how this affects our findings. For example, when is the consequence measure taken? Is i t immediately after arriving at work or after coming home? Is there a slow and deliberate unwinding effect that should be gauged in the area of commuting research? Thus, there may be an immediate effect of commuting, but after a critical period of time has passed other stressors at work or at home begin to play a more important role. Systematic investigation of this issue is definitely warranted.
A major bias in stress-strain research is the reliance on self-report data for determining objective stress, subjective stress, organisational, or personal outcomes. It has been argued in the psychological literature that relationships among variables may be artificially raised if the researcher uses only one source for all the study variables (Crampton & Wagner, 1994). This has occurred in several studies in the commuting stress field where even the physical impedance information was collected exclusively from participants. In order to minimise artificial inflation of the findings, it is suggested that some data, at the very least, be collected through sources other than the subject, including clock or odometer readings. The addition of physiological and/or health indicators as outcome measures to the usual self-report scales €or objective and subjective impedance is another suitable way to make the study multi-source.
Distortion in stress-strain studies can also be attributed to the influence of negative affectivity. According to Clark and Watson (1991), negative affectivity is a cognitive set that may underlie responses in various social and health phenomena. Researchers in the commuting field should consider partialling out this variable before doing any type of analysis between self-report stress and strain measures (Burke, Brief, & George, 1993).
Applications of the Model The suggested model may have several practical applications for the individual and organisation. By gaining a better understanding of the links that produce subjective stress and other negative outcomes during and after the commute, it may be possible for suitable forms of intervention to affect or change the process. Time management can be considered as a specific coping strategy allowing the individual to control several of the time-related components of the stressor. Thus, in the model we have included time management techniques as a potential moderator. As previous studies have indicated a role for such techniques in reducing stress, its judicious use may help to mitigate or alleviate some of the negative consequences of commuting.
170 KOSLOWSKY
Similarly, methods to increase control or predictability may affect strain responses. If the commuter can be made aware of special situations that exist on the way to work, such as traffic accidents or a closed road. both perception of control and predictability could be enhanced. It may very well be that by tuning in to a radio station in the morning (in many communities today special radio frequences in the morning and evening report on traffic patterns and even make recommendations on alternative routes) the expected negative effects of commuting can be ameliorated.
From an organisational perspective. reducing the negative impact of stressful stimuli may improve attitudes, reduce the probability of withdrawal behaviour. and increase performance. All of these variables have been shown to be potential outcomes of stress (Kahn & Byosiere. 1991). As commuting can be viewed as simply another source of stress, i t would seem that some of the findings from the general I/O stress literature are relevant here. too. The model in the present study allows management to identify in which type of individual and under what circumstances strain response can be expected to increase.
Although the Figure describes links in a model of the individual's commuter experience. the aggregate effect across all employees is the critical indicator for the organisation. As monetary and psychological costs are associated with new policies for dealing with commuting-related problems. changes must be designed and implemented so as maximise overall productivity and minimise negative individual consequences. In the past few years, management in many companies have instituted innovations that were. at least in part, a response to deteriorating commuting experiences. For example. flexitime, telecommuting, and subsidised car- pooling are techniques for making the commute easier or. in the case of telecommuting, eliminating it altogether (see Koslowsky et al., 1995, for a full discussion of these various techniques). Intelligent use of the model will permit the introduction of strategies and methods of intervention that help both the individual and the organisation.
Future Research
Following the basic premise and terminology set down by the Novaco et al. (1990,1991) research group, i t appears that negative consequences from the commuting experience result from both objective impedance and subjective impedance. In addition, variables such as control or time-urgency and subjective impedance are posited as moderator and mediator variables, respectively. Obviously, such research is only a first step in explaining the impact of commuting on personal and organisational outcomes. In order to properly test some of the relevant hypotheses in the model, multivariate techniques that consider the concurrent effects of the various elements are
COMMUTING STRESS 171
required. From an empirical, measurement perspective, it is recognised that it is difficult to examine simultaneously all the independent and moderator variables mentioned here. Nevertheless, examining parts of the model, building on it by rejecting or accepting certain links, and then continuing to other components, appears to be what is needed at this stage of the field's development.
Manuscript received May 1995 Revised manuscript received May 1996
REFERENCES Abdel-Salam. A,. Eyres, K.S.. & Cleary. J. (1991). Drivers' elbow: A cause of ulnar
neuropathy. Journal of Hand Surgery [Br]. 16(4). 43U37 . Aiello, J.R.. Derisi. D.T., Epstein. Y.M., & Karlin. R.A. (1977). Crowding and role of
interpersonal distance preference. Sociometry, 40.271-282. Baldassare. M.. & Katz. C. (1988). Orange Counry survey: 1988 Final reporr. lrvine Public
Policy Research Organization: University of California. Baron. R.M., & Kenny. D.A. (1986). The moderator-mediator variable distinction in social
psychological research: Conceptual. strategic, and statistical considerations. Journal of Personality and Social Psychology. 51. 1173-1 182.
Burke. M.J., Brief, A.P.. & George, J.M. (1993). The role of negative affectivity in understanding relations between self-reports of stressors and strains: A comment on the applied psychology literature. Joitrnal of Applied Psychology, 78,402472.
Campion. M. (1991). Meaning and measurement of turnover: Comparison of alternative measures and recommendations for research. Journal of Applied Psychology, 76.199-21 2.
Clark. L.A.. & Watson, D. (1991). General affective dispositions in physical and psychological health. In C.R. Snyder & D.R. Forsyth (Eds.), Handbook ofsocialand clinical psychology (pp.221-245). New York: Pergamon Press.
Crampton. S.M.. & Wagner, J.A. 111. (1994). Percept-percept inflation in micro- organizational research: An investigation of prevalence and effect. Journal of Applied Psychology, 79.67-76.
Evans, G.W. (1982). Environmental stress. London: Cambridge University Press. Evans, G.W., & Carrere. S. (1991). Traffic congestion. perceived control. and psycho-
physiological stress among urban bus drivers. Journal of Applied Psychology, 76.658463, Evans. G.W.. & Cohen. S. (1987). Environmental stress. In D. Stokols & I. Altman (Eds.).
Handbook of environmental psychology (pp.571410). New York: Wiley. Fox. M.L.. Dwyer, D.J.. & Ganster, D.C. (1991). Stress and control among nurses: Effects on
physiological outcomes. Proceedings of the Narional Academy of Management (pp.267- 271). Miami. F L National Academy of Management.
Frese. M. (1985). Stress at work and psychosomatic complaints: A causal interpretation. Journal of Applied Psychology. 70,314-328.
Frese. M., & Zapf. D. (1988). Methodological issues in the study of work stress: Objective vs. subjective measurement of work stress and the question of longitudinal studies. In C.L. Cooper & R. Payne (Eds.). Causes, coping, and consequences of stress at work. Chichester. UK: Wiley.
Gulian. E., Mathews. G.. Glendon. A.I., Davies, D.R.. & Debney. L.M. (1989). Dimensions of driver stress. Ergonomics. 32. 585-602.
Hanisch, K.A., & H u h , C.L. (1990). Job attitudes and organizational withdrawal: An examination of retirement and other voluntary withdrawal behaviors. Journalof Vocational Behavior. 39, 110-128.
172 KOSLOWSKY
Jackson, S.E. (1989). Doesjobcontrol control jobstress. In S.S. Sauter. J.J. Hurrell Jr. & C.L. Cooper (Eds.). Job control and worker healrh (pp.25-53). Chichester, UK: Wiley.
Kahn. R.L.. & Byosiere. M. (1991). Stress in organizations. In M. Dunnette (Ed.). Handbook of indusrrial and orgnni:arional psychology (pp.571450). Palo Alto. CA: Consulting Psychologists Press.
Kasl. S.V. (1987). Methodologies in stress and health: Past difficulties. present dilemmas, future directions. In S.V. Kasl & C.L. Cooper (Eds.). Srress and healrh: Issues in research methodology (pp.119-146). New York: Wiley.
King. A.C.. Winett, R.A.. & Lovett, S.B. (1986). Enhancing coping behaviors in at-risk populations: The effects of time management instruction and social support in women from dual-earner families. Behavior Therapy. 17. 57-66.
Kluger. A. (1992. November). Comniure predicrabiliry and strain. Paper presented at the second APAiNlOSH conference. Stress in the 90s. Washington. DC.
Knox. J.B. (1961). Absenteeism and turnover in an Argentine factory. American Sociological Rei,ien,. 16.424428.
Koslowsky. M. (in prep). Objecrive and subjeclive predicrors of cornnurfing srroin. Bar-Ilan University. Ramat Gan. Israel.
Koslowsky. M.. Kluger. A.N.. & Reich. M. (1995). Cornmuring stress; Causes, effecrs, and nierhods of coping. New York: Plenum.
Koslowsksy. M.. & Krausz. M. (1993). O n the relationship between commuting, stress, and attitudinal measures: A LISREL application. Journal of Applied Behavioral Science, 29. 485492.
Landy. F.J. (1992). Work design and stress. In G.P. Keita & S.L. Sauter (Eds.). Work and well being. (pp.115-158). Washington. DC: American Psychological Association.
Landv. F.J.. Rastegary. H.. Thayer. J.. & Colvin. C. (1991). Time urgency: The construct and its measurement. Joumnl of Applied Psychology. 76.644-657.
Leigh. J.P.. & Lust. J. (1988). Determinants of employee tardiness. Work and Occuparions. 15. 78-95
Loehlin. J.C. (1992). Latenr variable models. Hillsdale. NJ: Lawrence Erlbaum Associates Inc.
,Macan. T.M.. Shahani. C.. Dipboye, R.L.. & Phillips, A.P. (1990). College students’ time management: Correlations with academic performance and stress. Journal of Educarionol Psychology. 82.760-768.
Martin. J . (1971). Some aspects of absence in a light engineering factory. Occuparional
Montag. 1. & Comrey. A.L. (1987). Internality and externality as correlates of involvement in fatal driving accidents. Journal of Applied Psychology. 72.339-343.
Murphy. L.R. (1988). Workplace interventions for stress reduction and prevention. In C.L. Cooper & R. Payne (Eds.). Causes, coping and consequences ofsiress a1 work (pp.301-342). Chichester. UK: Wiley.
Nicholson. N.. & Goodge. P.M. (1976). The influence of social. organizational, and biographical factors on female absence. Journal of Managenienr Studies. 13. 234-254.
Novaco. R.W.. Kliewer. W.. & Broquet. A. (1991). Home environmental consequences of commute travel impedance. American Journal of Communiry Psychology. 19,881-909.
Novaco. R.W.. & Sandeen. B.A. (1992. November). Mirigaring rhe stress of comrnuring ro nork: Ridesharing and rhe inreracrional effecis of gender. Paper presented at the APAI NlOSH conference. Stress in the 90s. Washington. DC.
Novaco. R.W.. Stokols. D.. Campbell. J. & Stokols. J . (1979). Transportation. stress. and community psychology. American Journal of Cornmimiry Psychology. 7,361-380.
Novaco. R.W.. Stokols. D.. & Milanesi. L. (1990). Objective and subjective dimensions of travel impedance as determinants of commuting stress. American Journal of Communiry
Psychology. 45-77-89,
P s ~ c h o l o g ~ . 18. 231-257.
COMMUTING STRESS 173
Pop, P.O.. & Belohlav, J.A. (1982). Absenteeism in a low status environment. Academy of Management Journal. 25,677-683.
Schaeffer, M.H.. Street, S.W., Singer. J.E.. & Baum. A. (1988). Effectsof control on the stress reactions of commuters. Journal of Applied Social Psychology, 18.944-957.
Schriber, J.B.. & Gutek. B.A. (1987). Some time dimensions of work: Measurement of an underlying aspect of organizational culture. Journal of Applied Psychology. 72,642450.
Schulz, R. (1976). Effects of control and predictability on the physical and psychological well-being of the institutionalized age. Journal of Personality and Social Psychology, 33. 563-573.
Seligman. M.E.P.. & Miller, S.M. (1979). The psychology of power: Concluding comments. In L.C. Perlmuter & R.A. Monty (Eds.), Choices and perceived control. Hillsdale. NJ: Lawrence Erlbaum Associates Inc.
Stokols. D. (1992). Establishing and maintaining healthy environments: Toward a social ecology of health promotion. American Psychologist, 47,6-22.
Taylor. P.. & Pocock, C. (1972). Commuter travel and sickness: Absence of London office workers. British Journal of Preventive and Social Medicine, 26. 165-172.
Thomas, L.T.. & Ganster, D.C. (1995). Impact of family-supportive work variables on work-family conflict and strain: A control perspective. Journal of Applied Psychology, 80. 6-15.
Thornson, S.C. (1981). Will it hurt less if I can control it? A complex answer to a simple question. Psychological Bulletin, 90, 89-101.
Wright. L.. McGurdy. S., & Rogoll, G. (1992). The TUPA scale: A self-report measure for the type A subcomponent of time urgency and perpetual activation. Psychological Assessment. 4. 352-356.
memo3/Lassarre.pdf
is and
c t r l l s s t
r a o t ( m p f
e e a r o e a i t t
1
c i A o a t ( s m c “ w r t t o v d s b w b c s a
2
2
R
P
w o c s
i a T t t t q o c t a
d i h
v
w “ g t r
k d �
because in case of jam the gap becomes null and (d) is equal to the length (�) of the vehicle.
S. Lassarre et al. / Accident Analys
rossing the road occur in different proportions between ‘junc- ions’ and ‘mid-block locations’, according to the structure of the oad network. Hence, the relative numbers of accidents between ocations with and without pedestrian facilities depends on the ocal transport policy and the resources devoted to pedestrian afety. The consequences of an accident, measured by the injury everity, are a function of the impact speed, the vehicle design, he road design and the vulnerability of the pedestrian.
In the following analysis, it is assumed that pedestrians are at isk only when crossing a road. This assumption is quite realistic, s non-crossing accidents generally represent a small proportion f pedestrian accidents (Duncan et al., 2002). The characteris- ics taken into consideration include those relating to the traffic traffic volume and speed), the road design (road width, median, arked crosswalks) and the traffic signals (colour lights and
edestrian crossing signs). No distinction is made between dif- erent types of vehicles.
Nevertheless, these restrictions leave space for complexity, specially when considering the possibilities rising from differ- nt types of behaviour at crossings. For a pedestrian trip from given origin to a given destination, there may be different
outes, each one entailing different sequences of crossing options f different types, which in turn correspond to different micro- nvironments. The choice between these sequences of options is trade-off between the pedestrian’s perception of risk, his will-
ngness to take his time in reaching his destination and his desire o feel comfortable, and these parameters may vary according o the characteristics of the individual.
.2. What is exposure?
In this research, the definition of accident exposure when rossing roads is analogous to air pollution exposure when walk- ng in urban areas. In environmental epidemiology, the National cademy of Science (1991) defines exposure as “an event that ccurs when there is a contact at the boundary between a human nd the environment with a contaminant of a specific concen- ration for an interval of time”. On the road network, a direct physical) contact occurs only in case of an accident, that is to ay, a collision between a road user and a vehicle that generates echanical energy, which is the cause of the damage, during a
ertain amount of time. Briggs (2000) gave some reasons why a looser definition of exposure may often need to be applied hen health is considered in the sense of positive well-being,
ather than merely ill-health”. Moreover, there is a virtual con- act between a road user and an “atmosphere” generated by the raffic. The quality of the “atmosphere” depends on the presence f “contaminants” that correspond, in the traffic, to the moving ehicles and are described by a traffic volume and a speed. The ifficulty is to define an appropriate “concentration” suited to the ituation of road crossing. The time of exposure can be defined y the time spent by the pedestrian for crossing a road of a certain idth with a walking speed. The time spent in traffic has always
een a recommended indicator of what road safety specialists all a measurement of “risk exposure”. Moreover, the walking peed of a pedestrian depends on his or her age, the trip purpose, nd so on.
F H
Prevention 39 (2007) 1226–1238 1227
. Exposure assessment
.1. Selection of an exposure indicator
We start from a measure of accident risk proposed by outledge et al. (1974a,b, 1976):
c = � + vtc
d
here Pc is the accident risk of the crossing, � the average length f the vehicle, v the average speed of the flow, tc the average rossing time for a pedestrian and d is the average traffic gap, as hown in Fig. 1.
This indicator is meant to yield the proportion of space that s not available to the pedestrian for crossing the road freely nd safely (i.e. that is occupied by a virtual flow of vehicles). he space is equal to the length of the vehicle plus the distance
ravelled by the vehicle during the time taken by the pedes- rian to cross the road. It is also a measure of accessibility to he other side of the road. If there are long vehicles, travelling uickly and in large numbers, one cannot access the other side f the road because one faces a kind of “moving wall” and if one hooses to cross, one has an increased accident risk. However, here are some weaknesses in this indicator, which are analyzed nd assessed in the following paragraphs.
In particular, from the traffic theory, there exists a monotonic ecreasing function between speed (v) and density (k), measured n number of vehicles per kilometre. For a linear function, we ave:
= vf
( 1 − k
kJ
)
ith (vf) the free speed when the flow tends to zero, (kJ) the jam” density in saturation conditions. As (d) is equal to the ap between the front of the following vehicle to the rear of he preceding one plus the length of the vehicle, we have the elations:
= 1 and kJ = 1
ig. 1. Indicator by Routledge, Repetto-Wright and Howarth (following owarth).
1228 S. Lassarre et al. / Accident Analysis and
F 2
P
a P g
R c i o f b t c
P
w h t
v a t
t t F v c b t
t
b
P
i t m m s m v
fi o t
2 p
i e o s c t w
2
fl f I t T f d fl d o t c s o p a i
t
C
a
Li
ig. 2. Routledge and modified indicators function of speed (m/s) with vf = 0 m/s, � = 5 m, tc = 2 s with a linear function speed/density.
Then, the Routledge indicator becomes:
= k
( 1
kJ + tcv
) = kJ
( 1 − v
vf
) ( 1
kJ + tcv
)
= (
1 − v
vf
) (1 + tckJv)
This risk indicator is proportional to the density of vehicles nd to the speed of the flow. In saturation conditions, v = 0 and = 1. If there is no traffic, v = vf, P = 0. There is a maximum
reater than one at v = (vf/2) − (1/2kJtc) (Fig. 2). However, (P) is not a proportion as asserted erroneously by
outledge et al. (1974a,b), because the virtual vehicle lengths an overlap and in this case (P) is greater than 1. (P) can be nterpreted as the average number of virtual vehicles in one unit f length, which one can meet when crossing the road. In the ormula, there is a static and a dynamic part. The first is expressed y the ratio of two densities and the second is expressed by he number of vehicles that pass during the time the pedestrian rosses:
= k
( 1
kJ + tcv
) =
( k
kJ + tckv
) =
( k
kJ + tcq
)
ith (q) the traffic volume (measured in number of vehicles per our) and q = kv according to the fundamental relationship of raffic theory.
As explained above, when (P) is higher than 1, the flow of ehicles is a “wall”, which a pedestrian cannot cross. However, s the traffic is nearly saturated and the speed is low, opportuni- ies to cross between vehicles may still exist.
As it stands, the Routledge indicator has some drawbacks, herefore. The value P = 1, signifying that it is quite impossible o cross, is attained quite quickly when traffic density increases. urthermore, when the traffic is nearly saturated and the speed ery low, the value is still 1 because of the impossibility of rossing, although the risk is also low and the indicator should
e less than 1. The lack of accessibility may be high in dense raffic conditions, but the accident risk is relatively low.
To avoid this drawback, we propose to suppress the constant erm related to the saturation density in the equation, which
T
a m
Prevention 39 (2007) 1226–1238
ecomes a symmetric parabola, as shown in Fig. 2:
= ktcv = (
1 − v
vf
) tckJv = tcq
The “concentration” of vehicles expressed by the modified ndicator is now just equal to the average number of vehicles hat pass during the time taken to cross. The magnitude of the
odified indicator is smaller, however. For instance, in the sym- etric indicator, the “concentration” would be equal both at
peed v = 5 m/s and v = 15 m/s, while for the initial asym- etric indicator the “concentration” would be higher at speed = 5 m/s than forv = 15 m/s, because of the density of the traf- c. The symmetry assumption could be tested by means of field bservations and/or a roadside experiment, or by a computer or raffic simulator experiment.
.2. Exposure at junctions and mid-block locations, rotected or unprotected
The proposed exposure assessment algorithm is based on the dentification of traffic flows through which a pedestrian crosses; ach moment of pedestrian risk exposure consists of crossing a ne-lane traffic flow, qualified by a “concentration” (e.g. expo- ure) indicator (C) for a duration (t) equal to the time taken to ross and a width (L) equal to the crossing distance. The time aken to cross is equal to the width of the lane divided by the alking speed vc (= 1.4 m/s).
.2.1. Mid-block locations For each road crossed at mid-block, we consider number of
ows to be crossed, (m), as shown in Fig. 3. We suppose that or two-way roads, the flows are distinct: one in each direction. f there are more lanes in each direction, we create more dis- inct flows in each direction from the aggregated two-way flow. herefore, for a one-way road there is at least one flow to cross,
or a two-way road at least two flows, and so on. The flow in one irection is then broken down into distinct flows by lane. If two ows are physically separated (e.g. by a median), two indepen- ent crossings are considered (one from each direction) instead f a global one. Moreover, a marked crosswalk, offering protec- ion at mid-block crossings, may be available and needs to be onsidered. A vehicle–pedestrian interaction can also be con- idered at marked crosswalks, by imposing a speed reduction n the vehicle flow (e.g. by 20%). It is noted that this exam- le is just a suggestion; further measurements are necessary to ssess the level of alertness of the drivers in vehicle–pedestrian nteractions.
More specifically, where there is no separation, we calculate he (m) “concentrations”:
i = qivc(L1 + L2 . . . + Li),
nd the time:
i = vc
nd the speed vi, and note the presence or not of a arked crosswalk. Where there is a separation, we calculate
S. Lassarre et al. / Accident Analysis and Prevention 39 (2007) 1226–1238 1229
expo
(
C
a
C
a
T
a
m t t i l c t t z a
2
c f b a t t c f m t c a a
•
•
•
3
m t e m l
e v t t v f g p 2 i m
Fig. 3. Crossing situations and
m1) exposures:
1i = q1i vc(L11 + L12 . . . + L1i)
nd (m2) exposures:
2i = q2ivc(L21 + L22 . . . + L2i),
nd the times
1i = L1i
vc and T2i = L2i
vc
nd speeds and note the presence or not of a marked crosswalk. There are two special points to note. When leaving the pave-
ent, the exposure for crossing a far-side lane is higher than he exposure for crossing a near-side lane. In case of the same raffic volume on each lane, the exposure on the far-side lane s expected to be exactly twice the exposure on the near-side ane (i.e. 2qvL versus qvL). For that reason, we cumulate the rossing distances. One should consider carefully the order of he crossed flows from the departure roadside. Moreover, when here is a separation, the crossing distance should be reset to ero. By calculating all the individual flow exposures, we obtain micro-environmental exposure assessment.
.2.2. Traffic controlled junctions or mid-block locations At junctions with traffic lights, two phases for pedestrian
rossings can be considered: one when the lights are green or pedestrians (i.e. they are red for vehicles, with the possi- le exception of turning movements) and one when the lights re red for pedestrians (i.e. they are green for vehicles). When he lights are green for pedestrians, a pedestrian may be exposed o two types of risk: one resulting from the flow of turning vehi- les (from or into adjacent roads), and another one resulting rom vehicles not complying with the red light. Some turning ovements may be constrained by country regulations. When
he lights are red for pedestrians, a pedestrian is exposed if he hooses to disregard them and attempts to cross the road. We ssume that pedestrians do not cross outside marked crosswalks t junctions.
The calculation of exposure is then as follows:
When the lights are green for pedestrians and when non- compliant vehicles are taken into account, the exposure is
2 t w a
sures for mid-block locations.
calculated according to the same procedure and formulae as for mid-block locations. It is noted that this assumption has to be checked with data about red light running violations in relation to the traffic conditions. When the lights are green for pedestrians and vehicles turn- ing left or right are taken into account, the flow of traffic turning in each direction is measured, aggregated and dis- tributed proportionally to the number of lanes in that direction. Each lane of traffic is assumed to be one-way. The speed is taken as equal to the maximum of the speeds of the turning vehicles. The exposure is then calculated as for mid-block locations. When the lights are red for pedestrians and non-compliant pedestrians are taken into account, the exposure is assumed to be equal to the case in which the lights are green for pedes- trians and non-compliant vehicles are taken into account.
. Crossing behaviour modelling
According to the above, accident risk exposure can be esti- ated for any location (micro-environment) along a pedestrian
rip, where a pedestrian is likely to cross. However, in order to stimate the total exposure for an entire trip, it is necessary to odel pedestrian behaviour, in terms of the choice of crossing
ocation. Previous research on pedestrian movement in urban areas is
xtensive and ranges from modelling pedestrian behaviour and ehicle–pedestrian interactions, to accident analyses and evalua- ion of safety measures. However, few attempts have been made o model pedestrians’ crossing behaviour at the level of an indi- idual trip. Several theories and approaches have been proposed or pedestrian modelling. Most see crossing behaviour as largely overned by the gap-acceptance theory, which states that each edestrian has a critical gap to cross the road (Manuszac et al., 005; Hamed, 2001). Another interesting approach for estimat- ng crossing preferences concerns pedestrian level-of-service
odels (Sarkar, 1995; Baltes and Chu, 2002; Guttenplan et al.,
003). In addition, a promising approach for modelling pedes- rian crossing behaviour is offered by discrete choice models, hich correlate crossing decisions to a utility function (Evans
nd Norman, 1998; Hine and Russel, 1993).
1 is and Prevention 39 (2007) 1226–1238
s h a b e ( t F b
t a a w c e p
3
a w i b s c t r t
s
•
•
s
i t t m c c e t s
t i
F N g
t o c c o t w t o
t s d w d M p o
t t t d o F
230 S. Lassarre et al. / Accident Analys
However, an overall approach to pedestrians’ crossing deci- ions (i.e. the places where pedestrians are most likely to cross) as not yet been presented. More specifically, most studies nalyze crossing decisions at a particular location, while the ehaviour of pedestrians along an entire trip has not been xplored in detail. Moreover, crossing at uncontrolled locations mid-block crossing, jaywalking, etc.), which is a common pat- ern of behaviour, is not taken into account in most studies. inally, most studies are not designed to link observed crossing ehaviour to pedestrian risk exposure.
The pedestrian behaviour model outlined below provides a ool to estimate crossing probabilities for each location along ny pedestrian trip. It is based on existing models, which were dapted for the purposes of this research, while additional tools ere developed as well. The results of modelling pedestrian
rossing behaviour can be combined with corresponding risk xposure estimates to calculate the total weighted exposure of a edestrian along a trip.
.1. Basic principles
This approach is based on several principles. First of all, the nalysis aims to model “real-life” situations in an urban net- ork, where pedestrian trips may be more or less complex and
nclude several changes of direction. In this research, crossing ehaviour along a trip with several changes of direction is con- idered to be similar to those along a corresponding trip with no hange of direction. Road sections along a complex route are hus transformed to a single, linear, uni-directional sequence of oads. Accordingly, the origin and the destination of a trip are aken to be the beginning and end of this linear continuous route.
Based on the above, two categories of crossings can be con- idered, taking into account the network layout:
“Primary” crossings made at junctions or mid-block loca- tions (with change of direction) selected for the purpose of following the particular route. “Secondary” crossings made at junctions on either side of the road (without change of direction) while moving along sequential road links (i.e. additional crossings made as a con- sequence of following the route).
An example of “primary” and “secondary” crossings is pre- ented in Fig. 4a.
The proposed model yields probabilities for primary cross- ngs only; secondary crossings probabilities (i.e. whether these ake place on the origin side or on the destination side of the rip roadway) depend on and can be calculated from the pri-
ary crossings probabilities. For example, in Fig. 4, if a primary rossing takes place across the first road link, the first secondary rossing takes place on the destination side of the roadway. How- ver, if no primary crossing takes place across the first road link, hen the first secondary crossing would take place on the origin
ide of the trip roadway.
From the above definition, it is obvious that the classifica- ion into “primary” and “secondary” crossings does not mean to mply that the former are more important or more difficult ones
t i o d
ig. 4. (a) “Primary” and “secondary” crossings along a pedestrian trip. (b) umber of “primary” crossings along a pedestrian trip in relation to ori- in/destination.
han the latter. In practice, it is perfectly possible that a sec- ndary crossing involves crossing a major road and a primary rossing involves crossing a minor road. The specific classifi- ation is solely related to the nature of the crossings in terms f modelling approach; a “primary” crossing is probabilistic, in he sense that it can be made on various locations along the trip, hereas a “secondary” crossing is deterministic, in the sense
hat it can be made either on the origin or on the destination side f the roadway at a specific location along the trip.
The total number of primary crossings along a trip depends hus on the origin–destination combination in relation to the ide of the roadway along the trip. In particular, if the origin and estination are on the same side of the roadway, a pedestrian ould not have to cross; if the origin and destination are on ifferent sides of the roadway, a pedestrian would have to cross. oreover, in complex trips with several changes of direction,
edestrians are likely to make additional primary crossings, in rder to minimise their walking distance.
For example, in the top part of Fig. 4b, an example of a rip on which origin and destination are on the same side of he roadway is presented; a pedestrian could either walk along he origin side of the roadway (and not cross with change of irection at all) or cross with change of direction twice, in rder to reach the destination. Moreover, the bottom part of ig. 4b shows an example of a trip on which origin and des-
ination are on different sides of the roadway; in this case, it
s shown that a pedestrian would have to cross with change f direction either once, or three times, in order to reach the estination.
is and
c a
•
•
3
3
b t t l ( d t
L L
m s s t t l t ( ( t w v
c o c
U
u m t p c
n a b a t I
b a t M r p f
L L
w n f
3
p t b c i d s p
t e t l i
f In particular, the mid-block branch options of the nested logit model were not exploited so far, but were simply used to obtain the total marginal probability for mid-block. As mentioned above, these options are:
S. Lassarre et al. / Accident Analys
On the basis of the above, the total number of primary rossings N along a pedestrian trip can be estimated through distribution as follows:
If origin–destination are on the same side of the roadway, then N {0, 2, 4, . . .}. If origin–destination are on different sides of the roadway, then N {1, 3, 5, . . .}.
.2. Model development
.2.1. Model for a single road link The baseline case of one single road link (i.e. a road segment
etween junctions) is examined first. Assuming that the pedes- rian’s origin and destination are located on different sides of he road link, he or she has a choice between different crossing ocations: one of the two junctions or some location at mid-block between the two junctions). Therefore, pedestrians crossing ecisions are considered to follow a two-level hierarchical struc- ure:
evel 1: Junction or mid-block? evel 2: If junction, which one?
In order to model this decision process, we used a nested logit odel formulated by Chu et al. (2003). In that study, a crossing
cenario was presented to survey participants, who were asked to tate their crossing preferences. They were given several options o choose from: two options for crossing at a junction and up o four options for crossing at a mid-block location. The nested ogit model fitted to the data had thus a two-level structure. The op level had two branches: junctions (I) and mid-block locations M). The bottom level had two options in the junction branch one for each junction of the road link) and up to four options in he mid-block branch (B, cross first–walk later; C, jaywalk; D, alk first–cross later; F, use mid-block crosswalk). Explanatory ariables mainly concern road characteristics.
The calculation of probabilities also follows a two-level pro- edure. The utility UO for option (O) is defined as the sum f the products of all variables values with the corresponding oefficients.
tility function UO = ∑
(variable × coefficient)
The probability of a crossing option being chosen is the prod- ct of its marginal and conditional choice probabilities. The arginal probability represents the probability of choosing junc-
ions or mid-block options. The conditional probability is the robability of choosing a particular crossing option once the hoice has been made between junction and mid-block.
The model considered in the present research is part of this ested logit model, which is better adapted to the needs of the nalysis. In particular, the top level of the model (junction or mid-
lock) is used to obtain Level 1 marginal crossing probabilities, nd the “junction” branch of the bottom level of the model is used o obtain Level 2 conditional crossing probabilities for junction. t should be noted that the crossing options of the “mid-block”
F r
Prevention 39 (2007) 1226–1238 1231
ranch of the bottom level of the model are not considered, s they are not useful in this research; however, they are used o calculate the utilities and inclusive values of the top level.
oreover, option F – to use mid-block crosswalk – can be used in oads links with mid-block crosswalks as the Level 2 conditional robability for mid-block. The analysis would then have the ollowing structure:
Level 1: Junction or mid-block? evel 2.1: If junction, which one? evel 2.2: If at mid-block, on mid-block crosswalk or not?
It should be noted that a series of variables sensitivity tests ere carried out for the final model, allowing for minimizing the umber of variables. The crossing behaviour model considered or single road links is presented in Table 1.
.2.2. Model for a pedestrian trip If we consider a pedestrian trip along a sequence of, for exam-
le, three road links between six traffic controlled junctions, he resulting crossing probabilities produced by the crossing ehaviour model would have the general form shown in Fig. 5: rossing probabilities are significantly higher at junctions than n mid-block locations. We thus obtain a uniform probability istribution along the trip, as all links along the trip are con- idered to be equivalent, in the sense that no preference for a articular link (or road section) is taken into account.
However, it is necessary to account for the different weights hat different links or sections may have in the crossing prefer- nces of a pedestrian. This parameter is related not so much to he traffic and geometry parameters of the road that make it more ikely to be chosen than another, as to the natural tendency of an ndividual to behave in a particular way along a given route.
In order to incorporate this parameter in the model, some unctions of the nested logit model are used (Chu et al., 2003).
ig. 5. Model results: crossing probabilities (junction or mid-block) along three oad links.
1232 S. Lassarre et al. / Accident Analysis and Prevention 39 (2007) 1226–1238
Table 1 The hierarchical crossing behaviour model for a single link—overview of crossing options, variables and values
(
• • • •
u s o m i a f r
b o r B C D
t d p t ( a o i
t a t
trian trip are calculated by weighting the uniform crossing probabilities (junction or mid-block for each road link) with the non-uniform crossing probability distribution (crossing earlier
•) Value to be entered.
B: cross first and walk later. C: cross diagonally (jaywalk). D: walk first and cross later. F: use a mid-block crosswalk.
Options B, C, D can be considered to express an individ- al’s tendency to cross earlier or later (or randomly select a ection) along a link. Three utility functions correspond to these ptions. If these three functions are isolated from the rest of the odel, one can notice the following: when the walking distance
s different for the three options, different utility combinations re obtained. However, when the walking distance is the same or the three options, the same utility distribution is obtained, egardless of the value of the walking distance.
Therefore, considering a trip of a given length, along a num- er of links, and that all other variables are equal along the trip, ne can calculate the following probabilities on the basis of the
espective utility functions: : cross first and walk later P = 0.579 : cross diagonally (jaywalk) P = 0.064 : walk first and cross later P = 0.358
Fig. 6. Non-uniform crossing probability along a trip.
o
F r
One can thus consider that the probability B describes the endency to cross at the beginning of the trip, the probability D escribes the tendency to cross towards the end of the trip and the robability C describes the tendency to cross in the middle of the rip. By attributing these probability values to the starting point 0% of the distance), the ending point (100% of the distance) nd the middle of the trip (50% of the distance), respectively, ne can obtain a non-uniform probability distribution, as shown n Fig. 6.
In the equation, the value x corresponds to the percentage of he total length of the trip. The equation is generic and may be pplied to any pedestrian trip, in order to weight each link in erms of its location along the route.
The final crossing probabilities for each option along a pedes-
r later along the trip). The general form of the resulting final
ig. 7. Model results: final crossing probability distribution for a trip along three oad links.
is and
c F a i t F i w t f t
p T p b o r d
3
c
•
•
•
•
n r f l
r i t
t b ( t t 4 t M d N
f d
l s
•
•
p t l
s d o c h d a t
i t c t r p F d T i
4
4
d t c
w o a ( a
a i d
S. Lassarre et al. / Accident Analys
rossing probability distribution along the trip is presented in ig. 7. Fig. 7, also as Fig. 5, refers to the case of a pedestrian trip long three road links. Moreover, the estimated probabilities n Fig. 7 are obtained by weighting the estimated probabili- ies of Fig. 5 with the non-uniform probability distribution of ig. 6. While in Fig. 5 the interim locations are identified accord-
ng to their type (e.g. junction or mid-block) and the link on hich they are located (i.e. link 1, link 2 or link 3), in Fig. 7
he same locations are identified according to their distance rom the trip origin, as a percentage of the total length of the rip.
It should be also noted that the model ultimately yields inde- endent crossing probabilities, i.e. these do not add up to one. his is in accordance with our initial assumption that, in any edestrian trip, there may be additional crossings (over and eyond the minimum required). However, in the event that only ne crossing is to be considered, the final probabilities can be escaled to add up to one, and can thus be considered to be ependent.
.3. Summary and validation
A hierarchical methodology is used to estimate pedestrians’ rossing behaviour, consisting of the following steps:
Estimation of the total number of crossings along a trip, in relation to origin and destination parameters. Estimation of crossing probabilities at different locations along each road link. Estimation of crossing probabilities along the trip, in relation to the distance from origin and destination. Calculation of the weighted final crossing probabilities for each location along the trip.
On the basis of the above, a model for estimating the type, umber and location of crossings along a trip was developed. The esulting algorithm makes it possible to calculate probabilities or different pedestrian crossing choices along a trip through a imited yet sufficient number of variables.
Preliminary validation of the model yielded promising esults. However, there is some evidence that there is room for mprovement, mainly as regards a more precise calibration of he parameters.
In particular, the first step of the model, entailing the selec- ion of the total number of crossings, was validated on the asis of real data from a pedestrian survey in Lille, France WHO, 2006; Lassarre et al., 2005). It was confirmed that, if rip origin–destination are on the same side of the roadway, then he total number of crossings N follows a distribution N {0, 2, , . . .}. It also suggested that further calibration is necessary for he particular case of trips along a single direction (e.g. N = 0).
oreover, it was confirmed that, if trip origin–destination are on ifferent sides of the roadway, then the total number of crossings
follows a distribution N {1, 3, 5, . . .}. As far as the second step of the model is concerned, the trans-
ormation of the nested logit model proved to be sufficient for escribing the distribution of crossing probabilities along a road
2 T r g
Prevention 39 (2007) 1226–1238 1233
ink (junctions or mid-block). Validation was based on two data ets (WHO, 2006):
CCTV recordings of crossing decisions of 1870 pedestrians in Florence, Italy, on a road link between two uncontrolled junctions. A field survey in Athens, Greece, recording crossing deci- sions of 1793 pedestrians on a road link between two traffic controlled junctions.
The results showed non-significant differences among model redictions and observed behaviour. It should be noted, however, hat the performance of the model is somewhat improved for inks with traffic controlled junctions.
Finally, the non-uniform probability distribution of the third tep of the pedestrian behaviour model was compared with real ata from the Lille pedestrian survey. The basic assumptions f the model were found to be accurate (higher probability of rossing at the beginning or the end of the trip). There again, owever, there is room for improvement after calibration for ifferent area types, given certain particularities of the survey rea in Lille (i.e. an area around the university campus where here are only few traffic controlled junctions).
The above results adequately validate the proposed model. It s noted that the model includes a sufficient number of explana- ory variables, whose combination can capture the majority of onditions encountered in urban areas. Within this framework, he survey sites on which data were collected were typical and epresentative of different urban conditions. Consequently, the roposed model can be applied to the majority of urban settings. urther calibration, in order to account for special or local con- itions might be required and would certainly be interesting. he incorporation of interactions between pedestrians is also an
nteresting field for further improvement of the model.
. Feasibility demonstration
.1. An example
Fig. 8(1) shows three pedestrian trips in the Quartier Latin istrict of Paris. In addition, Fig. 8(2) shows the main pedes- rian traffic control facilities in the area (traffic lights, marked rosswalks).
We will look only at the crossings along the main network, here there is no separation of flows. According to the method- logy presented above, these trips can be considered along single direction, by taking into account the network layout
which affects the number of primary and secondary crossings), s shown in Fig. 9.
For each trip, the pedestrian crossing behaviour model is pplied as described above, in order to obtain primary cross- ngs probabilities. In Table 2, the uniform crossing probability istribution – i.e. the crossing probabilities (junction 1, junction
or mid-block) – are calculated for each road link of the trip. hen, the non-uniform probability distribution is estimated in
elation to the distance of each crossing option from the trip ori- in. The final crossing probabilities are obtained as the product
1234 S. Lassarre et al. / Accident Analysis and Prevention 39 (2007) 1226–1238
nd C on the study area—pedestrian facilities in the study area.
o fi t
t
e a
F s
Fig. 8. (1 and 2) Origin, route, destination of trips A, B a
f the uniform and non-uniform probability distributions. These nal (independent) probabilities can also be rescaled to add up
o one, in the event that they are considered to be dependent. Fig. 10 shows the final rescaled (dependent) probability dis-
ributions for each trip.
Furthermore, for each one of the primary crossing options of
ach trip, a measure of accident risk exposure can be obtained s in Table 3. For each junction or mid-block location, the infor-
ig. 9. Representation of origin, route and destination of trips A, B and C on a ingle linear direction. Fig. 10. Final (scaled) crossing probability distribution for trips A, B and C.
S. Lassarre et al. / Accident Analysis and Prevention 39 (2007) 1226–1238 1235
Table 2 Probabilities for the primary crossing options of trips A, B and C
Link Location Probability Distance from origin % Trip distance Non-uniform probability distribution
Final probabilities Rescaled probabilities
Trip A, o/d = 1 Link 1 J1 0.109 0 0.0 0.579 0.063 0.066
MB 0.101 35 10.6 0.402 0.041 0.043 J2 0.790 70 21.2 0.261 0.207 0.218
Link 2 J1 0.854 75 22.7 0.244 0.209 0.221 MB 0.103 130 39.4 0.105 0.011 0.011 J2 0.043 190 57.6 0.056 0.002 0.003
Link 3 J1 0.526 195 59.1 0.057 0.030 0.032 MB 0.280 225 68.2 0.077 0.022 0.023 J2 0.194 260 78.8 0.134 0.026 0.027
Link 4 J1 0.044 265 80.3 0.145 0.006 0.007 MB 0.093 295 89.4 0.228 0.021 0.022 J2 0.864 330 100.0 0.358 0.309 0.327
0.946 1.000
Trip B, o/d = 2 Link 1 J1 0.310 0 0.0 0.579 0.179 0.306
MB 0.381 95 26.4 0.206 0.078 0.134 J2 0.310 190 52.8 0.059 0.018 0.031
Link 2 J1 0.044 195 54.2 0.057 0.002 0.004 MB 0.093 240 66.7 0.072 0.007 0.011 J2 0.863 290 80.6 0.147 0.127 0.217
Link 3 J1 0.864 295 81.9 0.158 0.137 0.233 MB 0.093 325 90.3 0.237 0.022 0.037 J2 0.044 360 100.0 0.358 0.016 0.027
0.586 1.000
Trip C, o/d = 2 Link 1 J1 0.109 0 0.0 0.579 0.063 0.097
MB 0.101 35 10.8 0.399 0.040 0.063 J2 0.790 70 21.5 0.258 0.204 0.315
Link 2 J1 0.475 75 23.1 0.240 0.114 0.176 MB 0.051 162.5 50.0 0.064 0.003 0.005 J2 0.475 255 78.5 0.132 0.063 0.097
Link 3 J1 0.864 260 80.0 0.143 0.123 0.191 MB 0.093 290 89.2 0.226 0.021 0.032 J2 0.044 325 100.0 0.358 0.016 0.024
m p t t
i a c “ p t t i t m
a t o
4
c F s w
ation required is related to three flows: one in each direction, lus one turning movement in case of a junction. Additionally, he exposures for each of the secondary crossings along each rip are calculated as shown in Table 4.
Summarizing, a pedestrian’s risk exposure can be weighted n relation to the different crossing options encountered along
trip and to the behaviour of pedestrians when it comes to rossing decisions. These crossing decisions mainly concern primary” crossings, i.e. crossings that are necessary for the edestrian in order to reach his or her destination. Moreover, on he basis of these primary crossing decisions, one can estimate
he risk exposure corresponding to the “secondary” crossings, .e. the crossings that are a consequence of following the par- icular route. As a result, degrees of exposure for different
icro-environments can be combined with detailed information
w t l p
0.647 1.000
bout the presence of pedestrians in these micro-environments o enable an integrated exposure assessment to be carried ut.
.2. Discussion
The feasibility of the proposed approach can be better per- eived by examining the results for trip C in more detail (see ig. 10). In this trip, the origin and destination are on different ides of the roadway and therefore an odd number of crossings ould be expected. Assuming that only one primary crossing
ill be made, an intuitive approach would be to cross around
he beginning or towards the end of the trip, i.e. junction 2 of ink 1 or junction 1 of link 3. Indeed, the model yields increased robabilities for these particular options, thus reflecting the most
1236 S. Lassarre et al. / Accident Analysis and Prevention 39 (2007) 1226–1238
Table 3 Exposure for the primary crossing options of trips A, B and C
Location Main trip roadway 1 Main trip roadway 2 Turning movements
Volume Speed Time Exposure Volume Speed Time Exposure Volume Speed Time Exposure
Trip A Link 1 J1 50 50 3.6 0.050
MB 50 50 3.6 0.050 J2 (L) 50 50 3.6 0.050
Link 2 J1 (L) 50 50 3.6 0.050 MB 50 50 3.6 0.050 J2 50 50 3.6 0.050
Link 3 J1 550 40 5 0.764 MB 550 40 5 0.764 J2 550 40 5 0.764
Link 4 J1 680 35 5 0.944 MB 680 35 5 0.944 J2 680 35 5 0.944
Trip B Link 1 J1 200 35 3.6 0.198
MB 200 35 3.6 0.198 J2 200 35 3.6 0.198
Link 2 J1 550 40 5 0.764 MB 550 40 5 0.764 J2 (L) 550 40 5 0.764 100 25 5 0.139
Link 3 J1 (L) 680 35 5 0.944 MB 680 35 5 0.944 J2 680 35 5 0.944
Trip C Link 1 J1 50 35 3.6 0.050
MB 50 35 3.6 0.050 J2 (L) 50 35 3.6 0.050
Link 2 J1 (L) 500 40 3.6 0.496 500 40 3.6 0.992 MB 500 40 3.6 0.496 500 40 3.6 0.992 J2 (L) 500 40 3.6 0.496 500 40 3.6 0.992
Link 3 J1 (L) 680 35 5 0.944 MB 680 35 5 0.944
(
r n b a e
C t o c l ( m l 1 v a b
s c
A c i ( w i i 2 t j
J2 680 35 5 0.944
L) traffic light.
easonable strategy to be pursued in this case. It should also be oted that it is highly unlikely for this single primary crossing to e made in the middle of link 2, as link 2 should then be crossed gain as a secondary crossing, and this is also reflected in the stimated probability distribution.
Assuming now that three primary crossings are made on trip , this would actually correspond to an attempt to minimise
he distance walked and follow the shortest path between the rigin and the destination. In this case, it would be reasonable to onsider that one crossing would be made on each of the three inks. The first crossing would most likely be made at the second traffic controlled) junction of link 1 and the last crossing would ost likely be made at the first (traffic controlled) junction of
ink 3, while the second crossing could be made either at junction
or at junction 2 of link 2. The mid-block option of link 2 is
ery unlikely to be selected, as the crossing distance is greater long this section and the traffic is denser. Hence, it would only e selected in the event of a large gap in the flow of traffic.
p
d e
We thus find that the estimated probability distribution corre- ponds to intuitive behaviour, whatever the number of crossings onsidered.
A similar conclusion can be drawn from the analysis of trip (Fig. 10), which would require an even number of primary
rossings. In the case of two primary crossings, the shortest ntuitive path would include one crossing at the end of link 1 junction 2 of link 1 or junction 1 of link 2) and one crossing hen approaching the destination. This trend is fully reflected
n the estimated probability distribution, which also indicates an ncreased probability of selecting a second crossing at junction of the last link, which would represent a slight deviation from
he shortest path for the purpose of crossing at a traffic controlled unction. These results are in accordance with general walking
ractice.
That said, one particular aspect of these modelling results eserves further discussion, and that is mid-block crossings, specially those made in the middle of the trip, which appear to
S. Lassarre et al. / Accident Analysis and Prevention 39 (2007) 1226–1238 1237
Table 4 Exposure for the secondary crossings of trips A, B and C
Location of secondary crossing
Main trip roadway 1 Main trip roadway 2 Turning movements
Volume Speed Time Exposure Volume Speed Time Exposure Volume Speed Time Exposure
Trip A Origin side 1 (L) 500 40 3.6 0.496 500 40 3.6 0.992
2 600 35 5.0 0.833 3 130 35 5.0 0.181
Destination side 1 (L) 500 40 3.6 0.496 500 40 3.6 0.992
Trip B Origin side 1 (L) 500 40 3.6 0.496 500 40 3.6 0.992 180 25 3.6 0.179
Destination side
1 450 40 5.0 0.625 2 100 35 3.6 0.099 3 (L) 530 40 3.6 0.526 400 45 3.6 0.794 50 25 3.6 0.050
Trip C Origin side 1 (L) 500 40 3.6 0.496 500 40 3.6 0.992
2 (L) 50 35 3.6 0.050
Destination side
1 (L) 550 40 5.0 0.764 100 25 5.0 0.139 2 (L) 530 40 3.6 0.526 400 45 3.6 0.794 50 25 3.6 0.050
(
b t t u c t i
5
b l a o e p d t d i p c
o a a m p c b i c t
w h p d s r r p 4 w b
b c b t p p c fi
t i o r c i c
L) traffic light.
e penalized to some extent. This may partly be due to the fact hat the examples analyzed are located in an area of increased raffic. Another more obvious reason is the form of the non- niform distribution, which further discourages the mid-block rossings of the middle of the trip. Further validation of this dis- ribution will make it possible to calibrate the existing form and mprove it if necessary.
. Conclusion
The methodology described above yields estimated proba- ilities for specific crossing decisions made by pedestrians and inks these decisions to their accident risk exposure. From the ssumptions of the analysis and the models’ structure, it is bvious that the dynamics of pedestrian decisions can be mod- lled within an appropriate framework. The application of the edestrian behaviour model to three paths in the Quartier Latin istrict of Paris gives some promising results, which correspond o a realistic representation of pedestrian movements in accor- ance with the initial assumptions of the analysis. The model ntegrates variables related only to traffic conditions and the rotection offered to the pedestrian for crossing: traffic lights or rosswalks.
An assessment of the exposures for each crossing is made n the basis of the usual information available about traffic in n urban network: traffic volumes, densities and speeds by lane nd by turning movements. Lane crossing exposure, defined as a icro-environmental “concentration” of vehicles, is simply the
roduct of the time taken to cross, which could depend on the haracteristics of the pedestrian, and traffic volume, which could
e on an hourly or daily basis. For one trip or a set of trips, it s possible to come up with a distribution of the exposures by ombining these values with the probabilities of crossings and hen estimating an average exposure value.
e a c o
A microscopic simulation framework is under development, hich will have the following structure. As inputs, we would ave trips (which may be solely walk or by a mono or bi-mode lus a walk) and corresponding origin–destination pairs. These ata could be defined by specific addresses or be randomly elected within zones. Using simulation, we would describe the oute followed by the pedestrian during the trip, as a sequence of oad links that he or she has to go along. The number of crossings er trip would be randomly selected from a distribution N {0, 2, , . . .} or N {1, 3, 5, . . .) according to the side of the roadway on hich the origin and the destination are located, this roadway eing considered as a single, linear one.
Depending on the design and facilities of the road (link) to e crossed there are crossing options for each link: outside any rossing facilities (mid-block), on marked crosswalk at mid- lock, at junction (junction 1 or junction 2, crosswalk-marked or raffic controlled). By means of the set of the estimated crossing robabilities, a set of crossing choices can be attributed to a edestrian. These would then be weighted by the non-uniform rossing probability distribution for the entire trip, to obtain the nal crossing probability distribution.
After that, a crossing location would be randomly selected as he first crossing to be made along the trip. Once this selection s made, the chosen location would be considered as the origin f the rest of the trip, and the crossing probabilities would be e-calculated for the new trip. According to the total number of rossings selected, a location would be selected for each cross- ng and the modelling process would restart from the selected rossing location.
Accordingly, an amount of exposure would be assigned to
ach crossing option, which would enable the total exposure long the trip to be estimated, taking account also of “secondary” rossings. This approach, combining both traffic and epidemi- logical elements, would make it possible to aggregate risk
1 is and
e m
A
s S
R
B
B
C
D
E
G
G
H
H
J
L
M
N
R
R
R
238 S. Lassarre et al. / Accident Analys
xposure along real pedestrian trips passing through different icro-environments.
cknowledgement
This work has been supported by the European Commis- ion within the “HEARTS-Health Effects and Risks of Transport ystems” Project of the 5th Framework Programme.
eferences
altes, M., Chu, X., 2002. Pedestrian level of service for mid-block street crossings. In: TRB 81st Annual Meeting, Transportation Research Board, Washington.
riggs, D., 2000. Exposure assessment in spatial epidemiology. In: Elliott, P., Wakefield, J.C., Best, N.G., Briggs, D.J. (Eds.), Methods and Applications. Oxford University Press, Oxford.
hu, X., Guttenplan, M., Baltes, M., 2003. Why people cross where they do—the role of street environment. In: TRB 82nd Annual Meeting, Transportation Research Board, Washington.
uncan, C., Khattak, A., Hughes, R., 2002. Effectiveness of pedestrian safety treatments for hit-along-roadway crashes. In: The Proceedings of the TRB 81st Annual Meeting, Transportation Research Board, Washington.
vans, D., Norman, P., 1998. Understanding pedestrians road crossing deci- sions: an application of the theory of planned behaviour. Health Educ. Res. 13 (4).
uerin, M., Gosselin, P., Cordier, S., Viau, C., Quénel, P., Dewailly, E., 2003. Environnement et santé publique. Fondements et pratiques, Edisem, Canada.
S
W
Prevention 39 (2007) 1226–1238
uttenplan, M., Landis, B., Crider, L., McLeod, D., 2003. Multi-modal level of service—analysis at a planning level. In: TRB 82nd Annual Meeting, Transportation Research Board, Washington.
amed, M., 2001. Analysis of pedestrians behavior at pedestrian crossings. Saf. Sci., 38.
ine, J., Russel, J., 1993. Traffic barrier and pedestrian crossing behavior. J. Transp. Geogr. 1 (4).
ulien, A., Carré, J.-R., 2002. Cheminements piétonniers et exposition au risque (Risk exposure during pedestrian journeys). Recherche Transports Sécurité 76, 173–189.
assarre, S., Banos, A., Bodin, F., Bonnet, E., Papadimitriou, E., Zeitouni, K., 2005. The Lille case study. Deliverable 24 of the HEARTS project, INRETS.
anuszac, M., Manski, C., Das, S., 2005. Walk or wait? An empirical analysis of street crossing decisions. J. Appl. Econom. 20 (4), 529–548.
ational Academy of Science, 1991. Human Exposure Assessment for Air- borne Pollutants. Advances and Opportunities. National Academy Press, Washington, DC.
outledge, D., Repetto-Wright, R., Howarth, I., 1974a. A comparison of inter- views and observation to obtain measures of children’s exposure to risk and pedestrians. Ergonomics 17, 623–638.
outledge, D., Repetto-Wright, R., Howarth, I., 1974b. The exposure of young children to accident risk as pedestrians. Ergonomics 17 (4), 457–480.
outledge, D., Repetto-Wright, R., Howarth, I., 1976. Four techniques for mea- suring the exposure of young children to accident risk as pedestrians. In: Proceedings of the International Conference on Pedestrian Safety, Haifa,
Israel.
arkar, S., 1995. Evaluation of safety for pedestrians at macro- and microlevels in urban areas. Transportation Research Record No. 1502.
HO-World Health Organization, 2006. Health effects and risks of transport systems: the HEARTS project. WHO Regional Office for Europe.
- Measuring accident risk exposure for pedestrians in different micro-environments
- Introduction
- Main pedestrian accident scenarios
- What is exposure?
- Exposure assessment
- Selection of an exposure indicator
- Exposure at junctions and mid-block locations, protected or unprotected
- Mid-block locations
- Traffic controlled junctions or mid-block locations
- Crossing behaviour modelling
- Basic principles
- Model development
- Model for a single road link
- Model for a pedestrian trip
- Summary and validation
- Feasibility demonstration
- An example
- Discussion
- Conclusion
- Acknowledgement
- References
memo3/McNabola-PM_VOC_exposure_and_uptake.pdf
ilable at ScienceDirect
Atmospheric Environment 42 (2008) 6496–6512
Contents lists ava
Atmospheric Environment
journal homepage: www.elsevier .com/locate/atmosenv
Relative exposure to fine particulate matter and VOCs between transport microenvironments in Dublin: Personal exposure and uptake
A. McNabola*, B.M. Broderick, L.W. Gill Department of Civil, Structural and Environmental Engineering, Trinity College Dublin, Ireland
a r t i c l e i n f o
Article history: Received 12 October 2007 Received in revised form 20 March 2008 Accepted 2 April 2008
Keywords: PM2.5
Benzene Personal exposure Commuter transport Pollutant uptake Direct comparisons
* Corresponding author. Tel.: þ353 1 896 3321; fa E-mail address: [email protected] (A. McNabola)
1352-2310/$ – see front matter � 2008 Elsevier Ltd doi:10.1016/j.atmosenv.2008.04.015
a b s t r a c t
To compare the relative exposure to and uptake of air pollutants between modes of commuter transport, measurements of personal exposure to PM2.5 and traffic-related VOCs were obtained over an 18-month period. In total, 468 samples were recorded comprising journeys equally divided between the four main modes of commuter transport in Dublin, Ireland: the private car, pedestrian, public bus and cyclist. Samples were recorded along two fixed routes approaching/exiting the city centre at fixed times of the day (morning and evening peak traffic flows, 08:00–09:00 and 17:00–18:00). Samples were measured using a high flow gravimetric personal sampler for PM2.5 and a low flow vacuum operated bag sampler for VOCs. Sampling was always carried out simultaneously between two modes of transport to ensure a direct comparison regardless of meteorolog- ical and traffic conditions. Significant differences were found between the personal exposures recorded in the four modes investigated. The car commuter was found to have the highest exposure to VOCs, while the bus commuter was found to have the highest exposure to PM2.5. The pedestrian was consistently found to have the lowest exposure. Significant differences were also found between the two fixed routes investigated. How- ever, as there were differences in physiological states, exposure durations and exposure levels between the four modes of transport, it was deemed necessary to estimate the total uptake of pollutants by means of a numerical human respiratory tract model. The results showed the cyclist to have the highest deposition of PM2.5 in the lungs followed by the bus, pedestrian and car. The car passenger had the highest absorption of VOCs followed by the cyclist, pedestrian and bus. Hence, the findings of the human respiratory tract model give a significantly different impression of relative uptake of pollutants to the relative exposure concentrations found initially.
� 2008 Elsevier Ltd. All rights reserved.
1. Introduction
Studies addressing the personal exposure of the public to air pollution in transport microenvironments have steadily increased in the last decade (Kaur et al., 2007). Individual commuters have been shown to receive a signif- icant proportion of their daily exposure to air pollutants in
x: þ353 1 677 3072. .
. All rights reserved.
the relatively short period in which they are commuting to and from urban areas (Michaels and Kleinman, 2000; Schweizer et al., in press). Pollutants emitted in the trans- port microenvironment such as PM2.5, primarily emitted from diesel-engined vehicles, have been shown to be asso- ciated with adverse cardiac and respiratory health effects (Seaton et al., 1995; Oberdorster et al., 1995; Schwartz et al., 1996; Peters et al., 1999; Michaels and Kleinman, 2000; Oberdorster, 2000; Pope, 2000; Dockery, 2001; Peters and Pope, 2002; Peters et al., 2004). Pollutants emit- ted by petrol-engined vehicles such as benzene and 1,3
A. McNabola et al. / Atmospheric Environment 42 (2008) 6496–6512 6497
butadiene have been shown to be associated with an in- creased risk of developing various cancers such as leukae- mia (WHO, 2000; Hughes et al., 2001). The transport sector has been shown to be responsible for significant pro- portions of these pollutants in the urban atmosphere, with particularly elevated concentrations measured in close proximity to roadways (Kaur et al., 2007).
It is important that personal exposure studies are carried out to quantify the health risk of individual urban commuters/dwellers in addition to the ambient air quality monitoring normally carried out by a regulatory body in most major cities. Indeed, ambient concentrations from fixed site monitors have been shown to significantly under- estimate or to have little or no association with the personal exposure of population groups such as commuters (Fernandez-Bremauntz and Ashmore, 1993; Adams et al., 2001b; Gulliver and Briggs, 2004). Equally, exposure concentrations have shown a good deal of variability from location to location around the globe and as a result, personal exposure from one location cannot necessarily be applied to another (Kaur et al., 2007).
Passengers in public buses have generally displayed the highest personal exposure to PM2.5 compared to other modes, followed by private car commuters, with modes such as the cyclist and pedestrian displaying the lowest personal exposure (Adams et al., 2001a,b,c; Chan et al., 2002a; Gomez-Perales et al., 2004; Kaur et al., 2005b). Car commuters have generally been shown to be exposed to the highest concentrations of VOC air pollution, while non-motorised modes such as the cyclist and pedestrian have been shown to experience lower exposure (Kingham et al., 1998; Rank et al., 2001; Lau and Chan, 2003; Gomez-Perales et al., 2004; Ballesta et al., 2006; O’Donog- hue et al., 2007).
Few studies have considered the personal exposure of both particulate matter and VOCs together. Measuring both types of pollutant was considered important in the current study to give a broad representation of atmospheric traffic emissions experienced by commuters, particularly since, as discussed, the relative exposures of different modes appear to differ for the two types of pollutant. No comprehensive study of commuter personal exposure to air pollution has been carried out in Ireland and this study examines four modes of transport which account for over 90% of commuters in the Dublin City region (DTO, 2004). A recent review of the personal exposure literature in transport microenvironments identified the following: very few studies have considered walking and cycling, the most common forms of transport globally; very few studies have examined the intra-mode variability of personal exposure; and often important influencing factors such as meteorological and traffic variables are omitted (Kaur et al., 2007). The present study examines both pedestrian and cyclist exposure, intra-mode variability and compre- hensively accounts for various meteorological and traffic- related variables. Furthermore, the current study has made direct comparisons between modes by recording samples simultaneously from the same starting point to the same finishing point, making the data directly compa- rable under identical atmospheric and traffic conditions, thus, resulting in a more accurate measure of relative
exposure than the comparison of the mean exposure as carried out in other similar studies.
In addition, few investigators have attempted to quantify or acknowledge the fact that differing breathing rates and exposure durations exist between motorised and non-motorised modes of transport. Rank et al. (2001) noted that cyclists breathe at a higher rate than car com- muters and therefore the exposure concentrations of the two commuter subgroups are not directly comparable as the cyclist would breathe in a larger volume of air than the car commuter over the same duration. O’Donoghue et al. (2007) attempted to quantify this by calculating the total mass of pollutants inhaled by cyclists compared to bus passengers. While the exposure of the cyclist was lower than the exposure of the bus passenger and the duration of exposure generally shorter, the total inhaled mass of the cyclist was higher than that of the bus passenger, due to the higher breathing rate. Calculating the total inhaled mass of any pollutant serves as a surrogate indicator of the actual deposition of particles or absorption of gases which are dependent on a number of chemical, physical and physiological properties (McNabola et al., 2006, 2007). The chemical properties of reactivity and solubility of a gas determine whether the gas will be absorbed in the lungs entirely, partially or not at all (ICRP, 1994; Nazaroff and Alvarez-Cohen, 2001; McNabola et al., 2006, 2007). These chemical factors together with the exposure concentrations and breathing parameters determine the extent and regional distribution of gases absorbed in the lungs. The physical size of particulate matter determines its deposition in the lungs as well as breathing rates and exposure concentrations (ICRP, 1994; Nazaroff and Alvarez-Cohen, 2001; McNabola et al., 2006, 2007).
A variety of more accurate methods of calculating the deposition of particulate pollution or the adsorption of gas- eous pollution in the lungs have been reported by previous investigators (Koblinger and Hofmann, 1985, 1990; ICRP, 1994; Salma et al., 2002; Hofmann et al., 2002; Gemci et al., 2003; McNabola et al., 2006, 2007). The use of com- putational fluid dynamics to model flow and reactions in the lungs has been reported (Gemci et al., 2003; Boyle, 2004). CFD models vary from simple 2D models with a symmetrical shape (real lungs are not symmetrical) to 3D models based on accurate measurements. However, even the most complex 3D CFD models do not represent the entire population. The geometry, volume and surface area of each lung varies from person to person. Therefore, the results of absorption and deposition from a CFD model are not readily transferable to the entire population exposed, but do give an indication of the general trends of deposition of particles and the absorption of gases.
Another method is the Monte Carlo lung deposition model, which uses a Monte Carlo simulation to randomly select a pathway through the lungs and predict the deposi- tion of particles along these pathways. The pathways are based on geometrical data of the lungs recorded extensively by previous investigators (Raabe et al., 1976). However, this model does not consider the absorption of gases and vapours into the lungs which form a major part of air pollution exposure for the urban commuter and therefore would only give partial information on the actual
A. McNabola et al. / Atmospheric Environment 42 (2008) 6496–65126498
uptake of pollutants (Koblinger and Hofmann, 1990, 1985; Hofmann et al., 2002; Salma et al., 2002).
Another modelling method which is representative of the entire population and accounts for gases as well as par- ticles has been developed by the International Commission on Radiological Protection (ICRP) called the human respira- tory tract (HRT) model (ICRP, 1994). This model predicts the deposition of particles and absorption for various degrees of physical activity, for both males and females of varying height, weight, age, etc. This model has been used in previ- ous investigations to assess the uptake of pollutants such as benzene and 1,3 butadiene in the lungs as well as particu- late matter (McNabola et al., 2006, 2007) and is used in this investigation to quantify the relative uptake of pollutants between modes of transport based on exposure measure- ments recorded in Dublin, Ireland.
2. Methodology
2.1. Field study and experiment design
A field study was carried out over an 18-month period on two fixed routes in Dublin city from January 2005 to June 2006. Sampling was undertaken only on weekdays (not including national/bank holidays) during peak traffic in the morning (08:00–09:00) and evening (17:00–18:00). Sampling was carried out on four modes of transport: the private car, public bus, cycling and walking on both routes, with the samples being evenly distributed between the modes throughout the sampling period, as shown in Fig. 1. All samples were collected in parallel between two modes on the same route, starting their journey from the same point at the same time and also finishing the journey at the same point (but generally at a different time, depending on traffic conditions). Logistic constraints on the number of volunteers available to carry out the samples and the number of personal samplers meant that only two
Jan 05
Feb 05
Mar 05
Apr 05
Jun 05
Sep 05
Nov 05
Dec 05
Jan 06
Feb 06
Mar 06
Apr 06
May 06
Jun 06
0
5
10
15
20
25
S a m
p l e C
o u
n t
Mode Car Cyclist Bus Pedestrian
Fig. 1. Sample distribution timeline.
joint samples could be carried out at any one time. There- fore each data point on one mode had a corresponding point on another mode allowing six direct exposure com- parison ratios, between the four modes, to be formulated on each route. The six comparisons were the following: cyclist/pedestrian (cyc/ped), car/pedestrian (car/ped), bus/ pedestrian (bus/ped), bus/cyclist (bus/cyc), bus/car (bus/ car) and car/cyclist (car/cyc).
An experiment design was carried out to estimate the number of samples required to achieve an accurate comparison between the modes based on the mean and standard deviation data obtained by previous investigators using similar sampling methods and in similar climates (Adams et al., 2001b; O’Donoghue et al., 2007). The number of samples required per comparison was estimated to be 12, which was increased to 15 to make the study more rig- orous and hence a total of 90 samples were required for the 6 comparisons on each route. Over the sampling period, however, a total of 194 valid pairs of direct comparison samples were recorded (388 individual samples of PM2.5
and VOCs). Samples were collected throughout the year across
a range of seasonal conditions as can be seen from Fig. 1. Meteorological and traffic data were recorded for each sample, meteorological data were obtained from the local monitoring authority’s regional weather station located some 15–18 km from the sampling routes. Traffic data were collected using induction loops located at traffic light junctions along both routes. Meteorological data were also recorded en route using a Nielsen-Kellerman Kestrel porta- ble weather station which was carried by the commuters, together with the air sampling equipment during each sample run.
2.2. Fixed route and time study – route selection
Dublin is located around a bay on the east coast of Ireland and has three main approaches to the city centre, from the south, north and west. Two sampling routes were selected, one approaching from the north and one approaching from the west, Fig. 2 shows the location of both the routes. The route approaching from the north, Route 1, was approximately 3 miles in length and ran from Trinity College Dublin in the city centre along the quays in the direction of the N3, a major commuter corri- dor. The quay area of Dublin is one of the most congested parts of the city centre. The roadway on Route 1 comprises two lanes of traffic, two footpaths, one bus lane and one cycle lane in either direction. Traffic flow in either direction is separated by the River Liffey over most of its length.
Route 2 was a slightly longer route, approximately 3.5 miles in length, which ran from Trinity College Dublin in the city centre, along Dame Street, and westwards to St. James Hospital. Route 2 was less consistently congested than Route 1 with variable amounts of traffic congestion throughout its length. The route comprised one bus lane, one cycle lane and 1–2 lanes of traffic in each direction. The general cross-sectional profile of Route 2 could be typ- ified by the urban street canyon, with tall buildings on either side of a relatively narrow roadway, compared to Route 1 which was much wider than the building heights
Fig. 2. Sampling routes.
A. McNabola et al. / Atmospheric Environment 42 (2008) 6496–6512 6499
on either side. Hence, the two routes chosen had signifi- cantly different geometry, orientation and overall levels of traffic congestion.
2.3. Sampling equipment
The sampling equipment used comprised two sets of PM2.5 personal samplers and two sets of VOC personal sam- plers. The PM2.5 personal sampler used was the high flow personal sampler (HFPS) which was recently developed specifically for the short-term sampling of PM2.5 in trans- port microenvironments (Adams et al., 2001c) and which has been used successfully in a number of commuter expo- sure studies (Adams et al., 2001a,b; Gomez-Perales et al., 2004; Kaur et al., 2005a,b). The HFPS is a mobile, high flow, gravimetric sampler designed to collect particulate matter with a 50% cut point of 2.5 mm equivalent aerody- namic diameter. This size selection is achieved using a porous foam plug specifically designed for the collection of PM2.5 at a high flow rate of 1 m3 h�1 (Vincent et al., 1993; Adams et al., 2001c). The porous foam plug prevents larger particles from reaching the filter paper on which the sam- ple is collected. The high flow pumps used were Casella Vortex Ultra-Flow pumps set at the beginning of each sam- ple at a flow rate of 1 m3 h�1 in order to achieve sufficient mass change on the filter and in order to achieve the correct cut point of 2.5 mm. Samples of at least 20 min duration were required to obtain sufficient particulate mass on filters. The sampling inlet used was a GSP sampler which was adapted internally to incorporate the porous foam plug and filter cassette as described in full by Adams et al. (2001c). The sampling inlet was placed in the breathing zone of the commuter during sampling by attaching it to the shoulder of each volunteer.
VOC samples were collected using an SKC universal pump and Vac-U-Chamber containing SKC Tedlar 1 l sam- pling bags. This method of sampling has also been recently developed specifically for mobile sampling of VOCs and has been successfully used in a number of personal exposure studies to date (McNabola et al., 2006; O’Donoghue et al.,
2007). This method of VOC sampling operates by connect- ing the 1 l bag sample to the outside environment in a sealed vacuum chamber from which air is drawn out by the universal pump. As a result of the chamber being evac- uated, the sampling bag inside fills with ambient air at a constant flow rate. The flow rate is set at the start of each sample so that the sampling bag will be at least 50% full at the end of each sample (400 ml of sample is required for analysis). Again, the sampling inlet was placed in the breathing zone of the commuter and all samples were ana- lysed within a 15 h period to avoid degradation of the sample.
2.4. Fixed route and time study – modes
A fixed sampling procedure was set for each of the four modes of transport investigated. For each sample on each mode the sampling vessels and filters were prepared prior to sampling. The sampling equipment was turned on at the beginning of the journey within the mode in question only, i.e. waiting for buses or walking to/from the car was not in- cluded. At the end of the sample run the equipment was turned off and the duration was noted. It should be noted that any samples compromised due to faults such as batter- ies running out; sampling bags not being opened; insuffi- cient sample volume/sampling time; or incorrect flow rates were discarded.
For the car commuters, the sampling equipment was placed on the passenger seat of the car and the sampling inlets were placed in the breathing zone of the driver. The portable weather meter was fixed to the exterior of the vehicle and the relative wind speed (relative to the motion of the car), temperature, pressure and humidity at 10 s intervals were recorded throughout the sample. All windows and vents were closed during sampling to restrict variations due to these parameters and neither vehicle used air conditioning. On Route 1 a 1994 diesel-engined Landrover was used for all the car commuter samples and on Route 2 a 1993 diesel-engined Nissan Vannette was used for all car commuter samples. Car commuter samples
A. McNabola et al. / Atmospheric Environment 42 (2008) 6496–65126500
typically lasted 20–45 min depending on traffic conditions, with a mean duration of 35 min.
The pedestrian samples were collected by carrying the sampling pumps in a small bag worn on the back of the vol- unteers. The sampling inlets were placed in the breathing zones of the pedestrian and the portable weather meter was also fixed to the shoulder of the volunteer in the free air stream. The pedestrian walked along the centre of the footpath adjacent to the incoming traffic during morning samples and adjacent to the outgoing traffic in the evening on each route. Average journey times for the pedestrian were 25–30 min.
Similarly, the cyclist and bus commuters both carried the sampling equipment in small back packs in the same fashion as the pedestrian. The bus commuter chose a seat at random, if any were available, or else stood on the lower floor of the bus. Windows and ventilation settings on the public bus were beyond the control of the volunteers and therefore varied from sample to sample. Buses were closed shell, double-decker, diesel-engined vehicles without air conditioning. Samples on the bus varied in duration from 20 to 45 min, depending on traffic conditions. The cyclist travelled in the cycle lanes (which were located in the bus lanes adjacent to the footpath) with the incoming traf- fic in the morning and outgoing traffic in the evening. The durations of the cyclist samples were the shortest of the four modes examined and were typically 20 min on both routes.
2.5. Intra-mode variability studies
Variability within modes was also examined to determine whether individual commuters could make small adjustments to their travel behaviour to reduce their exposure without changing modes. Alterations in the route of the pedestrian on Route 1 were examined by simulta- neously comparing pedestrians walking along the centre of the footpath as normal and pedestrians walking along ‘‘the boardwalk’’, a recently constructed promenade which overhangs the River Liffey and affords the pedestrian an extra 2–3 m distance from the traffic emission.
The effect of travelling from less congested areas to the city centre was investigated by extending the length of Route 1 to a less congested suburban area. This was carried out by taking a sample on Route 1 as normal and then taking a second sample starting at the end of Route 1 and finishing in a suburban area further north of the city centre. The exposures of both the car commuter and cyclist were evaluated using this procedure, which allowed a compari- son between urban and suburban commuting.
The effect of maintaining a larger inter-car distance, than normal, to the preceding vehicle in idling traffic was also investigated for car commuters. Here the effect of an inter-car spacing of approximately 2 m rather than 1 m was investigated, with the reduction in exposure due to increased dispersion between source and receptor being evaluated.
Finally, the effect on car commuter exposure of a single refuelling stop was investigated by collecting a number of samples in petrol stations while refuelling one of the cars used in the study. Samples were recorded for the time it
took to refill the vehicle (5–10 min) but only VOCs were taken in this case as the sampling time was too short to obtain an accurate PM2.5 sample.
2.6. Sample analysis
Sample analysis comprised two separate processes, gravimetric analysis for PM2.5 samples and gas chromatog- raphy for VOC samples. The gravimetric analysis was carried out by pre- and post-weighing a 37 mm diameter Teflon (PTFE) filter, with pore size 2 mm, on a Cahn C33 six figure microbalance with an accuracy of �0.6 mg. The weighing room was not fully environmentally controlled and it was therefore necessary to make adjustments for changes in air density. Temperature, pressure and relative humidity were recorded during pre- and post-weighing and adjustments were made as described by Koistinen et al. (1999). Electrostatic effects in the weighing of samples were minimised by using a Staticmaster containing an alpha particle emitting source of radiation, Po-210. The Staticmaster was placed in the weighing chamber of the microbalance and was used to ionise the air in the chamber and the filters being analysed. The microbalance was cali- brated before each weighing session using a calibration weight as specified by the manufacturers. Filters were equilibrated prior to pre-weighing for at least 24 h and again after sampling for at least 24 h. Filters were weighed at least three times and the average value was taken (if the three weights agreed to within 5 mg of each other). Filters were always handled with clean non-serrated tweezers to avoid contaminating the sample and field blanks were taken with every sample to minimise sampling and labora- tory errors. Any weight change recorded on the field blanks was subtracted from the samples. The analysis procedure adopted during the investigation followed closely that de- scribed by Koistinen et al. (1999) to ensure quality control in the analysis.
Gas chromatography (GC) was carried out using a Perkin Elmer ozone precursor monitoring system. The system comprised an Autosystem GC containing two capillary columns and two flame ionisation detectors (FIDs), an auto- matic thermal desorption unit (ATD 400) complete with sampling accessory and controlling hardware and software as well as analysis software. Helium was used as the carrier gas, while the combustion of the sample in the FIDs was achieved using zero air and hydrogen. The sample vessels were prepared prior to sampling by repeatedly flushing with zero air (up to 15 times) followed by evacuation and sealing. The analysis system was calibrated on a weekly basis throughout the sampling period using a calibration gas containing known quantities of VOCs produced and certified by the UK National Physics Laboratories. This monitoring laboratory has previously been used in the analysis of VOCs for various fixed site and personal expo- sure studies (Keating et al., 1998; Broderick and Marnane, 2002; McNabola et al., 2006; O’Donoghue et al., 2007).
The particular VOCs analysed for data collection were benzene and 1,3 butadiene for their association with traffic and for their adverse health effects. Concentrations of eth- ylene and acetylene were also noted for their strong asso- ciation with traffic, both compounds often regarded as
A. McNabola et al. / Atmospheric Environment 42 (2008) 6496–6512 6501
traffic markers (Borbon et al., 2003). In contrast, ethane concentrations were also noted as they are not associated with traffic and should not be expected to vary significantly between modes of transport.
2.7. Data analysis
A dataset for the entire sampling period was compiled in the statistical software package SPSS (v12.0.1), which was divisible into subgroups of data between the two routes and four modes, as well as between seasons, etc. Each sample in the dataset comprised the following vari- ables: date, time, mode, route, concentrations of PM2.5, benzene, 1,3 butadiene, ethylene, acetylene and ethane, wind speed and wind direction, temperature, precipitation, sunshine hours, pressure, relative humidity, traffic num- bers, relative wind speed, and idle time. The resulting 21�388 matrix was subsequently analysed for descriptive statistics and mean comparison tests were carried out to investigate statistically significant (or otherwise) differ- ences in the data.
2.8. Human respiratory tract modelling
The numerical mechanics of the human respiratory tract (HRT) model used in this investigation to estimate pollut- ant uptake have been described in detail in a number of previous investigations (ICRP, 1994; McNabola et al., 2006, 2007). In brief, the HRT model estimates the total absorption of pollutants in the lungs cumulatively over the exposure duration. The model takes into account the concentration of the pollutant, the duration of exposure, the breathing parameters, the physical properties and the chemical properties of solubility and diffusivity of the pol- lutant. The lungs are divided into five regions in which cumulative pollutant absorption/deposition is calculated as listed below.
� Extrathoracic One (ET1) – anterior nasal passages � Extrathoracic Two (ET2) – from the nose and mouth to
the larynx � Bronchial (BB) – the trachea and bronchi � Bronchiolar (bb) – the bronchioles and terminal
bronchioles
Table 1 Route 1 summary statistics (all samples)
Mode PM2.5 (mg m�3) Benzene (ppb) B
Car Mean 82.73 2.01 N 46 45 4 Standard deviation 44.84 1.29
Cyclist Mean 88.14 1.72 N 56 42 4 Standard deviation 61.54 0.72
Bus Mean 128.16 1.61 N 44 27 2 Standard deviation 68.08 0.80
Pedestrian Mean 63.45 1.35 N 48 37 3 Standard deviation 38.17 0.63
� Alveolar/Interstitial (AI) – the alveoli and interstitial tissues
Following the input of the relevant parameters and data the model gave outputs of the deposition or absorption of the pollutant in question in each of the five regions in series starting with ET1 on inhalation to AI and finishing with ET1
on exhalation. These regional depositions/absorptions were then plotted on a bar chart to display the uptake of pollutants through the lungs cumulatively over the expo- sure duration.
3. Results
3.1. Mode comparison study: January 2005–June 2006
As a preliminary overview of the observed exposure concentrations experienced by commuters on each of the modes on Routes 1 and 2, Tables 1 and 2 show the mean and standard deviation data considering all samples. The bus commuter was shown to have the highest exposure to PM2.5 on both routes and the car was shown to have the highest exposure to VOCs on both routes. The pedes- trian’s personal exposure was consistently the lowest of the four for both types of pollutant. The cyclist’s personal exposure was generally the second highest of the four with the exception of PM2.5 on Route 2. Statistically signif- icant differences in these mean concentrations were found between the modes for all pollutants (P> 0.01) except eth- ane (P¼ 0.212), indicating, as expected, that ethane is not associated with traffic emissions and does not vary between modes of transport. These mean concentrations are not directly comparable, however, as they are not sum- maries of samples taken under the same conditions. Greater differences were found in direct comparisons of paired exposure concentrations measured simultaneously, shown later in Table 5.
3.2. Daily variation
Table 3 shows that samples taken in the evening dis- played a higher mean exposure than samples taken in the morning. Statistically significant differences were found for the majority of pollutants between morning and
utadiene (ppb) Ethane (ppb) Ethylene (ppb) Acetylene (ppb)
0.79 13.21 7.32 4.36 5 45 45 45 0.58 8.50 9.67 2.95
0.54 6.97 8.83 5.10 2 42 42 42 0.19 3.77 4.55 3.13
0.54 4.14 8.46 5.27 7 27 27 27 0.24 2.28 4.07 2.08
0.40 8.21 6.18 3.32 7 37 37 37 0.17 7.05 4.50 2.05
Table 2 Route 2 summary statistics (all samples)
Mode PM2.5 (mg m�3) Benzene (ppb) Butadiene (ppb) Ethane (ppb) Ethylene (ppb) Acetylene (ppb)
Car Mean 88.95 1.95 1.18 8.84 13.26 9.90 N 44 42 42 38 38 38 Standard deviation 34.10 1.11 1.00 6.55 9.09 8.95
Cyclist Mean 71.61 1.54 0.48 13.05 4.92 4.79 N 48 43 43 42 42 43 Standard deviation 46.94 0.81 0.26 10.52 2.91 4.077
Bus Mean 103.81 1.31 0.47 17.56 4.08 3.67 N 45 45 45 45 45 45 Standard deviation 46.71 0.64 0.17 11.43 3.48 2.49
Pedestrian Mean 66.27 0.99 0.41 11.38 3.65 3.54 N 46 43 43 42 42 42 Standard deviation 42.80 0.56 0.26 6.91 2.09 2.39
A. McNabola et al. / Atmospheric Environment 42 (2008) 6496–65126502
evening samples, with the exception of ethane (PM2.5
P¼ 0.04; benzene P¼ 0.07; butadiene P¼ 0.05; ethane P¼ 0.60; ethylene P¼ 0.27; acetylene P¼ 0.05), indicating that the effect is greater for traffic-related pollutants.
3.3. Seasonal variation
Examining the seasonal variation (Table 4) of the exposure concentrations showed that while traffic-related VOC concentrations were generally higher in winter months than in summer, PM2.5 was higher in summer than in winter. Samples taken in autumn were too few in number to make any meaningful comparisons with the other seasons. Only ethylene displayed a statistically signif- icant difference between the seasons (P¼ 0.02), the other pollutants displayed low P values and may not have reached the significance level due to the small number of samples taken in autumn.
3.4. Direct comparison ratios
The study was designed to allow direct comparison of modal exposure under identical traffic and meteorological conditions. To this end, all samples were collected in pairs on two different modes, but on the same route and at the same time. Thus, a direct comparison ratio (DCR) was cal- culated from each pair of measurements. Six mean DCRs were calculated from the direct comparison of the four modes on each route, resulting from approximately 15–20 pairs of samples per comparison. These DCRs allow the comparison on relative exposure due to the choice of mode only as variations due to traffic volumes and meteorological conditions are controlled. Table 5 presents
Table 3 Daily variations in exposure concentrations
PM2.5 (mg m�3) Benzene (ppb) Bu
Morning Mean 80.87 1.48 N 198 172 17 Standard deviation 48.28 0.96
Evening Mean 92.09 1.66 N 179 152 15 Standard deviation 56.76 0.86
these mean DCRs for both routes and all pollutants measured.
While the mean PM2.5 exposure concentrations experienced by the car commuter and cyclist were shown to be quite similar (Tables 1 and 2), the DCRs (Table 5) indi- cate consistently larger differences in exposure. Similarly, where the mean 1,3 butadiene exposure concentrations experienced by the bus commuter and cyclist were shown to be similar (Tables 1 and 2), greater differences were again identified using the direct comparison. The difference in the comparison of the mean exposures and the DCRs highlights the importance of directly comparing modes of transport in studies of this nature.
3.5. Relative exposure factors
As sample analyses were carried out in parallel between two modes of transport, the DCRs only support comparison between two modes at any one time. A more comprehen- sive comparison could have been made if all modes were sampled simultaneously, but this was not logistically feasi- ble as described earlier. However, a direct comparison between all the modes can still be made using the data in Table 5 by treating each mode as an unknown variable and forming four sets of simultaneous equations for each pollutant. These simultaneous equations are then solved to obtain a set of relative exposure factors (REFs) which quantify the exposure of commuters on each mode relative to that of the pedestrian. An example of one such equation for PM2.5 is shown in Eq. (1) and the complete set of REFs is presented in Table 6.
tadiene (ppb) Ethane (ppb) Ethylene (ppb) Acetylene (ppb)
0.56 10.63 6.56 4.46 2 168 169 169 0.51 8.98 6.01 3.98
0.67 11.14 7.35 5.46 2 150 150 151 0.53 8.51 6.82 4.97
Table 4 Seasonal variations in exposure concentrations
Season PM2.5 (mg m�3) Benzene (ppb) Butadiene (ppb) Ethane (ppb) Ethylene (ppb) Acetylene (ppb)
Spring Mean 89.07 1.47 0.60 11.76 6.45 5.47 N 198 177 177 173 174 174 Standard deviation 57.35 0.83 0.59 9.72 4.60 5.10
Summer Mean 91.70 1.60 0.61 15.14 5.80 4.31 N 74 65 65 64 64 64 Standard deviation 40.63 1.17 0.50 7.47 10.05 4.50
Autumn Mean 110.20 1.45 0.42 4.57 9.56 4.35 N 4 8 8 8 8 8 Standard deviation 87.49 0.52 0.15 1.36 3.91 1.55
Winter Mean 75.60 1.78 0.65 5.68 8.78 4.24 N 101 74 74 73 73 74 Standard deviation 48.56 0.86 0.36 3.58 5.85 2.71
A. McNabola et al. / Atmospheric Environment 42 (2008) 6496–6512 6503
ð3�BusÞ ¼ ð2:66� carÞþ ð3:09� PedÞþ ð1:39� cycÞ (1)
3.6. Inter-route variation
The variation in exposure between the two routes was examined to investigate the effect of the differing levels of traffic, wind speed and canyon geometry on the personal exposure of commuters. The variation of exposure concen- tration between the two routes can be seen in Table 7, where exposure to PM2.5, benzene and ethylene was found to be higher on Route 1 than on Route 2. Statistically signif- icant differences were found for benzene (P¼ 0.01) and ethylene (P¼ 0.05). The higher exposure on Route 1 was due to higher traffic volumes and a lower mean wind speed during sampling. The lower mean wind speed during sam- pling on Route 1 (mean¼ 5.4 m s�1) was purely by chance and was found to have a statistically significant difference to the wind speed during sampling on Route 2 (mean¼ 6.0 m s�1).
3.7. Intra-pedestrian comparisons
Intra-pedestrian sample analyses were carried out on Route 1 where a deviation in the normal route of the pedes- trian was possible. Samples were recorded simultaneously by two pedestrians travelling along Route 1: the first pedestrian walking the route as normal and the second pedestrian deviating along the boardwalk. Pedestrians on the boardwalk were 2–3 m further away from the incoming traffic and thus a reduction in their exposure was found. Fig. 3 shows the exposure of the pedestrian on the footpath
Table 5 Direct comparison ratios – mean values
Route 1
PM2.5 Benzene Butadiene Ethylene Acetylene
Cyc/ped 1.41 1.18 1.30 1.36 1.42 Car/cyc 1.49 1.51 1.79 1.36 1.16 Bus/car 2.66 0.68 0.66 0.74 0.47 Bus/ped 3.09 1.28 1.25 1.29 1.00 Bus/cyc 1.38 1.08 1.14 1.25 1.27 Car/ped 1.34 2.84 3.51 2.64 3.43
to be 2.8 times that of the pedestrian on the boardwalk for PM2.5 and 1.8 times for benzene.
The relationship between the two pedestrian exposures, for both pollutants, can be seen to be approximately linear, indicating that the reduction in exposure between the two is almost entirely due to the increased dispersion afforded to the pedestrian on the boardwalk who is further away from the emission sources.
3.8. Intra-car comparisons
Comparisons were made between car commuters on Route 1, in which one car commuter drove normally while the second car maintained a larger distance (approximately 2 m or half a car length) to the vehicle in front, when idling in stationary traffic. As a large portion of the car commu- ter’s journey was spent in stationary, congested traffic it was hypothesised that a large portion of the car commu- ter’s exposure is associated with emissions from the preceding vehicle. Maintaining a larger distance between vehicles allows greater dispersion to take place and the car commuter’s exposure to be reduced. Fig. 4 shows the resulting reduction in exposure observed in the 10 pairs of samples measured.
An analysis of these data shows that the car commuter driving in traffic normally experienced a 30% higher expo- sure to PM2.5 and a 40% higher exposure to benzene, on average. Thus showing that, as for the intra-pedestrian comparison, commuter actions which increase the distance between the source of emissions and the receptor have significant effects on the reduction of their personal expo- sure to air pollutants.
Route 2
PM2.5 Benzene Butadiene Ethylene Acetylene
1.75 1.48 1.44 1.47 1.80 2.85 1.81 2.03 1.78 1.52 2.14 0.61 0.45 0.78 0.57 2.14 1.28 2.19 1.60 2.01 3.21 1.29 1.46 1.21 0.97 1.57 3.63 2.98 15.54 8.38
Table 6 Relative exposure factors
Pollutant Route1 Route 2
Bus Car Cyc Ped Bus Car Cyc Ped
PM2.5 3.25 1.66 1.62 1 3.92 2.32 1.26 1 Benzene 1.44 2.33 1.35 1 1.67 3.02 1.48 1 Butadiene 1.56 2.79 1.39 1 1.97 3.47 1.49 1 Ethylene 1.55 2.2 1.38 1 4.19 8.93 3.31 1 Acetylene 1.48 2.84 1.67 1 2.64 5.73 2.75 1
A. McNabola et al. / Atmospheric Environment 42 (2008) 6496–65126504
3.9. Inner city: suburban comparisons
Samples of car commuters and cyclists on an extended version of Route 1 gave a comparison between the exposure concentrations in the inner city section of the route com- pared to the more suburban section of the route (Table 8).
Inner city commuting conditions were shown to be associated with higher personal exposures for both the car commuter and cyclist as would be expected due to greater traffic congestion. This difference in exposure could also be due, in part, to the more open nature of the suburban section of the route, while the urban section is more enclosed by tall buildings. Significance tests car- ried out on the difference in exposure between the inner city and suburban samples on the bicycle reveals P values of 0.01, 0.09 and 0.07 for PM2.5, benzene and 1,3 butadi- ene, respectively, while the car commuter samples gave P values of 0.01, 0.06 and 0.08. The fact that benzene and butadiene in both cases failed to display significant differences was assumed to be due to insufficient sample number size.
3.10. Refuelling station exposure
Five VOC air quality samples were taken in petrol stations while refuelling, the samples were typically of 5 min duration and were recorded while refuelling the vehicle only. The mean exposure concentrations recorded were 225 ppb for benzene and 195 ppb for butadiene. These values considerably exceed the mean exposure of any of the modal exposures while commuting. Although this was an acute exposure occurring 1–2 times a week for the typical private car commuter, it is one which will add to the overall mean exposure of the car commuter on a long-term basis while modes such as the cyclist, pedestrian and bus passenger do not experience this exposure.
Table 7 Inter-route variation
PM2.5 (mg m�3) Benzene (ppb) Bu
Route 1 Mean 89.83 1.70 N 194 151 15 Standard deviation 58.79 0.95
Route 2 Mean 82.36 1.45 N 183 173 17 Standard deviation 45.22 0.87
3.11. HRT model outputs – benzene
Fig. 5 shows the results of the model predictions based on input data from Table 1. Summing the absorption in each region resulted in the highest total uptake for the car commuter, followed by the cyclist, pedestrian and bus com- muter (see Table 9). The car commuter showed the largest amount of absorption in region ET1, due to the high mean exposure concentration resulting in the highest concentra- tion gradient across the lungs of the four modes. The cyclist showed the highest absorption in the alveolar region where absorbed pollutants would directly enter the bloodstream, followed closely by the pedestrian. This was the case due to the high breathing rates for both the pedestrian and (more so) the cyclist.
3.12. Model outputs – 1,3 butadiene
Fig. 6 shows the results of the lung model calculations for the gaseous pollutant 1,3 butadiene. Again, due to the high concentration gradient the car commuter has the highest absorption in ET1 and the highest overall absorp- tion, followed by the cyclist, pedestrian and bus commuter (see Table 9). However, in the case of this highly soluble pollutant, the absorption occurs at the top of the respira- tory system and the pollutant does not progress deeper into the lungs as a result.
3.13. Model outputs – PM2.5
For PM2.5 the lung model shows the cyclist to have the highest total deposition followed by the bus commuter, pedestrian and car commuter (see Table 9). The car com- muter, who was previously shown to have the highest uptake of the two VOC pollutants is shown to have the low- est total deposition, due to his relatively low exposure to PM2.5 and low breathing rate Fig. 7.
tadiene (ppb) Ethane (ppb) Ethylene (ppb) Acetylene (ppb)
0.58 8.63 7.67 4.48 1 151 151 151 0.39 6.98 6.48 2.75
0.63 12.89 6.26 5.33 3 167 168 169 0.62 9.68 6.28 5.59
Fig. 3. Footpath/boardwalk pedestrian comparison: (a) PM2.5 exposure concentrations, (b) benzene exposure concentrations.
A. McNabola et al. / Atmospheric Environment 42 (2008) 6496–6512 6505
The bus commuter, whose exposure concentration to PM2.5 was notably higher than the other modes, is shown by the model to have the second highest total deposition over a typical commute. This was due to the low breathing rate and relatively short duration of the bus commuter. Although the pedestrian had the lowest exposure concen- tration and only a slightly elevated breathing rate, due to the relatively long duration the pedestrian’s total deposi- tion was shown to be the second lowest. Pedestrians with shorter duration commutes would clearly have a lower total uptake of pollutants in all cases.
3.14. Cumulative predictions including refuelling
Scaling up the lung model predictions to consider a period of 1 month of commuting which includes a realistic number of refuelling stops for car commuters was also examined. A number of samples of benzene were recorded during the experimental phase of the investigation in refu- elling stations for a duration of 5–10 min while refuelling.
These concentrations were extremely high with a mean concentration from five samples of 225 ppb for benzene. The predictions of the lung model were scaled up to include a morning and evening commute, 5 days a week, for 4 weeks for each mode. For the car commuter two refuelling stops of 5 min durationwere also included. Fig. 8 shows the results of the model’s predictions in this scenario which shows that the relative uptake of benzene remains the same as the pre- vious case but the car commuter’s absorption can be seen to considerably outweigh that of the other modes.
4. Discussion
Significant between-mode, between-route, seasonal and daily variations in the PM2.5 and VOC exposures of commuters in Dublin were observed over an 18-month pe- riod. The intra-mode variability of a number of modes was also highlighted. The relative uptake of pollutants was also highlighted in contrast to the relative personal exposures.
Fig. 4. Inter-car comparison: (a) PM2.5 exposure concentrations, (b) benzene exposure concentrations.
A. McNabola et al. / Atmospheric Environment 42 (2008) 6496–65126506
4.1. Between-mode comparison
The mean exposure levels for car, bus, cyclist and pedestrian commuters were shown to have statistically sig- nificant differences. The highest mean VOC exposure was
Table 8 Inner city:suburban comparison
Cyclist
PM2.5 (mg m�3) Benzene (ppb) Bu
Inner city Mean 118.94 2.05 0 N 10 10 10 Standard deviation 27.47 0.92 0
Suburban Mean 60.50 1.30 0 N 10 10 10 Standard deviation 15.78 0.95 0
observed for the car commuter followed by the cyclist, bus passenger and pedestrian, with the cyclist and bus pas- sengers’ mean exposure levels being quite similar. The highest mean PM2.5 exposure level was observed for the bus passenger followed by the cyclist, car commuter and
Car
tadiene (ppb) PM2.5 (mg m�3) Benzene (ppb) Butadiene (ppb)
.51 114.02 2.19 0.78 10 10 10
.19 34.37 1.27 0.56
.34 71.28 1.28 0.42 10 10 10
.19 32.80 0.64 0.26
Inhalation Exhalation
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
ET1 ET2 BB bb AI bb BB ET2 ET1 Lung Region
B e
n z e
n e
p
p b
Car Bus Cyclist Pedestrian
Fig. 5. Cumulative benzene absorption – comparison between modes.
A. McNabola et al. / Atmospheric Environment 42 (2008) 6496–6512 6507
pedestrian, although the car commuter and cyclist dis- played quite similar mean exposure levels. The high mean VOC exposure of car commuters may be due to their road position, located closest of the four modes to the location where the majority of VOC emissions originate (in the main traffic lanes close to car exhausts). Similarly the bus commuter’s high exposure to PM2.5 may be explained by his proximity to the particle-laden exhaust emissions of other buses in the bus lanes. In contrast, the pedestrians’ low exposure to both types of pollutants may be due to their location on the footpaths, where they would be afforded more dispersion. This investigation highlights the importance of considering both petrol and diesel de- rived pollutants as their relative exposure concentrations have been shown to differ between modes.
This pattern of relative exposure agrees with similar investigations in other locations. In the UK, where a number of studies have been carried out mostly on particulate exposure, much lower mean exposure to PM2.5 has been observed. Bus and car commuters, and cyclists in London have been reported to be exposed to mean PM2.5 concen- trations of 39, 37.7 and 34.5 mg m�3, respectively (Adams et al., 2001b). An investigation of PM2.5 exposure in buses in Hong Kong however, gave similar mean concentrations to those reported here: 93–97 mg m�3 (Chan et al., 2002a). An examination of particulate exposure in Guangzhou, China on buses and cars also showed similar concentrations of 145 and 73 mg m�3, respectively (Chan et al., 2002b). The reasons behind the variation in levels between geographi- cal locations are unclear and may be due, in part, to differ- ences in the method of PM2.5 measurement employed in
Table 9 Total absorption/deposition by mode and pollutant
Mode Benzene (ppb) 1,3 Butadiene (ppb) PM2.5 (mg)
Car 0.63 0.75 16.73 Bus 0.26 0.28 21.83 Cyclist 0.51 0.52 23.06 Pedestrian 0.46 0.46 17.35
different studies, but the generally consistent findings on the relative exposure of modes are clear.
Short-term investigations of commuter exposure to benzene and 1,3 butadiene have previously been carried out in Dublin. An investigation of cyclist and bus commuter exposure consisting of 14 samples collected over a 2-week period gave mean exposures to benzene of 1.6 ppb for the cyclist and 2.2 ppb for the bus commuter (O’Donoghue et al., 2007). The cyclist concentration was similar to the mean exposure found in the current investigation, but the mean bus commuter exposure was higher. This may be due to the fact that this short-term study was carried out in the month of February when weather conditions are colder, resulting in higher emissions. A comparison of cy- clists and car commuters in Copenhagen, Denmark, found higher benzene concentrations of 4.4 ppb for car com- muters and similar concentrations of 1.6 ppb for cyclists (Rank et al., 2001). However, concentrations of benzene in a comparison of six European cities have been found to be lower in Dublin than in other major cities (Ballesta et al., 2006).
4.2. Direct comparison ratios
Forming DCRs for each of the paired samples recorded and compiling their mean values revealed greater differ- ences between the modes than was apparent when com- paring overall mean exposure concentrations. Specifically, the exposures of car commuter and cyclist were shown to differ to a greater extent than indicated by the overall mean concentrations. The differences between the two measures highlight the importance of using direct compar- isons between modes when investigating relative expo- sure. Factors such as traffic congestion, emission rates and atmospheric conditions vary constantly and by carrying out simultaneous sampling of pairs of modes, the influence of these factors is reduced.
The DCRs confirm that the pedestrian experienced the lowest exposure of the four modes investigated. The
B u
t a
d i e
n e
p
p b
Inhalation Exhalation
ET1 ET2 BB bb AI bb BB ET2 ET1 Lung Region
Car Bus Cyclist Pedestrian
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Fig. 6. Cumulative 1,3 butadiene absorption – comparison between modes.
A. McNabola et al. / Atmospheric Environment 42 (2008) 6496–65126508
pedestrian’s exposure to PM2.5 is shown to be considerably lower than the bus commuter’s and the pedestrian’s expo- sure to VOCs is shown to be considerably lower than the car commuter’s. Smaller differences between the modes were found for the car/cyclist, cyclist/pedestrian and bus/cyclist comparisons, but they are larger than the mean exposure would suggest.
4.3. Relative exposure factors
Forming simultaneous equations from the matrix of DCRs leads to a set of relative exposure factors that allow comparison of all four modes simultaneously. These REFs show the pedestrian to have the lowest relative exposure to all pollutants. The bus commuter’s exposure to PM2.5
was 3–4 times higher than the pedestrian, while the car commuter’s exposure to VOCs 2–3 times higher than the
0
2
4
6
8
10
12
14
16
18
20
P M
2 . 5
u
g
Inhalation Ex
ET1 ET2 BB bb AI Lung R
Fig. 7. Cumulative PM2.5 absorption –
pedestrian. A similarity was noted in the REFs for the VOCs, particularly on Route 1, indicating that these pollut- ants are linked and are likely to originate from the same source. The commuter’s exposure to traffic-related VOCs on Route 1 was highest in the car, followed by the bus, cyclist and pedestrian (with the exception of acetylene for which the cyclist is higher than the bus commuter). On Route 2, the car was again shown to have the highest REFs for all traffic-related VOCs and the pedestrian was the lowest. The exposure of the bus passenger was gener- ally higher than that of the cyclist with the exception of acetylene for which the REFs are similar.
The REFs reflect the findings of studies of this nature carried out in Dublin and other cities. The car commuter’s exposure to traffic-related VOCs could be expected to be the highest due to his position in the traffic lanes, whereas the bus passenger and cyclist could be expected to have
halation
Car Bus Cyclist Pedestrian
bb BB ET2 ET1 egion
comparison between modes.
0
5
10
15
20
25
30
35
B e
n z e
n e
p
p b
ET1 ET2 BB bb AI bb BB ET2 ET1 Lung Region
Inhalation Exhalation
Car Bus Cyclist Pedestrian
Fig. 8. One-month cumulative benzene absorption including refuelling stops – comparison between modes.
A. McNabola et al. / Atmospheric Environment 42 (2008) 6496–6512 6509
similar exposure levels due to their shared road position. The cyclist is located slightly further away from traffic emis- sions than the bus which may explain why the cyclist expo- sure is more often lower than that of the bus commuter.
4.4. Seasonal variations
Concentrations of VOCs were found to be higher in colder winter/autumn conditions than in summer, although only ethylene showed statistically significant differences, while PM2.5 concentrations were higher in summer. These higher exposure concentrations in the colder conditions of winter may be due to cold start emis- sions (Taylor and Fergusson, 1997), while temperatures during summer conditions in Ireland seldom reach high enough levels to induce evaporative emissions on a regular basis. Previous investigations in London, UK, also found PM2.5 exposure to be higher in summer than in winter (Adams et al., 2001b).
4.5. Between-route comparison
Comparing the two routes, the more consistently congested Route 1 displayed higher mean concentrations of pollutants. This was due, in part, to the traffic volumes and congestion on the route but also due to a lower mean wind speed during sampling on Route 1 compared to Route 2. Previous investigations of between-route variability in the UK have shown significant differences between highly congested and less congested routes (Kingham et al., 1998; Adams et al., 2001b).
The relative exposure factors displayed more difference between the modes on Route 2 than on Route 1. This could be explained by the differing geometry of the routes, Route 2 is typical of the urban street canyon geometry in cross- section and the prevailing wind direction is at approximately 45� to the direction of the street. Thus, a street canyon vortex is likely to be formed, transporting pollutants towards some traffic lanes and away from
others. Route 1 is more open in cross-section, its width being considerably longer than its height due to the River Liffey dividing the incoming and outgoing traffic. Thus vortices are unlikely to dominate the dispersion of the pol- lutants, which are instead more directly exposed to the ac- tion of the wind.
4.6. Daily variation
Significantly higher concentrations of pollutants were generally found for evening commutes compared to morn- ing commutes during the sampling campaign. These higher concentrations could be explained by an accumulation of ambient concentrations throughout the day in the city cen- tre. Pollutants such as the traffic-related VOCs and PM2.5
are not persistent in the environment and have atmospheric half lives due to reaction in the environment. Butadiene for example has a half life of 0.24–1.9 days and as such it is clear how such a pollutant could accumulate with constant emissions from traffic throughout the day, result- ing in higher personal exposures in the evening samples (Hughes et al., 2001).
The prevailing wind being south westerly in direction could also result in higher concentrations on the leeward side of the street than the windward side as a result of street canyon vortices and hence a higher exposure for evening traffic than morning traffic as shown in Fig. 9.
4.7. Intra-mode comparisons
In general the intra-mode comparison studies showed that individual commuters could take small steps to in- crease the distance between source and receptor, thereby reducing their exposure to air pollutants by a significant proportion. Pedestrians who took the boardwalk were shown to have a reduced exposure relative to pedestrians on the footpath, who were closer to the traffic. This finding could be extended to pedestrians who take quieter side streets or diversions through parks, etc. as part of their
Fig. 9. Canyon cross-section.
A. McNabola et al. / Atmospheric Environment 42 (2008) 6496–65126510
route. Similarly cyclists who avoid the main transport corridors could also reduce their exposure. Car commuters in idling congested traffic were shown to have a lower exposure if they maintained a larger than normal distance (an extra 1 m) between their car and the preceding vehicle. Widespread adoption of this measure could result in dis- ruption to traffic as the number of vehicles on each section of roadway would be reduced. However, relocating the ven- tilation inlet of a private car away from the exhaust of the preceding vehicle, to roof level for example, could have the same effect. Similarly relocating the exhaust of vehicles to roof level as is the case on some large HGVs could have a similar effect.
4.8. Uptake of pollutants
Taking into account the physiological differences between pedestrians, cyclists, bus and car commuters, the HRT model predicts the car commuter to have the highest relative uptake of VOCs but the lowest relative uptake of PM2.5. Therefore, the car as a modal choice retains its unat- tractive position in terms of the health impacts from VOC air pollution but not from particulate air pollution.
The pedestrians had a similar trip duration to the car commuter but their exposure to PM2.5 was lower. As a result of the elevated physiological state of the pedestrian the model yields the second lowest relative uptake in this case. As the trip duration for the car and pedestrian are similar it would lead one to conclude that the elevated breathing rate is primarily responsible for the pedestrian’s higher PM2.5 uptake. However, the trip duration can be shown to have a higher impact on pollutant uptake for commuters breathing at higher rates than at lower rates for equal exposure concentrations, as shown in Section 4.9. In terms of VOC pollution the pedestrian was also predicted by the model to have the second lowest uptake for both benzene and butadiene. The pedestrian is widely accepted as having the lowest exposure to air pollutants in the urban transport environment. However, conclusions that this low exposure results in a lowest health impact rel- ative to all other modes has been shown by the model’s
predictions to be unfounded as a result of higher breathing rates.
The bus commuter was reported as having the highest mean exposure to PM2.5, experiencing concentrations sig- nificantly higher than the other modes. However, compared to the cyclist and pedestrian, the bus commuter sits on the bus in a state of rest and the mean duration of exposure in this case was relatively short. The model pre- dicts the bus commuter to have the second highest uptake of PM2.5 in this case, while the cyclist is shown to have the highest. Therefore, while a high exposure to PM2.5 is pres- ent in buses, the bus commuter’s health does not appear to be as highly impacted as other more physically exerting modes. In terms of VOC air pollution, the bus commuter was reported to have a low exposure and coupled with the factors of breathing rate and duration (as mentioned above), the results show that the bus commuter would have the lowest uptake of VOC air pollution relative to the other modes. Therefore, in terms of minimising one’s health impact from air pollution while commuting, the bus can be seen to be an attractive modal choice, as both VOC and (to a lesser extent) PM2.5 uptake was shown to be low.
The cyclist, with a significantly higher breathing rate than the pedestrian, is predicted by the model to have the highest uptake of PM2.5 and the second highest uptake of VOCs, despite having a short commute duration. The cyclist’s mean exposure concentration was second highest for both PM2.5 and VOCs and therefore considering the health impacts from both types of air pollutant the cyclist as a modal choice appears to be most damaging relative to the other modes.
4.9. Duration of exposure
The duration of exposure is an important factor in the uptake of pollutants for each mode, the longer the time spent commuting the larger the total uptake. However the trip duration appears to be a more important factor for the modes which require higher breathing rates as shown in Fig. 10. Plotting the uptake of benzene with equal exposure concentrations for differing breathing rates against duration shows that the uptake of those performing mild exercise increases with duration at a higher rate than those at rest. The mean durations of exposure used here could certainly differ for individual commuters, but it can be seen from Fig. 10 that a lower or higher duration for pe- destrian and cyclist would have a larger influence on their total uptake than for the bus or car commuter.
4.10. Comparison of relative uptake and relative exposure
The findings of the lung model for relative uptake can influence the modal choice of commuters in a different manner to the relative exposure concentrations. The rela- tive exposure taken in isolation would suggest that those with a low exposure such as the pedestrian and cyclist would suffer less health impacts from air pollution than those with higher exposure such as the car and bus com- muters. However, taking into account the physical activities of each commuter and their duration of exposure shows
0
0.5
1
1.5
2
2.5
0 10 20 30 40 50 60 Duration (min)
B e n
z e n
e ( p
p b
)
at rest mild exercise
Fig. 10. Pollutant uptake vs. exposure duration at high and low breathing rates.
A. McNabola et al. / Atmospheric Environment 42 (2008) 6496–6512 6511
that those performing higher levels of physical exertion have a high uptake of pollutants. The relative uptake findings portray the bus in a more positive light than the exposure concentrations would suggest. The pedestrian health impacts are shown to be worse than previously believed but remains a good modal choice relative to the cyclist, car and (for PM2.5 uptake) bus commuters. The car commuter remains a poor modal choice in terms of air pol- lution health impacts and the cyclist is also shown to have negative health impacts relative to the other modes. The health benefits of cycling and walking are, of course, not addressed by the model and could negate some of the impacts of pollutant uptake while commuting (Wonga et al., 2007).
4.11. Regional distribution of pollutants
The regional distribution of the pollutants in the lungs was shown by the model to vary according to pollutant type. For benzene, a gaseous pollutant with low solubility, the majority of the pollutant is carried to the alveolar region of the lungs were it would directly enter the blood- stream. Where the concentration gradient is high (as for the car) a significant proportion of the benzene is absorbed in fluids of the extrathoracic regions of the lungs, from where it would be transferred by the lungs to the mouth (to be swallowed or expectorated). For butadiene, a highly soluble pollutant, the inhaled concentration is almost en- tirely absorbed in the extrathoracic region and only a small proportion is transferred deeper into the lungs under higher breathing rates. For benzene 60–80% of the pollut- ant is retained in the lungs depending on the breathing rate while all of the inhaled butadiene is absorbed. PM2.5
is primarily deposited in the lower extrathoracic and bronchial regions of the lungs, with a very small fraction progressing deeper into the lungs.
4.12. Refuelling stops
Refuelling stops were not shown by the model to alter the relative uptake of benzene for the car compared to other modes as the car was already the mode with highest
absorption. However a significant increase in the total uptake of benzene is notable compared to other modes and this would clearly add to the health impacts of the pri- vate car as a modal choice.
5. Conclusions
Personal exposure levels of PM2.5 and VOCs have been measured for pedestrians, cyclists, private car users and bus passengers in central Dublin during peak traffic congestion, in a comprehensive, long-term multi-modal transport exposure study. Mean exposure levels across the four modes were shown to vary to a statistically signif- icant degree and mean exposure concentrations were shown to be generally higher on the more congested Route 1 than on Route 2. Direct comparison ratios and relative exposure factors formulated from simultaneous sampling of pairs of modes showed that the personal exposure of dif- ferent commuters was underestimated by the sample means of modal exposure concentration. They also showed a greater degree of difference between the modes on Route 2 than on Route 1 due to differences in street cross-section. Bus commuters were shown to have the highest exposure to PM2.5 while car commuters were shown to have the highest exposure to VOCs. The pedestrian was consistently found to experience a lower exposure to both types of pol- lutant than the car commuter, bus passenger or cyclist. An investigation of intra-mode variability showed that indi- vidual commuters regardless of their mode could reduce their personal exposure while commuting by taking small steps to increase the distance between the source and the receptor during their journey.
The predictions of the lung model have shown the relative uptake of pollutants to differ from the relative exposure to pollutants for the car commuter, bus passenger, pedestrian and cyclist. The model shows both the car commuter and cyclist to be associated with high uptake of pollutants relative to the pedestrian and bus commuter with lower uptake of pollutants. The investigation highlights that the exposure concentrations of commuter subgroups performing different levels of physical activity are not directly comparable. The length of the commute was shown to be a more important factor for the cyclist and pedestrian than for the car and bus commuters due to their increased breathing rates. The added effect of refu- elling stops was also shown to increase the health impacts for users of the private car as a modal choice, relative to the bus, cyclist and pedestrian.
Acknowledgements
The authors would like to thank the Centre for Transport Research (PRTLI programme) and the ETI project (ERTDI programme), Trinity College Dublin for funding this investigation.
References
Adams, H.S., Niewenhuijsen, M.J., Colvile, R., 2001a. Determinants of fine particle (PM2.5) personal exposure levels in transport microenviron- ments, London, UK. Atmospheric Environment 35, 4557–4566.
A. McNabola et al. / Atmospheric Environment 42 (2008) 6496–65126512
Adams, H.S., Kenny, L.C., Niewenhuijsen, M.J., Colvile, R.N., McMullen, M. A.S., Khandelwal, P., 2001b. Fine particle (PM2.5) personal exposure levels in transport microenvironments, London, UK. The Science of the Total Environment 279, 29–44.
Adams, H.S., Kenny, L.C., Niewenhuijsen, M.J., Colvile, R.N., Gussman, R.A., 2001c. Design and validation of a high-flow personal sampler for PM2.
5. Journal of Exposure Analysis and Environmental Epidemiology 11, 5–11.
Ballesta, P.P., Field, R.A., Connolly, R., Cao, N., Caracena, A.B., De Saeger, E., 2006. Population exposure to benzene: one day cross sections in six European cities. Atmospheric Environment 40, 3355–3366.
Borbon, A., Fontaine, H., Locoge, N., Veillerot, M., Galloo, J.C., 2003. Developing receptor oriented methods for non methane hydrocarbon characterization in urban air: part II source apportionment. Atmospheric Environment 37, 4065–4076.
Boyle, F., 2004. Computational fluid dynamics in engineering. The Engineers Journal 58, 112–116.
Broderick, B.M., Marnane, I.S., 2002. A comparison of the C2–C9 hydrocar- bon compositions of vehicle fuels and urban air in Dublin, Ireland. Atmospheric Environment 36 (6), 975–986.
Chan, L.Y., Lau, W.L., Lee, S.C., Chan, C.Y., 2002a. Commuter exposure to particulate matter in public transportation modes in Hong Kong. Atmospheric Environment 36, 3363–3373.
Chan, L.Y., Lau, W.L., Zou, S., Cao, Z., Lai, S., 2002b. Exposure levels to carbon monoxide and respirable suspended particulate in public transportation modes while commuting in urban area of Guangzhou, China. Atmospheric Environment 33, 3363.
Dockery, D.W., 2001. Epidemiological evidence of cardiovascular effects of particulate air pollution. Environmental Health Perspectives 109 (S4), 483–489.
DTO (Dublin Transport Office), 2004. Road User Monitoring Report. Fernandez-Bremauntz, A.A., Ashmore, M.R., 1993. Exposure of commuters
to carbon monoxide in Mexico City. Atmospheric Environment 29, 525–532.
Gemci, T., Shortall, B., Allen, G.M., Corcoran, T.E., Chigier, N., 2003. A CFD study of the throat during aerosol drug delivery using heliox and air. Journal of Aerosol Science 34 (9), 1175–1192.
Gomez-Perales, J.E., Colvile, R.N., Nieuwenhuijsen, M.J., Fernandez- Bremauntz, A., Gutierrez-Avedoy, V.J., Paramo-Figueroa, V.H., Blanco-Jimenez, S., Beuno-Lopez, E., Mandujano, F., Bernabe- Cabanillas, E., Ortiz-Segovia, E., 2004. Commuter exposure to PM2.5, CO and benzene in the public transport in the metropolitan area of Mexico City. Atmospheric Environment 38, 1219–1229.
Gulliver, J., Briggs, D.J., 2004. Personal exposure to particulate air pollution in transport microenvironments. Atmospheric Environment 38 (1), 1.
Hofmann, W., Asgharian, B., Winkler-Heil, R., 2002. Modeling intersubject variability of particle deposition in human lung. Journal of Aerosol Science 33 (2), 219–235.
Hughes, K., Meek, M.E., Walker, M., Beauchamp, R., 2001 1,3 Butadiene: human health aspects. In: Concise International Chemical Assessment Document 30, WHO, Geneva.
ICRP (International Commission on Radiological Protection Task Group), 1994. Human respiratory tract model for radiological protection. Annals of the ICRP 24 (66), 1–3.
Kaur, S., Nieuwenhuijsen, M.J., Colvile, R.N., 2007. Fine particulate matter and carbon monoxide exposure concentrations in urban street trans- port microenvironments. Atmospheric Environment 41 (23), 4781– 4810.
Kaur, S., Nieuwenhuijsen, M.J., Colvile, R.N., 2005a. Pedestrian exposure to air pollution along a major road in central London, UK. Atmospheric Environment 39, 7307–7320.
Kaur, S., Nieuwenhuijsen, M.J., Colvile, R.N., 2005b. Personal exposure of street canyon intersection users to PM2.5, ultra fine particle counts and CO in central London, UK. Atmospheric Environment 39, 3629– 3641.
Keating, D., Marnane, I., Misstear, B.D.R., Broderick, B., 1998. An ad- vanced air pollution monitoring system. Engineers Journal 52 (3), 47–49.
Kingham, S., Meaton, J., Sheard, A., Lawrenson, O., 1998. Assessment of exposure to traffic related fumes during the journey to work. Transport Research Part D 3 (4), 271–274.
Koblinger, L., Hofmann, W., 1990. Monte Carlo modeling of aerosol depositions in human lungs. Part 1: simulation of particle transport in stochastic lung structure. Journal of Aerosol Science 21 (5), 661–674.
Koblinger, L., Hofmann, W., 1985. Analysis of human lung morphometric data for stochastic aerosol deposition calculations. Physics in Medicine and Biology 30 (6), 541–556.
Koistinen, K.J., Kosua, A., Tenhola, V., Hanninen, O., Jantunten, M.J., Oglesby, L., Kuenzli, N., Georgoulis, L., 1999. Fine particle (PM2.5) measurement methodology, quality assurance procedure and pilot results of the EXPOLIS study. Journal of the Air and Waste Manage- ment Association 49, 1212–1220.
Lau, W.L., Chan, L.Y., 2003. Commuter exposure to aromatic VOCs in public transport modes in Hong Kong. The Science of the Total Environment 308, 143–155.
Michaels, R.A., Kleinman, M.T., 2000. Incidence and apparent health significance of brief airborne particle excursions. Aerosol Science and Technology 32, 93–105.
McNabola, A., Broderick, B.M., Johnston, P., Gill, L.W., 2006. Effects of the smoking ban on benzene and 1,3 butadiene levels in pubs in Dublin. Journal of Environmental Science and Health, Part A 41, 799–810.
McNabola, A., Broderick, Gill, L.W., 2007. Optimal cycling and walking speed for minimum absorption of traffic emissions in the lungs. Journal of Environmental Science and Health, Part A 42 (13).
Nazaroff, W.W., Alvarez-Cohen, L., 2001. Environmental Engineering Science. John Wiley & Sons Publishing.
Oberdorster, G., Gelein, R.-M., Ferin, J., Weiss, B., 1995. Involvement of ultrafine particles. Inhalation Toxicology 7, 111–124.
Oberdorster, G., 2000. Pulmonary effects of inhaled ultrafine particles. International Archives of Occupational Environmental Health 74, 1–8.
O’Donoghue, R.T., Gill, L.W., McKevitt, R.T., Broderick, B.M., 2007. Exposure to hydrocarbon concentrations while commuting or exercising in Dublin. Environment International 33, 1–8.
Peters, A., Von Klot, S., Heifer, M., Trentingalia, I., Hormann, A., Wichmann, H.E., 2004. Exposure to traffic and the onset of myocardial infarction. New England Journal of Medicine 351 (17), 1721.
Peters, A., Perz, S., Doring, A., Stieber, J., Koenig, W., Wichmann, H., 1999. Increases in heart rate during an air pollution episode. American Journal of Epidemiology 150 (10), 1094–1098.
Peters, A., Pope, C.A., 2002. Cardiopulmonary mortality and air pollution. The Lancet 360 (9341), 1184–1185.
Pope, C.A., 2000. Review: epidemiological basis for particulate air pollution health standards. Aerosol Science and Technology 32, 4–14.
Raabe, O.G., Yeh, H.C., Schum, G.M., Phalen, R.F., 1976. Tracheobronchial geometry: human, dog, rat, hamster. Lovelace Foundation for Medical Education and Research, Report No. LF-53.
Rank, J., Folke, J., Homann-Jespersen, P., 2001. Differences in cyclists and car drivers exposure to air pollution from traffic in the city of Copen- hagen. The Science of the Total Environment 279, 131–136.
Salma, I., Balashazy, I., Hofmann, W., Zaray, G., 2002. Effects of physical exertion on the deposition of urban aerosols in the human respiratory system. Journal of Aerosol Science 33, 983–997.
Schwartz, J., Dockery, D.W., Neas, L.M., 1996. Is daily mortality associated specifically with fine particles. Journal of the Air and Waste Manage- ment Association 46, 927–939.
Schweizer, C., Edwards, R.D., Bayer-Oglesby, L. Time spent in traffic across Europe. Environmental Health Perspectives, in press.
Seaton, A., MacNee, W., Donaldson, K., Gooden, D., 1995. Particulate air pollution and acute health effects. The Lancet 345, 176–178.
Taylor, D., Fergusson, M., 1997. Road User Exposure to Air Pollution: Liter- ature Review. Environmental Transport Association.
Vincent, J.H., Aitken, R.J., Mark, D., 1993. Porous plastic foam filtration media: penetration characteristics and applications in particle size- selective sampling. Journal of Aerosol Science 24 (7), 929–944.
Wonga, C.M., Oua, C.Q., Thacha, T.Q., Chaua, Y.K., Chana, K.P., Hoa, S.Y., Chunga, R.Y., Lam, T.H., Hedleya, A.J., 2007. Does regular exercise protect against air pollution-associated mortality? Preventive Medicine 44, 386–392.
WHO (World Health Organisation), 2000. Air Quality Guidelines, second ed. In: WHO European Series No. 91. WHO Regional Office for Europe, Copenhagen.
- Relative exposure to fine particulate matter and VOCs between transport microenvironments in Dublin: Personal exposure and uptake
- Introduction
- Methodology
- Field study and experiment design
- Fixed route and time study - route selection
- Sampling equipment
- Fixed route and time study - modes
- Intra-mode variability studies
- Sample analysis
- Data analysis
- Human respiratory tract modelling
- Results
- Mode comparison study: January 2005-June 2006
- Daily variation
- Seasonal variation
- Direct comparison ratios
- Relative exposure factors
- Inter-route variation
- Intra-pedestrian comparisons
- Intra-car comparisons
- Inner city: suburban comparisons
- Refuelling station exposure
- HRT model outputs - benzene
- Model outputs - 1,3 butadiene
- Model outputs - PM2.5
- Cumulative predictions including refuelling
- Discussion
- Between-mode comparison
- Direct comparison ratios
- Relative exposure factors
- Seasonal variations
- Between-route comparison
- Daily variation
- Intra-mode comparisons
- Uptake of pollutants
- Duration of exposure
- Comparison of relative uptake and relative exposure
- Regional distribution of pollutants
- Refuelling stops
- Conclusions
- aclink2
- blink1
memo3/Robinson.pdf
A
O h c C
d s c m c a ©
K
1
h s t p b s c
2
s c n w
0 d
Accident Analysis and Prevention 39 (2007) 86–93
Bicycle helmet legislation: Can we reach a consensus?
D.L. Robinson ∗ c/o Prof JSF Barker Building, Trevenna Rd, Armidale, NSW 2351, Australia
Received 4 January 2006; received in revised form 12 June 2006; accepted 13 June 2006
bstract
Debate continues over bicycle helmet laws. Proponents argue that case-control studies of voluntary wearing show helmets reduce head injuries. pponents argue, even when legislation substantially increased percent helmet wearing, there was no obvious response in percentages of cyclist ospital admissions with head injury—trends for cyclists were virtually identical to those of other road users. Moreover, enforced laws discourage ycling, increasing the costs to society of obesity and lack of exercise and reducing overall safety of cycling through reduced safety in numbers. ountries with low helmet wearing have more cyclists and lower fatality rates per kilometre. Cost-benefit analyses are a useful tool to determine if interventions are worthwhile. The two published cost-benefit analyses of helmet law
ata found that the cost of buying helmets to satisfy legislation probably exceeded any savings in reduced head injuries. Analyses of other road afety measures, e.g. reducing speeding and drink-driving or treating accident blackspots, often show that benefits are significantly greater than osts. Assuming all parties agree that helmet laws should not be implemented unless benefits exceed costs, agreement is needed on how to derive
onetary values for the consequences of helmet laws, including changes in injury rates, cycle-use and enjoyment of cycling. Suggestions are made
oncerning the data and methodology needed to help clarify the issue, e.g. relating pre- and post-law surveys of cycle use to numbers with head nd other injuries and ensuring that trends are not confused with effects of increased helmet wearing.
2006 Elsevier Ltd. All rights reserved.
c s e f n c s s
m w p l l b
eywords: Bicycle; Helmet; Legislation; Head injury; Case-control; Bias
. Introduction
Hagel and Pless (2006) criticised evidence against bicycle elmet legislation (Curnow, 2005) arguing that large population tudies of the effects of helmet laws provide weaker evidence han case-control studies. The two sources of data are com- ared and discussed, along with what information should ideally e collected to provide the best possible evaluation and under- tanding of helmet legislation, including effects related to risk ompensation or reduced safety in numbers.
. Case-control studies
Serious problems in the methodology of analysing self-
elected samples came to light after publication of randomised ontrol trials showing hormone replacement therapy (HRT) sig- ificantly increased the risk of heart disease. Yet a review of hat were considered the best quality observational studies (11
∗ Tel.: +61 267 726475. E-mail address: [email protected].
w T w ( s a t
001-4575/$ – see front matter © 2006 Elsevier Ltd. All rights reserved. oi:10.1016/j.aap.2006.06.007
ase-control studies, 16 prospective studies, 3 cross-sectional tudies) concluded that HRT decreased the risk by 50% (Lawlor t al., 2004a). The problem was attributed to the same sort of dif- erences between users and non-users of HRT being present in early all studies, leading to difficulties in correctly adjusting for onfounders. Similar misleading results were also reported for tudies of other self-selected populations, e.g. users of vitamin upplements (Lawlor et al., 2004b).
Evidence suggests that cyclists who choose to wear helmets ay differ substantially from those who do not. Helmet earers are more likely to ride in parks, playgrounds or bicycle aths than city streets (DiGuisseppi et al., 1989), obey traffic aws (Farris et al., 1997), wear fluorescent clothing and use ights at night (McGuire and Smith, 2000). These factors affect oth the risk of colliding with motor vehicles, and impact speed hen collisions occur. The former was evident in the data of hompson et al. (1996) in that non-helmeted cyclists collided ith motor vehicles 41% more frequently than helmet wearers
OR = 1.50, P < 0.0001). The latter was demonstrated by a tudy of bike/motor vehicle collisions. The authors (Spaite et l., 1991) concluded: “This implies that non-users of helmets end to be in higher impact crashes than helmet users, since the
sis an
i b
b a a o w o e w w
f 3 a s c o o c b ( r D e b b a a
i t h h e c T f
i r f b
3
i w I t e 2 i a r t b c a d
i i t m P 1 s a
i e p
F w f s
D.L. Robinson / Accident Analy
njuries suffered in body areas other than the head also tend to e much more severe”.
Bike/motor vehicle collisions caused a majority (34 out of 62) rain injuries >AIS2 in the case-control study of Thompson et l. (1996). The authors attempted to adjust for age and whether motor vehicle was involved (reporting no significant effect of ther factors), but did not consider impact speed in collisions ith motor vehicles, although this significantly affects the risk f head injury (Janssen and Wismans, 1985). Thus any differ- nces in head injuries due to differences in impact speed between earers and non-wearers, as observed by Spaite et al. (1991), ould be incorrectly attributed to helmets. There may also have been difficulties in correctly adjusting
or other confounders. Thompson et al. (1989) reported only age categories: <15, 15–24 and >25. However, a subsequent
nalysis of a subset of the same data (Thompson et al., 1990) howed that 83% of children aged 0–4 suffered head injury, ompared to 42% of 5–9 year olds and 23% of 10–14 year lds. Such large differences indicate that age adjustment in the riginal study may have been inadequate. Another indication of onfounding in Thompson et al. (1989) was the vast discrepancy etween helmet wearing of children in the control (CC) group 21.1%; n = 478) and an observational study (OS) of children iding round the same city in the same year (3.2%; n = 4501, iGuisseppi et al., 1989). The larger OS study was intended to
stimate population helmet wearing rates. If the sole difference etween CC and OS cyclists was that the former fell off their ikes, it would imply that helmet wearing was associated with seven-fold increase in the risk of falling off the bike, negating ny benefit of helmets.
Curnow (2005) argued that fear of death or chronic disabil- ty (which he defined as brain injuries of severity AIS4-6) was he main motive for wearing helmets. However, the majority of ead injuries treated in emergency departments (73% of the 757 ead injuries in the study of 3390 injured cyclists by Thompson
t al., 1996) did not involve brain injury. Brain injuries >AIS2 omprised only 8% of head injuries (Thompson et al., 1996). he Cochrane review (Thompson et al., 2003) calculated odds
or brain injury >AIS2 from at most 90 such injuries in two stud-
1 T c i
ig. 1. Comparison of two road safety interventions in Victoria, Australia. (a, left) Per ith numbers of non-head injuries to bicyclists (BNonHd) and pedestrians (PNonH
rom collisions with motor vehicles for bicyclists (B%DSHI) and pedestrians (P%DS peeding and drink-driving. Sources: ATSB (2002) and the Victorian Transport Accid
d Prevention 39 (2007) 86–93 87
es (4.2% and 1.8% of injuries in Thompson et al., 1989, 1996, espectively). The small numbers and potential problems of con- ounding noted above suggest that the conclusions concerning rain injury >AIS2 should be treated with caution.
. Helmet law studies
A published review examined data from enforced helmet laws n all jurisdictions where legislation increased percent helmet earing (%HW) by at least 40 percentage points within a year.
n contrast to the 90 brain injuries >AIS2 in the Cochrane review, he helmet-law review included 10,479 head injuries severe nough to appear in hospital admissions databases (Robinson, 006). In five jurisdictions with hospital admissions data, %HW ncreased from a pre-law average of 35% to a post-law aver- ge of 84%. If, as claimed by the Cochrane review, helmets educe serious head injuries by 63–88%, an increase from 35% o 84% helmet wearing would reduce percent head injury (%HI) y 39–62%. It would be impossible to miss such large, sudden hanges in time series data. Yet there was little or no notice- ble response in %HI to the changes in %HW, leading to serious oubts about the benefits of helmet legislation (Robinson, 2006).
Fig. 1a illustrates some of the factors involved, contrast- ng %HW in Victoria, Australia with (1) numbers of non-head njuries from bike/motor vehicle collisions to cyclists and pedes- rians and (2) percentages of serious injuries in collisions with
otor vehicles involving death or serious head injury (%DSHI). olice enforced the helmet law. Surveys in the first month (July 990) showed 94%, 87% and 89% of primary and secondary chool children and adults wore helmets, compared to 65%, 37% nd 44% in March 1990 (Sullivan, 1990).
The obvious and sharp decline in numbers of non-head njuries (Fig. 1a) coinciding exactly with legislation can be xplained by noting that numbers counted in identical pre- and ost-law observational surveys declined by 36% (Robinson,
996); non-head injuries declined because cycle-use declined. here was also a more gradual change in the ratio of adult to hild cyclists. Helmet laws discouraged children (42% reduction n the first year) more than adults (29% reduction), resulting in an
cent without helmets (%NonHel) before and after helmet legislation, compared d) and percentages of serious injuries involving death or serious head injury
HI). (b, right) Numbers of pedestrian fatalities and timing of campaigns against ent Commission.
8 sis an
i t i i t ( p b
c t b w s r t b H R o
4
s t G a r n o a t d p m
d a a o
d f w e
h o s t s
5
a t h a i h “ t
i d c l t c h i h
t a c o
T R p
C
2 4 6 8 9 9 9
O a r v o
8 D.L. Robinson / Accident Analy
ncreasing proportion of adults (50%, 55% and 60%) counted in he 1990, 1991 and 1992 surveys respectively. This was reflected n an increasing proportion of injuries to adults (36% pre law, ncreasing to 42%, 46% and 50% in 1991, 1992 and 1993 respec- ively). Adults had lower %DSHI (19%) than children <18 years 29%). The gradual decline in %DSHI of cyclists relative to edestrians is not consistent with the change in %HW, but might e explained by the increasing proportion of adult cyclists.
If legislation did not increase risk taking or discourage ycling, helmet law studies would compare identical popula- ions of cyclists before and after legislation, the only change eing increased helmet wearing. The absence of obvious benefit ould imply that helmets are largely ineffective. However, all
tudies of enforced helmet legislation that measured cycle-use eported substantial declines. Thus it is possible to conclude only hat helmet laws did not reduce the risk of injury per cyclist, ut not whether this is due to risk compensation (Adams and illman, 2001), reduced safety in numbers (Jacobsen, 2003; obinson, 2005), or over-optimistic predictions of the benefits f helmets in preventing serious head injuries.
. Experimental evidence
Experimental evidence shows that the brain is particularly usceptible to damage from rotations, which can shear connec- ions between neurones and even blood vessels. For example, ennarelli et al. (1972) subjected 12 squirrel monkeys to linear
ccelerations with peak levels 665–1230 g, and 13 primarily to otational accelerations in the range of 348–1025 g. Contact phe- omena were minimised by the design of the apparatus. None f the monkeys receiving linear acceleration was concussed, but ll 13 receiving rotational acceleration suffered concussion and he group had a high incidence of brain injuries such as sub- ural haematoma, subarachnoid haemorrhage and intracerebral etechial haemorrhage. Thus rotational accelerations can cause ore severe brain injuries than linear accelerations. Corner et al. (1987) measured rotational accelerations when
ummies wearing bicycle helmets went over the handlebars t 45 km/h and hit a smooth surface. Compared with a toler- nce of 1800 rad s−2 for concussion and 4500 rad s−2 for onset f vein rupture, measurements (averaging 58,000 rad s−2) were
a T w p
able 1 elationship between crash severity, represented by the percentage of non-helmet wea revented, for a model with an odds ratio of 0.31
rash severity (%HIN) Odds non-helmeted (ON) Odds helmeted (=0
0 0.25 0.08 0 0.67 0.21 0 1.50 0.47 0 4.00 1.24 0 9.00 2.79 5 19.00 5.89 9 99.00 30.69
dds are calculated as percent with head injury (%HI) divided by percent without HI, nd ON are odds for helmeted and non-helmeted cyclists respectively. An OR of 0.31 isk ratio (RR) = %HIH/%HIN and the risk reduction (RRD) or percent HI prevented = ehicle crashes, the overall percent head injury prevented will be much less than wha f head injuries”.
d Prevention 39 (2007) 86–93
escribed as “enormous”. No comparable results were reported or non-helmeted dummies, but other experiments showed that earing bike helmets increased rotational accelerations (Corner
t al., 1987). The lack of evidence for significant reductions in serious
ead injury following helmet legislation, small numbers of seri- us brain injuries and problems of confounding in case-control tudies, together with experiments such as the above, suggests hat the effect of helmets on rotational injuries requires further tudy.
. Odds ratios versus risk of head injury
Odds ratios (OR) for helmet efficacy are commonly described s reductions in risk. For example, Thompson et al. (1989) state hat “riders with helmets had an 85% reduction in their risk of ead injury (odds ratio 0.15;. . .)”. Sacks et al. (1991) cited this s evidence that, over a 5-year period, 2500 out of 2985 head njury deaths to US cyclists could be prevented if all cyclists wore elmets. The American Academy of Pediatrics (2001) stated that The bicycle helmet is a very effective device that can prevent he occurrence of up to 88% of serious brain injuries”.
The above predictions ignore the fact that the risk of head njury (and risk reduction calculated from odds ratios) is highly ependent on the circumstances of the crash. In bike-only rashes when the rear wheel skids, common impact sites are egs, hips, arms or shoulders. Lacking the extra size and mass of he helmet, bareheaded cyclists rarely hit their heads in minor rashes, let alone sustain serious head injuries. In contrast, a ead-on collision with a vehicle travelling at more than 80 km/h s likely to cause death or serious head injury, irrespective of elmet wearing.
Fitting odds ratios by logistic regression is equivalent to fit- ing a relationship between the risk of head injury for helmeted nd non-helmeted cyclists that depends on the severity of the ollision (see Table 1). So, for collisions so severe that 95% f non-helmeted cyclists would suffer serious brain injuries,
n OR = 0.31 (the Cochrane review estimate for head injury, hompson et al., 2003) means that 85.6% of helmeted cyclists ould be similarly afflicted, i.e. only 10% of injuries would be revented.
rers with head injury (%HIN), and risk reduction (RRD) or percent head injuries
.31 × ON) %HI helmet wearers Percent HI prevented (RRD)
7.2 64.0 17.1 57.2 31.7 47.1 55.4 30.8 73.6 18.2 85.5 10.0 96.8 2 2
i.e., O = %HI/(100 − %HI). The odds ratio, OR, is defined as OH/ON where OH
implies that OH = 0.31 × ON. %HIN can then be calculated from OH, to give the 100(1 − RR). Because most serious head injuries are from serious bike/motor
t people are led to believe when an OR of 0.31 is described as “preventing 69%
sis an
o m a m h o w
h w 1 t D h a T t
6
c i o t c
5 o N I h a h
a f % 3 t c l n
i d c n t b V d
n f t
5 c s f t p w t h
1 a h h t ( t
s a o o r r a i m m m a h
d t s t d i r t m a p a b
N s i C l m i
D.L. Robinson / Accident Analy
This explains why estimated reductions in head injuries based n data from emergency departments may not be appropriate for ore serious crashes. For example, McDermott et al. (1993),
nalysing hospital admissions data, reported that 29% of hel- eted adult cyclists (and 18% of helmeted child cyclists) had
ead injuries. If, as was claimed, helmets prevented 63–88% f all head injuries, expected injury rates in adult non-wearers ould be 78–241%, a far cry from the 38% actually observed. A detailed investigation of cyclists with severe or fatal
ead injuries (Corner et al., 1987) found that all fatalities ere caused by bike/motor vehicle collisions. For 13 of the 4 non-helmeted cyclists who died, there was no indication hat a helmet might have made any difference to the outcome. espite head protection, helmeted cyclists frequently suffer ead injuries. Nearly all deaths and debilitating head injuries re caused by bike/motor vehicle collisions (Kraus et al., 1987). he costs and benefits of helmets should therefore be compared
o those for other road safety measures.
. Cost-benefit analyses: helmet laws
Two groups of researchers used helmet law data to investigate osts and benefits. Neither considered the cost of the reduction n cycling because of helmet legislation, nor the consequences f reduced safety in numbers. Despite this, neither concluded hat the benefits of legislation were likely to have exceeded the osts.
In New Zealand, the estimated cost of helmets in the first years of the law was NZ$7.51 million, mainly for purchase
f helmets, 27 times greater than the estimated reduction of Z$0.28 million in hospital costs (Taylor and Scuffham, 2002).
t was also noted that there was no significant effect of increased elmet wearing in models that fitted a time trend (Scuffham et l., 2000), so even this modest reduction in hospital costs may ave been over-estimated.
For Western Australia, initial modelling was unable to detect ny change in %HI of cyclists compared to other road users. Dif- erent models were tried; some indicated possible reductions in
HI compared to pedestrians, the benefits ranging from about 0%–109% of the cost of the helmet law (A$21.6 million over he 7-years 1992–1998). The authors (Hendrie et al., 1999) con- luded: “In monetary terms, it is unlikely that the helmet wearing egislation would have achieved net savings of any sizeable mag- itude”.
Even though cycling to work was estimated to reduce mortal- ty by 40% (Andersen et al., 2000), the two cost-benefit analyses iscussed above ignored the loss of health benefits from reduced ycling. Some effects are hard to value, such as the inconve- ience or discomfort of wearing a helmet, or pain from a wound o the head. The fact that some cyclists choose to wear helmets, ut others do not, plus the 36% decline in counts of cyclists in ictoria with helmet laws, suggests that these intangibles have ifferent values to different cyclists.
For a more intuitive understanding of risks vs intangibles, ote that the annual total of 8502 fatalities or hospital admissions or non-cycling road injuries in Victoria in 1989 (the year before he helmet law) equates to one death or hospital admission every
(
r (
d Prevention 39 (2007) 86–93 89
17 person-years. At that time, Victoria had about 2 million yclists (Cameron et al., 1994) and 430 cyclist hospital admis- ions with some form of head injury (not necessarily the reason or admission). Even if universal helmet wearing prevented half hese head injuries, it would represent only one head injury er 9302 cyclist-years. In reality, even these modest benefits ere not achieved (Fig. 1a), perhaps because of risk compensa-
ion, reduced safety in numbers, poor fitting or incorrectly worn elmets.
Curnow (2005) compared 1988 (before any helmet law) with 994 (when all states had enforced laws); cyclist, pedestrian nd all road user deaths fell by 35%, 36% and 38% respectively; ead-injury deaths fell by 30%, 38% and 42%. Thus, despite elmet legislation, the reductions for cyclists were no greater han for other road users. Factoring in the reduction in cycling Fig. 1), cyclists may even have been worse off with helmet laws han without them.
Cost-benefit analyses are a common feature of many road afety decisions, e.g. whether to treat accident blackspots. Usu- lly, predicted savings in injury costs (amortised over a period f several years) must be several times greater than the cost f implementation. Some measures are reasonably effective. A eview of area-wide traffic calming schemes estimated that they educe the number of injury accidents on residential streets by an verage of 25% and 10% on main roads (Elvik, 2001). However, ndividual schemes vary, so continuous evaluation and assess-
ent is needed to ensure resources are used in the most effective anner and predicted benefits are realised. Some traffic calming easures have also been shown to generate other benefits such
s increased pedestrian activity leading to increased physical ealth (Morrison et al., 2004).
Both cyclists and pedestrians should benefit from measures irected at improving overall road safety. A 42% fall in pedes- rian fatalities in Victoria was attributed to campaigns against peeding and drink-driving that coincided almost exactly with he bicycle helmet law (Fig. 1b, Powles and Gifford, 1993). Road eaths also fell substantially in other Australian states (includ- ng New South Wales and South Australia) that adopted similar oad safety campaigns. One estimate, from UK literature, was hat accident costs in Victoria were reduced by an estimated £100
illion for an outlay of £2.3 million (A$5 million, see Powles nd Gifford, 1993). This compares with at least $39 million for urchase of cycle helmets to satisfy Victoria’s helmet law. The bove suggests that measures to improve overall road safety can e more cost effective than helmet legislation.
Recent research highlighted the importance of ‘Safety in umbers’. Injury and fatality rates per kilometre cycled are
ubstantially higher – more than five times for higher for fatal- ties – in countries where fewer people cycle (Jacobsen, 2003). ountries with the low helmet wearing have more cyclists and
ower fatality rates per cycle-km. (BHRF, 2006). Thus hel- et legislation may be counter-productive and actually increase
njuries per cyclist if, as in Australia, cycling is discouraged
Robinson, 2005).
It has been argued that cyclists would need to increase their isk-taking four-fold to overcome the protection of helmets Thompson et al., 2003). Table 1 demonstrates the fallacy of
9 sis an
t o o w
( i b t h
v o a e c y h t e w s a
7
s
(
(
(
(
(
(
(
r m M h D r c
% l s w t S ( ( s helmets were an important factor, the trend should have reversed when helmet wearing returned to pre-law levels. More realistic estimates might have been obtained if the initial research had considered point 3.
0 D.L. Robinson / Accident Analy
his argument. For crashes severe enough to cause injury to 95% f non-wearers, even if helmet wearing increased risk-taking by nly 10%, there would be the same number of head injuries as ithout helmets, but 10% more non-head injuries. Less-serious crashes have a lower percentage of head injuries
e.g. 11% of non-wearers, Maimaris et al., 1994). A 10% ncrease in crashes per km cycled (due to reduced safety in num- ers or risk compensation) would result in the same or more (in he case of reduced safety in numbers) injuries per km, even if elmets prevented 80% of head injuries.
Costs and benefits have been calculated for helmets for motor ehicle users (already required by law to wear seatbelts). Based n real-life crash-data suggesting they could prevent 28%, 40% nd 26% of minor, moderate and severe brain injuries, McLean t al. (1997) estimated that helmets for Australian motor vehi- le occupants would reduce injury costs by $1.9 billion (over 5 ears, all vehicles equipped with airbags) or $2.2 billion (only alf the fleet with airbags). This compares with $0.78 billion o equip the entire population with helmets (20 million at $39 ach, the cost used by Hendrie et al., 1999). Yet helmets are not idely accepted by motorists (except racing car drivers), pre-
umably because intangible costs (e.g. comfort or convenience) re deemed to outweigh the benefits.
. General principles
The following general principles should have widespread upport:
1) Any legislation (including helmet laws) should not be enacted unless the benefits can be shown to exceed the costs. Ideally, the benefits should be greater than from equivalent ways of spending similar amounts of money on other road safety initiatives.
2) Helmet legislation should be evaluated in terms of the effect on cycle-use, injury rates per km cycled, and changes in percentages of hospitalised cyclists with head and brain injuries (%HI). Because public health and loss of liberty are involved, it should be the responsibility of governments who pass such legislation to ensure adequate funds are set aside for evaluation.
3) Trends are a common feature of %HI data for all road users, not just cyclists. %HI data for cyclists should therefore be compared with the same statistics for other road users. Models should also be able to differentiate between gradual changes over time that do not correspond with the changes in helmet wearing and consequences of increased %HW.
4) Odds ratios should not be described as “percentage reduc- tions in head injury”. Instead, the fitted models should be used to predict and report head injury rates and risk ratios for crashes of different severities, e.g. bike only crashes requiring emergency treatment, bike/motor vehicle
crashes requiring admission to hospital for short or long periods. Given the discrepancy in %HW of cyclists report- ing falls from bikes and observational surveys of people cycling, case-control studies should also survey and report data on observed %HW of the cycling population.
F C ( M
d Prevention 39 (2007) 86–93
5) Surveys of cycle-use should use the same sites and obser- vation periods and be conducted at the same time of year, ideally the year before and year after legislation, to avoid confounding with long-term trends. Labour costs could be reduced by use of automatic counters on cycleways and road camera footage.
6) Risk of injury for different groups (e.g., helmet wearers ver- sus non-wearers) should be evaluated by comparing num- bers of injuries with estimated cycle-use. Changes in risks per km cycled should be compared with similar data for other road users.
7) If the benefits of helmet laws cannot be shown to exceed the costs by similar ratios to other road safety initiatives, the legislation should be repealed.
These principles could provide a framework for future esearch to evaluate and compare results, and so facilitate agree- ent. For example, excluding Australia and New Zealand, a edline search revealed 23 papers reporting consequences of
elmet laws; 65% considered only the effect on helmet wearing. espite the fact helmet laws are intended to reduce head injury
ates, only two reported significant declines in %HI, but neither ompared head injury and helmet wearing rates.
One (Macpherson et al., 2002) reported declining trends in HI for child cyclists in Canadian provinces that passed helmet
aws, compared with those that did not. However, the trends tarted before helmet legislation and continued after helmet earing should have stabilised. It therefore seemed unlikely that
he trends were a consequence of legislation (Robinson, 2003). ubsequent data confirm this. Ontario’s law was not enforced Burdett, 2002) and %HW returned to pre-law levels by 1999 Macpherson, 2006). The dominant feature (Fig. 2) is a relatively mooth downward trend, unrelated to published %HW rates. If
ig. 2. Percent helmet wearing (%HW) from surveys of children in Ontario, anada and percentage of child cyclist hospital admissions with head injuries
%HI). Sources: Macpherson et al., 2002; CIHI, 2003; Parkin et al., 2003; acpherson, 2006.
sis an
i a c b t s t e g r
a 2 o o 1 b % t c b s t w
c i d ( ( t ( c i m d f i
m t l n ( s r m w h b e m h s
s
d a ( b c 2
P M 1 s t a ( u ( e t H u m (
t c s s
8
s r r i t b l h h c p
w t t l g m c r o f
D.L. Robinson / Accident Analy
Similar difficulties were noted for a study of voluntary wear- ng (Cook and Sheikh, 2003). A subsequent analysis of the same nd later data found identical trends in %HI for boy and girl yclists, despite increasing %HW for girls and falling %HW for oys (Hewson, 2005). Although trends for cyclists and pedes- rians were not identical, the author considered the evidence trongly suggestive that the difference could not be due simply o helmet wearing. Greater recognition of point 3 might have ncouraged Cook and Sheikh (2003) to make use of the diver- ent %HW trends in boys and girls to test whether helmets were esponsible, instead of assuming this was the case.
The only other paper (out of the 23 noted above) reporting significant reduction in %HI following legislation (Lee et al., 005) did not even consider trends and provided no information n %HI by year. Instead, average %HI for various categories f cyclists for 1991–1993 were compared with averages for 994–2000. There was a reduction for children in California ut not adults. However, there is no evidence of a divergence in HW of adult and child cyclists. %HW of adults attending a
rauma centre in California increased by a similar number of per- entage points as %HW of children (Ji et al., 2006), suggesting oth may have followed similar trends. Point 2, that legislation hould be evaluated, would encourage governments to collect he required data, making it unnecessary to speculate on helmet earing and how this might have affected %HI rates. To be convincing, there should be a clear response to the
hange in %HW (point 3), as apparent for non-head injuries n Victoria (Fig. 1a), presumably because enforced legislation iscourages cycling. Hagel and Pless (2006) criticised Curnow 2005) for not citing Macpherson et al. (2003), Cook and Sheikh 2003) or Scuffham et al. (2000). However, as noted above, he first two failed to correctly account for trends. The third Scuffham et al., 2000) reported that the addition of a time-trend aused the helmet wearing proportion to become insignificant, mplying that the reduction in %HI might be due either to hel-
ets or time trends. The continued declines in %HI in Ontario, espite a return to pre-law %HW, illustrates the potential con- usion created by not correctly accounting for trends, or uncrit- cally citing such work.
Point 4 advocates reporting risk ratios, not odds ratios. Hel- ets and helmet laws continue to be promoted by claims in
he popular press that “Cyclists who wear helmets are 85% ess at risk of serious head injury” (McKee, 2006). However, on-helmeted cyclists rarely have injuries only to the head McDermott et al., 1993), so helmeted cyclists who crash will till need medical treatment for non-head injuries. Consequently, isk reductions (RRD, see Table 1) provide a more accurate esti- ate of the benefit of helmets. The crude RRD for voluntary earing (25% for head injuries of adult cyclists admitted to ospital, McDermott et al., 1993; 56% for concussion or other rain injury of cyclists in non-motor vehicle crashes treated in mergency rooms, Thompson et al., 1996) convey a different essage to the well-known 85% claim. Realised benefits from
elmet laws will be even less, because of poor fit, risk compen- ation and reduced safety in numbers.
Point 5 above notes the important requirement that surveys hould use the same sites and observation periods and be con-
w c c a
d Prevention 39 (2007) 86–93 91
ucted at the same time of year, ideally the year before and fter helmet laws. Hagel and Pless (2006) cite a detailed report Finch et al., 1993) listing pre- and post-law counts in Mel- ourne from identical surveys in May 1990 and 1991. Teenage ycle-use fell by 44% and numbers of adults counted fell by 9%.
Rather than report the 29% fall in adults counted, Hagel and less (2006) draw conclusions from a tenuous comparison of the ay 1991 survey with a much earlier survey (December/January
987/1988). Other data, including cycling injuries in Victoria, how a marked annual cycle (Carr et al., 1995) so it is not possible o reliably estimate changes in cycle-use by comparing surveys t different times of year. For teenagers, the same comparison 1987/1988 versus 1991) yields an estimated 8% fall in cycle- se, nothing like the accepted value of 44%. Hagel and Pless 2006) criticised Curnow (2005) for failing to present all relevant vidence, then omit data from the 1990 survey (counts of adults) hat refute their argument. Instead of easy to understand counts, agel and Pless (2006) cite only estimated cycle use, with an nbelievable value of 60 million hours per week in a city of 3 illion – 20 hours per week for every man, woman and child
Cameron et al., 1994)! General acceptance of points 1–7 could provide a framework
o help reduce confusion. Non-compliance (e.g. claims based on ounts of cyclists at different times of year, or long-term data eries with no allowance for trends) would be noted as such and o identified as less reliable data.
. Conclusions
Neither helmet law evaluations nor case-control studies hould be considered in isolation. To view helmets in the cor- ect perspective, we need to understand and explain both sets of esults, and experimental data. The strongest helmet law stud- es are those with large increases in %HW in a short period of ime. They show trends in %HI for hospital admissions data, ut no obvious response to the large increases in %HW with egislation. The lack of response might be due to an inability of elmets to prevent the more serious head injuries associated with ospital admissions, reduced safety in numbers due to reduced ycle use, risk compensation, or other changes in the cycling opulation.
In the largest case-control study, 92% of head injuries from ere of severity AIS1 or AIS2, suggesting they involved wounds
o the head or concussions. Helmets presumably prevent wounds o the head. Although case-control data show wearers also have ower rates of concussions and other brain injuries, evidence sug- ests that helmet wearers were less likely to have collided with otor vehicles and tend to be in lower impact bike/motor vehicle
rashes than non-wearers. As demonstrated by the contradictory esults from randomized control trials and observational studies f HRT, it can be very difficult to disentangle effects of con- ounded variables. The obvious confounding of helmet wearing
ith socioeconomic status, attitude to risk and other factors asso-
iated with less frequent and lower impact bike/motor vehicle ollisions leads to considerable difficulties in the interpretation nd understanding of case-control data for helmet wearing.
9 sis an
v l t t h t w
o s 1 c r i a c b b
r d r a r
R
A
2 A
A
B
B
C
C
C
C
C
C
D
E
F
F
G
H
H
H
J
J
J
K
L
L
L
M
M
M
M
M
M
M
M
2 D.L. Robinson / Accident Analy
A majority of brain injuries >AIS2 are caused by bike/motor ehicle collisions. Traffic calming, enforcement of drink-driving aws, cyclist and driver education, or other measures to reduce he frequency and severity of bike/motor vehicle collisions, may herefore represent more cost-effective ways of reducing serious ead injuries to cyclists than helmet laws. Indeed, countries with he lowest fatality rates per cycle-km also have the lowest helmet earing rates. Information from all sources is required to reach a consensus
n whether helmet legislation is beneficial—case-control data, tudies of helmet laws and experimental data (Corner et al., 987; Gennarelli et al., 1972; Janssen and Wismans, 1985). All osts and benefits need to be considered, including the cost of educed cycling, reduced enjoyment of cycling, reduced safety n numbers, and the ability of helmets to protect against minor nd serious brain injuries. These need to be compared to the ost of other road safety initiatives to reduce the incidence of ike/motor vehicle collisions that cause a majority of >AIS2 rain injuries.
Consensus can be reached if future research considers all elevant aspects of the problem. Adherence to the principles escribed above would allow relevant concerns to be addressed, esulting in the most cost-effective way of increasing the safety nd popularity of cycling and maximising both health and envi- onmental benefits.
eferences
dams, J., Hillman, M., 2001. The risk compensation theory and bicycle hel- mets. Inj. Prevent. 7 (2), 89–91.
001. Pediatrics 108 (4), 1030–1032. ndersen, L.B., Schnohr, P., Schroll, M., Hein, H.O., 2000. All-cause mortal-
ity associated with physical activity during leisure time, work, sports, and cycling to work. Arch. Intern. Med. 160 (11), 1621–1628.
TSB, 2002. Road Fatalities Australia 2001 Statistical Summary, Australian Transport Safety Bureau.
HRF, 2006. Bicycle Helmet Research Foundation Safety in Numbers Graph, http://www.cvclehelmets.ore/ (accessed June 2006).
urdett, A., 2002. Butting heads over bicycle helmets. eCMAJ http://www.cmai. ca/cgi/eletters/167/4/338#150 (accessed March 2006).
ameron, M.H., Vulcan, A.P., Finch, C.F., Newstead, S.V., 1994. Mandatory bicycle helmet use following a decade of helmet promotion in Victoria, Australia—an evaluation. Accid. Anal. Prevent. 26 (3), 325–337.
arr, D., Skalova, M., Cameron, M., 1995. Evaluation of the bicycle helmet law in Victoria during its first 4 years, Rpt 76 Monash Univ Ace Res Centre Melbourne.
IHI, 2003. Injury Hospitalizations (includes 2000-01 and 2001-02 data), Cana- dian Institute for Health Information.
ook, A., Sheikh, A., 2003. Trends in serious head injuries among English cyclists and pedestrians. Inj. Prevent. 9, 266–267.
orner, J.P., Whitney, C.W., O’Rourke, N. and Morgan, D.E., 1987. Motorcycle and bicycle protective helmets: requirements resulting from a post crash study and experimental research. Canberra, Federal Office of Road Safety, Report CR 55.
urnow, W.J., 2005. The Cochrane collaboration and bicycle helmets. Ace. Anal. Prevent. 37 (3), 569–573.
iGuisseppi, C.G., Rivara, F.P., Koepsell, T.D., 1989. Bicycle helmet use by children. Evaluation of a community-wide helmet campaign. JAMA 262,
2256–2261.
lvik, R., 2001. Area-wide urban traffic calming schemes: a meta-analysis of safety effects. Accid. Anal. Prevent. 33 (3), 327–336.
arris, C., Spaite, D.W., Criss, E.A., Valenzuela, T.D., Meislin, H.W., 1997. Observational evaluation of compliance with traffic regulations
P
P
d Prevention 39 (2007) 86–93
among helmeted and non-helmeted bicyclists. Ann. Emerg. Med. 29 (5), 625–629.
inch, C, Heiman, L., Neiger, D., 1993. Bicycle use and helmet wearing rates in Melbourne, 1987–1992: the influence of the helmet wearing law. Melbourne, Rpt 45, Monash Univ Ace Res Centre.
ennarelli, T.A., Thibault, L.E., Ommaya, A.K., 1972. Pathophysiological Responses to Rotational and Translational Accelerations of the Head, SAE Paper No. 720970. 16th Stapp Car Crash Conf., Society of Automotive Engi- neers, 1972.
agel, B.E., Pless, B., 2006. A critical examination of arguments against bicycle helmet use and legislation. Accid. Anal. Prevent. 38 (2), 277–278.
endrie, D., Legge, M., Rosman, D., Kirov, C, 1999. An economic evalu- ation of the mandatory bicycle helmet legislation in Western Australia, http://www.officeofroadsafetv.wa.gov.au/Facts/papers/bicvcle helmet legis- lation.html (accessed October 2004).
ewson, P.J., 2005. Investigating population level trends in head injuries amongst child cyclists in the UK. Accid. Anal. Prevent. 37 (5), 807–815.
acobsen, P.L., 2003. Safety in numbers: more walkers and bicyclists, safer walking and bicycling. Inj. Prevent. 9 (3), 205–209.
anssen, E.G., Wismans, J.S.H.M., 1985. Experimental and mathematical sim- ulation of pedestrian–vehicle and cyclist–vehicle accidents, Proceedings of the 10th International Technical Conference on Experimental Safety Vehi- cles, Oxford.
i, M., Gilchick, R.A., Bender, S.J., 2006. Trends in helmet use and head injuries in San Diego County: the effect of bicycle helmet legislation. Accid. Anal. Prevent. 38 (1), 128–134.
raus, J.F., Fife, D., Conroy, C., 1987. Incidence, severity, and outcomes of brain injuries involving bicycles. Am. J. Public Health 77 (1), 76–78.
awlor, D.A., Davey Smith, G., Ebrahim, S., 2004a. Commentary: The hor- mone replacement-coronary heart disease conundrum: is this the death of observational epidemiology? Int. J. Epidemiol. 33 (3), 464–467.
awlor, D.A., Davey Smith, G., Kundu, D., Bruckdorfer, K.R., Ebrahim, S., 2004b. Those confounded vitamins: what can we learn from the differences between observational versus randomised trial evidence? Lancet 363 (9422), 1724–1727.
ee, B.H.Y., Schofer, J.L., Koppelman, F.S., 2005. Bicycle safety helmet legisla- tion and bicycle-related non-fatal injuries in california. Accid. Anal. Prevent. 37 (1), 93–102.
acpherson, A.K., 2006. An evaluation of the effectiveness of bicycle hel- met legislation (powerpoint presentation, available at http://www.circl. pitt.edu/home/webinars/ppt/macphersonwebinar.ppt) (accessed June 2006).
acpherson, A.K., To, T.M., Macarthur, C., Chipman, M.L., Wright, J.G., Parkin, P.C., 2002. Impact of mandatory helmet legislation on bicycle- related head injuries in children: a population-based study. Pediatrics 110 (5), e60.
aimaris, C., Summers, C.L., Browning, C., Palmer, C.R., 1994. Injury patterns in cyclists attending an accident and emergency department: a comparison of helmet wearers and non-wearers. BMJ 308, 1537–1540.
cDermott, F.T., Lane, J.C., Brazenor, G.A., Debney, E.A., 1993. The effec- tiveness of bicyclist helmets: a study of 1710 casualties. J. Trauma 34 (6), 834–844.
cGuire, L., Smith, N., 2000. Cycling safety: injury prevention in Oxford cyclists. Inj. Prevent. 6 (4), 285–287.
cKee, I., 2006. Using your head can help to avoid any serious injury, The Scotsman http://living.scotsman.com/health.cfm?id=376662006 (accessed March 06).
cLean, A., Fildes, B., Kloeden, C, Digges, K., Anderson, R., Moore, V. and Simpson, D., 1997. Prevention of Head Injuries to Car Occupants An Inves- tigation of Interior Padding Options, Federal Office of Road Safety: Rpt CR160. Available at http://www.monash.edu.au/muarc/reports/atsb 160.pdf.
orrison, D.S., Thomson, H., Petticrew, M., 2004. Evaluation of the health effects of a neighbourhood traffic calming scheme. J. Epidemiol. Community Health 58 (10), 837–840.
arkin, P.C., Khambalia, A., Kmet, L., Macarthur, C., 2003. Influence of socioe- conomic status on the effectiveness of bicycle helmet legislation for children: a prospective observational study. Pediatrics 112 (3), e192–e196.
owles, J.W., Gifford, S., 1993. Health of nations: lessons from Victoria, Aus- tralia. BMJ 306, 125–127.
sis an
R
R
R
R
S
S
S
S
T
T
T
T
D.L. Robinson / Accident Analy
obinson, D.L., 1996. Head injuries and bicycle helmet laws. Accid. Anal. Prevent. 28, 463–475.
obinson, D.L., 2003. Confusing trends with the effect of helmet laws. Pediatrics P3Rs, http://pediatrics.aappublications.Org/cgi/eletters/110/5/e60ff429 (accessed March 2006).
obinson, D.L., 2005. Safety in Numbers in Australia: more walkers and bicy- clists, safer walking and bicycling. Health Promot. J. Aust. 16 (1), 47– 51.
obinson, D.L., 2006. No clear evidence from countries that have enforced the wearing of helmets. BMJ 332, 722–725.
acks, J.J., Holmgreen, P., Smith, S.M., Sosin, D.M., 1991. Bicycle-associated head injuries and deaths in the United States from 1984 through 1988. How many are preventable? J. Amer. Med. Assoc. 226, 3016– 3018.
cuffham, P., Alsop, J., Cryer, C., Langley, J.D., 2000. Head injuries to bicy- clists and the New Zealand bicycle helmet law. Accid. Anal. Prevent. 32, 565–573.
paite, D.W., Murphy, M., Criss, E.A., Valenzuela, T.D., Meislin, H.W., 1991. A prospective analysis of injury severity among helmeted and non-
T
d Prevention 39 (2007) 86–93 93
helmeted bicyclists involved in collisions with motor vehicles. J. Trauma 31, 1510–1516.
ullivan, G., 1990. Initial effects of mandatory bicycle helmet wearing legis- lation, Report IR 90-15 July 1990 Vicroads information services Burwood Road Hawthorn, Vic 3122.
aylor, M., Scuffham, P., 2002. New Zealand bicycle helmet law-do the costs outweigh the benefits? Injury Prevent. 8, 317–320.
hompson, D., Rivara, F., Thompson, R., 2003. Helmets for preventing head and facial injuries in bicyclists (Cochrane Review), In: The Cochrane Library, Issue 3. Oxford: Update Software.
hompson, D.C., Rivara, F.P., Thompson, R.S., 1996. Effectiveness of bicycle safety helmets in preventing head injuries A case-control study. JAMA 276 (24), 1968–1973.
hompson, D.C., Thompson, R.S., Rivara, F.P., 1990. Incidence of bicycle-
related injuries in a defined population. Am. J. Public Health 80 (11), 1388–1390.
hompson, R.S., Rivara, F.P., Thompson, D.C., 1989. A case-control study of the effectiveness of bicycle safety helmets. N. Engl. J. Med. 320 (21), 1361–1367.
- Bicycle helmet legislation: Can we reach a consensus?
- Introduction
- Case-control studies
- Helmet law studies
- Experimental evidence
- Odds ratios versus risk of head injury
- Cost-benefit analyses: helmet laws
- General principles
- Conclusions
- References
memo3/Sposato.pdf
Transportation Research Part F 15 (2012) 581–587
Contents lists available at SciVerse ScienceDirect
Transportation Research Part F
journal homepage: www.elsevier .com/locate / t r f
The influence of control and related variables on commuting stress
Robert G. Sposato, Kathrin Röderer, Renate Cervinka ⇑ Medical University of Vienna, Centre for Public Health, Vienna, Austria
a r t i c l e i n f o
Article history: Received 29 May 2011 Received in revised form 16 April 2012 Accepted 8 May 2012
1369-8478/$ - see front matter � 2012 Elsevier Ltd http://dx.doi.org/10.1016/j.trf.2012.05.003
⇑ Corresponding author. Address: Medical Univers 14016034923; fax: +43 140160934903.
E-mail address: [email protected]
a b s t r a c t
Existing research on commuting stress has shown that it is affected by variables such as control, predictability, the duration of the commute and impedance. The present study investigates the impact of several factors on the stress that commuters experience and clarifies possible relations between variables. For the purpose of this study, an online ques- tionnaire was completed by 363 commuters of the Vienna region. The relative strength of the relationship between predictors and commuting stress was determined by multiple regression analysis. Results suggest that control is the most powerful predictor of commut- ing stress, followed by the duration of the commute, predictability and impedance. Control significantly interacts with the duration of the commute and predictability. Based on these findings a research model is proposed, clearly depicting the moderating impact of the dura- tion of the commute and incorporating a clear distinction between predictability and control.
� 2012 Elsevier Ltd. All rights reserved.
1. Introduction
Commuting to work has more often than not become part of an employee’s daily routine. In spite of this, up until the last decade relatively little effort has been devoted to studying the impact of these daily journeys on the commuter’s health (Klu- ger, 1998; Koslowsky, Kluger, & Reich, 1995). Recently, a considerable body of research has emerged, in particular dealing with factors causing workers to experience stress on their commute (Evans & Wener, 2006, 2007; Evans, Wener, & Phillips, 2002; Wener & Evans, 2011).
One of the best-researched phenomena in the environmental stress literature is control and its relationship with stress (Bollini, Walker, Hamann, & Kestler, 2004; Evans & Cohen, 1987; Troup & Dewe, 2002). Studies have investigated the effect of this variable in the commuting environment and proposed several underlying working mechanisms. There are two main conceptualizations of control in literature on commuting stress. One understands control as a variable moderating the stress- ful effects of the journey to work (Schaeffer, Street, Singer, & Baum, 1988), whereas another conceptualises control as a medi- ator (Evans & Carrere, 1991). The most recent conceptual work in this particular field on control was conducted by Kluger (1998). He found his operationalisation of control, commute variability, to be the strongest correlate of commute strain. This result is further corroborated by Wener and Evans (2011), who found their measure of control to be most highly correlated with commuting stress. Important for the present understanding of control in commuting stress research is that Kluger (1998) confirmed an additive effect of (lack of) control and raised the hypothesis that it mainly causes psychological strain. By proposing a direct relationship between control and stress he follows the lead of Evans and Carrere (1991) and breaks with the concept of control as a variable only moderating the effect of commute characteristics on stress. This development can be seen as a major improvement to the theoretical framework concerning control and its effects, but is still inconsistent
. All rights reserved.
ity of Vienna, Institute of Environmental Health, Kinderspitalgasse 15, 1090 Vienna, Austria. Tel.: +43
.at (R. Cervinka).
582 R.G. Sposato et al. / Transportation Research Part F 15 (2012) 581–587
in several respects. A weak point is that what Kluger (1998) claims to be an additive effect of control is in his theory again mediated by additional variables. He argues that low variability allows for aversive events to be predictable and predictabil- ity then serves as a form of perceived control. This reasoning somehow contradicts his view of commute variability as a new operationalisation of control and leaves the two constructs, control and predictability, in a theoretical proximity that is nei- ther explained, nor researched.
Partly motivated by Kluger’s (1998) findings, predictability was investigated by Evans et al. (2002). Their understanding of predictability is that it serves as a form of cognitive control when behavioural control can not be exercised. They reported elevated physiological stress for commuters who perceived their commute to be less predictable. A natural- and quasi exper- iment by Wener, Evans, Phillips, and Nadler (2003) reconfirmed the correlation between unpredictability and stress. In their study on the difference in stress experienced by car- and train commuters Wener and Evans (2011) were able to show that predictability partly accounts for elevated stress levels of car commuters. Control instead, did not yield any significant effect as a mediating variable. Unfortunately this study did not investigate possible interactive effects between control and predictability.
In the course of the above-mentioned study Wener et al. (2003) also found that physiological stress increased with the duration of the commute, which constitutes another emerging variable in the commuting stress research field. While early studies reported incoherent findings on the effects of commuting duration (Novaco, Stokols, Campbell, & Stokols, 1979; Stokols, Novaco, Stokols, & Campbell, 1978), Evans and Wener (2006) confirmed their earlier findings proving the negative effect of the duration of the commute on physiological and self-reported measures of stress. They criticised the tendency in research to focus on the intensity of stressors while neglecting the impact of duration of exposure.
The duration of the commute also plays a significant role in the concept of commute impedance by Stokols et al. (1978). Commute impedance1 is best illustrated as traffic congestion (Novaco et al., 1979; Stokols et al., 1978) and occurs when people are physically hindered in moving between two or more points. It is defined by the distance travelled and the time spent in transit. The researchers found significant between-group effects (high-, medium- and low-impedance) in subjective reports of traffic congestion and annoyance, mood and residential adaptation, but no significant effect for physiological stress measures (Novaco et al., 1979; Stokols et al., 1978). In a later study Novaco, Stokols, and Milanesi (1990) found impedance to be related to job satisfaction, illness, absences from work and overall incidence of colds or flu. As depicted in the last two paragraphs, the duration of the commute and impedance are two closely related variables but differ in an important aspect. While impedance can be understood as a commute characteristic, duration can only be a moderating variable. Based on our understanding of the literature we propose that the effect of duration of the commute on stress measures should be seen as a moderating effect of time potentiating the impact of certain stressors.
Another variable considered in the commuting stress literature is the number of commute stages (changes between modes of transportation and public transport lines). According to Taylor and Pocock’s (1972) results the number of commute stages is a better predictor of absenteeism than the time travelled. On the other hand Wener et al. (2003) clearly showed that any effect of the number of commutes stages on commuting stress is mediated by differences in the duration of the commute.
Summarising the argument, commuting stress studies have only sparsely addressed more than one commute character- istic and their effect on commuting stress at a time. This has led to a situation where several variables stand alone without clarification of their relative power in influencing commuting stress and with a disregard to possible intercorrelations. This can also be seen as a direct consequence of considerably little advancement on the theoretical framework in this particular research field since Kluger (1998). While the singular studies are rather unaffected in their quality by this increasingly clus- tered research field, it makes a comprehensive understanding relatively difficult and decreases accessibility for new research efforts.
The present study addresses these issues in several respects. First, we investigate the relative impact of a wide range of previously studied variables on commuting stress: control, predictability, impedance, duration of the commute, commute stages and two new variables. We argue that beyond the duration of the commute two other time variables – commute days per week and years of commuting – should be considered in order to capture a possible effect of long-term exposure to commut- ing. This is a proposition supported by early research from Stokols et al. (1978) who found the number of months on route to be significantly correlated with systolic and diastolic blood pressure. We expect to find a negative effect of commuting days per week commuting and years of commuting, just as any effect of an environmental stressor grows stronger with prolonged exposure. Second, we examine the nature of the relationship between two pairs of closely related variables, namely control and predictability; and duration of the commute and impedance. Third, we study if and to what extent interactive effects exist for control, predictability and duration of the commute.
With regard to the fragmented research field we try to improve the existing theoretical framework by proposing a new comprehensive commuting stress model. This shall serve as a guiding point for future research and create a more straight- forward access to the argument.
1 This definition, the successively mentioned results and the term impedance subsequently refer to the concept of objective/physical impedance and do not take into account the concept of subjective/perceived impedance.
R.G. Sposato et al. / Transportation Research Part F 15 (2012) 581–587 583
2. Method
2.1. Participants
366 commuters (53.7% women) with a mean age of 37.42 years (SD = 10.07) were recruited via mailing list to participate in an online survey. The majority of respondents were regular employees (91.3%) and small numbers of self- employed work- ers/professionals (8.7%). The bulk of participants was well educated (66.7% graduate workers, 24.6% college degree) and had a median monthly income exceeding 1500 €. Of the commuters 76.9% travelled to work five days a week, with an average of 4.71 commute days per week (SD = 0.618). Furthermore participants reported to have been commuting for a mean of 7.03 years (SD = 8.08). The average commuting time was 37 min one way (SD = 22.70), covering a mean distance of 19.53 km (SD = 24.02). Official data from the Austrian Ministry for Transport, Innovation and Technology (Herry Consult GmbH, 2007) suggest a mean commuting time of 23 min and Statistik Austria (2004) report that 41% of their respondents in the Vienna region commute 16–30 min, indicating that our sample has slightly elevated commuting times when com- pared to an average commuter in this region.
33.9% of the respondents stated they habitually commuted by bicycle or by foot, 41% used public transport and 25.1% re- lied on a car for their daily journey to work. 91.2% of commuters reported their workplace was within a 130 km radius of Vienna, with 87.9% of them commuting to the city of Vienna.
2.2. Procedure
Starting in March 2010 an email with general information on the study, an invitation to participate, and the appeal to forward the email was sent out via the employees’ mailing list of the Medical University of Vienna and to advanced students attending a class on Environmental Psychology with the University of Vienna, thus producing a convenience sample. The on- line questionnaire was then available until July 2010 when data collection was complete.
The survey consisted of three sections and took about 35 min to complete. In order to avoid missing data due to time lim- itations, respondents were given the possibility to save their progress and continue at a later point. Furthermore, participants could comment on the survey, or issues they thought were important, by sending an email to the research-team via an embedded link at the end of each survey.
2.3. Measures
Basic sociodemographic data on gender, age, income and occupation was gathered. Furthermore data specific to the com- mute to work were collected: The time travelled in minutes one way (duration of the commute), commute days per week, years of commuting, the travelled distance in kilometre one way, means of transport used (which were then regrouped into three groups, namely: public transport, motorised transport and pedestrian and cyclists), number of commutes stages one way (changes between and within means of transport) and finally the monthly costs of the commute. Hence the nature of the study, data were all self-reported and no physiological data were collected, a point that is further addressed in the discussion.
Commuting stress was measured using the stress on the commute-scale (a = .94 in the present study) and consisted of ten 5-point Likert-type items (1 = ‘‘I strongly disagree’’ to 5 = ‘‘I strongly agree’’), such as ‘‘overall, commuting is stressful for me’’ and ‘‘commuting to work takes effort.’’ This scale was originally designed by Evans et al. (2002) and had been derived from items used by Kluger (1998) and Novaco et al. (1990).
The control on the commute-scale (a = .78 in the present study) was based on four Likert-type items (1 = ‘‘I strongly dis- agree’’ to 5 = ‘‘I strongly agree’’) taken from an unpublished scale by Evans and Wener (personal communication, 24 June, 2009) and included items like ‘‘I can control how long it will take me to get to work’’ and ‘‘I can choose at what time I com- mute to work’’.
The predictability on the commute-scale (a = .53 in the present study) was based on four Likert-type items (1 = ‘‘I strongly disagree’’ to 5 = ‘‘I strongly agree’’), also taken from Evans et al. (2002). Sample items included ‘‘commuting to work is con- sistent on a day to day basis’’ and ‘‘I can usually predict what time I will get to work.’’ The German versions of the stress-, control-, and predictability on the commute-scales were already used in a previous study (Sposato, 2011).
As Novaco and Gonzalez (2009) point out, the best way to index impedance still remains to be ascertained. Novaco et al. (1990) define their impedance measure in terms of distance and time and subsequently group participants into three groups: High-, medium- and low- impedance subjects. Due to the methodological approach we chose, we assessed imped- ance similarly to Kluger (1998) using an index derived from distance (km) and duration (min) of the commute. To create a balanced measure, we first standardised both distance (km) and duration (min) of the commute and then averaged the variables into one single measure for each participant. To avoid theoretical confusion and staying with the original denom- ination by Kluger (1998) this variable is going to be termed length of the commute, which denotes both duration and distance.
584 R.G. Sposato et al. / Transportation Research Part F 15 (2012) 581–587
3. Results
The results of the initial correlation analysis of all predictor variables and the outcome measure are presented in Table 1. The findings suggest that: (a) Commuting stress is correlated with all predictor variables except for the commute days per week. (b) Length of the commute and duration of the commute are expectedly strongly correlated and therefore show a similar pattern of negative correlations with control and commute stages. (c) The variable commute stages is negatively correlated with control and predictability and positively correlated with duration of the commute and length of the commute. (d) The var- iable years of commuting is positively correlated with control and negatively correlated with predictability.
To further investigate the relationship between the predictor variables and the outcome measure commuting stress, the data were analysed using multiple regression controlling for age, gender, transport mode, income and monthly costs of the com- mute. After a first exploratory regression had shown no effect for the variables income, age, transport mode, gender, commute stages and commute days per week, they were excluded from further analysis. All remaining predictor variables were analysed using hierarchical regression. The remaining control variable costs of the commute and variables from previous research were entered in the first block, followed by our newly added variable years of commuting in the second block and the interaction terms in the last block. To ensure the stability of the regression model and to avoid the problem of multicollinearity we ap- plied residual centering for the interaction terms (Little, Bovaird, & Widaman, 2006). Results are presented in Table 2.
The multiple regression model accounted for 58.4% of variance and yielded control (b = �.38, p < .000) as the most pow- erful predictor for commuting stress, followed by the duration of the commute (b = .25, p < .000), predictability (b = �.21, p < .000), the impedance measure: length of the commute (b = .14, p = .033), costs of the commute (b = .12, p = .015) and the years of commuting (b = �.08, p = .021). The interaction test yielded two significant interactions: duration � control (b = .10, p = .005) and predictability � control (b = .09, p = .030).
4. Discussion
A central finding of this study evolves around the variable control. It has previously been shown to be crucial to the field of environmental stress research (e.g. Cohen, 1980; Glass & Singer, 1972), and its role in commuting stress literature seems to be of similar importance. The effect of control we report here is sizeable (b = �.377, p < .000) and in line with Kluger (1998), as well as Wener and Evans (2011), who found their operationalisation of control to be the most highly correlated variable with their strain measures. The effect we found is stable even after controlling for potentially influential variables such as length of the commute and predictability, reconfirming the view of an additive effect of control on a more solid base.
Concerning the relationship of control and predictability we did not find a significant correlation for the two variables. However our results corroborate recent findings by Wener and Evans (2011), yielding a similar pattern of correlations for predictability & control and commuting stress. Further analysis entering the two variables separately into the regression model also showed that they account for distinct amounts of variance. These findings should prompt a change in our tradi- tional understanding of the relationship between predictability and control.
Seligman and Miller (1979) suggest that people seek predictability when control is not available. Thus, in low control envi- ronments such as public transportation, the important issue becomes, whether one can predict certain events or not. Sim- ilarly other researchers argue that predictability serves as a form of cognitive control in situations where little to no behavioural control is possible (Evans et al., 2002; Koslowsky et al., 1995). Our findings do not support this idea of low-con- trol as a precondition for the effects of predictability. It is clearly shown that predictability has a distinct effect on commuting stress, regardless of the level of control.
We postulate that there is a major difference between control and predictability. Our view is that predictability is a mea- sure reflecting temporary incidents, which force the commuter to adapt, rather than constituting a constant feature of the journey to work, such as control.
Table 1 Number of valid observations (n), mean, standard deviation (SD), correlations, and reliabilities (in bold along the diagonal), if available.
n Mean SD 1 2 3 4 5 6 7
Outcome measure 1. Stress 363 2.26 0.92 .94
Commute Characteristics 2. Duration 363 37.0 22.70 .59**
3. Length of the commute 358 �0.01 0.93 .65** .90** .88 4. Control 363 3.16 0.99 �.56** �.35** �.56** .78 5. Predictability 363 3.96 0.64 �.24** �.07 �.07 �.11* .53 6. Commutes stages 363 .32** .46** .40** �.21** �.12*
7. Days 363 4.71 0.62 �.07 �.08 �.10 .09 .05 �.01 8. Years 363 7.03 8.08 �.11* .03 .02 .12* �.12* .05 .14*
* p < .05. ** p < .01.
Table 2 Hierarchical regression of commuting stress on commute characteristics, the control variable and interaction terms.
B SEB b R2 DR2 P<*
Block 1 .56 .56 .000 Constant 3.84 0.311 .000 Control �.36 0.040 �.389 .000 Duration of the commute .24 0.055 .227 .000 Predictability �.27 0.051 �.191 .000 Length of the commute .16 0.063 .164 .011 Costs .13 0.052 .115 .017
Block 2 .57 .01 .036 Constant 1.78 0.263 .000 Control �.35 0.040 �.377 .000 Duration of the commute .24 0.054 .226 .000 Predictability �.28 0.051 �.196 .000 Length of the commute .17 0.062 .169 .008 Costs .13 0.052 .118 .013 Years �.01 0.004 �.075 .036
Block 3 .58 .02 .002 Constant 1.80 0.26 .000 Control �.35 0.04 �.38 .000 Duration of the commute .26 0.05 .25 .000 Predictability �.30 0.05 �.21 .000 Length of the commute .13 0.06 .14 .033 Costs .13 0.05 .12 .015 Years �.01 0.00 �.08 .021 Duration � control .12 0.04 .10 .005 Duration � predictability .10 0.06 .06 .127 Predictability � control .12 0.06 .09 .030
* The p value here is given for the F statistic for the amounts of explained variance per block and for the T statistic of the regression-coefficients per block.
R.G. Sposato et al. / Transportation Research Part F 15 (2012) 581–587 585
This is best shown by the results for the interaction effects of both control and predictability with the duration of the commute. The fact that the relationship of predictability and stress does not seem to be affected by the duration of the com- mute lends support to the idea that predictability and control can be seen as distinct concepts. To fully investigate the rela- tionship of control and predictability, an interaction term for the two variables was entered into the regression model yielding a significant interaction. As we applied residual centering, a significant interaction represents a distinct effect beyond the variance explained by the singular variables. This means that the combination of control and predictability syn- ergistically enhances the effect of the variables as single predictors. In other words, low control will aggravate the effects of low predictability and vice versa.
Summarising the above, we propose to revise our understanding of the relationship of predictability and control in three respects. First, the effect of predictability is not related to low control as a precondition. Second, we argue that predictability constitutes a temporary and intermittent stressor, whereas control acts as a constant stressor. Third, our findings prove that there is an interactive effect of predictability and control on commuting stress, which goes beyond the effect of the singular variables.
Further we found that the duration of the commute is significantly correlated with commuting stress and significantly interacts with control. These results are noteworthy for several reasons: (1) They corroborate findings by Evans and Wener (2006) and Wener et al. (2003) on the importance of the duration of the commute for commuting stress. (2) They shed light on the relationship between two commute characteristics, namely the duration of the commute and impedance. The dura- tion of the commute, simple as it is, has mostly been overlooked in commuting research. This is partly due to the above men- tioned research focus on the intensity of stressors rather than on exposure, but also because of the attention given to the impedance concept. Our methodological approach allowed us to explore the relative strength of these two interlinked con- cepts in predicting commuting stress. Stokols et al. (1978) originally introduced impedance as a measure superior to the duration of the commute as it does not only take into account the time spent on the travel to work, but also how this time relates to the distance travelled. Contrary to this, our findings suggest that the duration of the daily trip to work is a better predictor of commuting stress than our impedance measure length of the commute. However, our results clearly show that even after controlling for the highly correlated duration of the commute, length of the commute still has a significant impact on commuting stress, substantiating the contribution of the impedance concept to the commuting stress model. However it is important to mention that, highly correlated variables can create elevated levels of multicollinearity in multiple regression models and thus affect the coefficient estimates for highly correlated variables. In our case and despite the high level of cor- relation for length of the commute and duration of the commute, multicollinearity indicators (e.g. VIF) were well within lim- its, indicating sound coefficient estimation. In any case it is advised to consider these results with the necessary precaution. (3) The fact that only control shows a significant interaction with duration of the commute, while predictability does not, gives further evidence to our distinction between predictability and control.
586 R.G. Sposato et al. / Transportation Research Part F 15 (2012) 581–587
A variable, whose correlation with commuting stress turned non-significant when controlling for the other commute characteristics, is the number of commute stages. Further analysis using a stepwise method confirmed that the loss of signif- icance occurs when duration of the commute is added to the equation. This finding is in accord with Wener et al. (2003) and shows that benefits from a lower number of commute stages are mediated by shorter travels.
An interesting result of the present study is the effect costs have on commuting stress. We found an increase of commut- ing stress with rising costs of the commute. This variable has not been previously investigated and was initially thought of as a control variable. It is important to note that, due to the statistical analysis we carried out, the effect we found is not med- iated by related variables such as the duration of the commute or length of the commute. A possible explanation might be that the increase in costs for better train seats, or a better car brings relatively little change to fundamental dimensions of commuting such as delays and traffic jams. Furthermore anecdotal data suggest that the basic mechanism of longer dis- tances increasing the costs of the trip to work confronts workers with the commuting reality of paying more and more for something less and less enjoyable and therefore may cause annoyance and strain. The exact relationship and possible underlying mechanisms however remain to be investigated in future research.
Finally, we found a relationship between the years of commuting and commuting stress. Interestingly the significant effect we found was opposite to what we expected: With the years of commuting, commuting stress seems to be diminishing. A plausible explanation for this might be selection and adaptation processes that take place over time. This notion may be re- flected in results by Novaco et al. (1979), who found a significant effect of impedance for residential adaptation. Similarly, commuters, who find their way to work to be exceedingly stressful, might opt for a change of their situation. They can do so by finding a home closer, or in a more favourable location relative to their workplace, by changing their workplace, or by developing strategies to help them cope with their stressful commute. This kind of relationship calls for further examination in longitudinal and more in depth studies.
The present research design carries certain methodological shortcomings. As a cross-sectional design the possibilities to determine causality are limited. Our online-questionnaire was self administered and not necessarily in temporal proximity to the actual commute, that respondents were asked to rate, as it was done in similar studies (e.g. Wener & Evans, 2011). This may have affected the answers respondents gave due to lacking immediacy, as well as potentially increasing the risk of biased replies affected by mood, attitudes or particular incidents prior to the completion of the questionnaire. For self-report measures in general, percept-percept inflation skewing the actual correlations is another problem to be considered. These issues could be resolved in a true experiment or a longitudinal study. A further step towards stronger results could be achieved by amending our self-report measures with physiological ones. A mitigating fact however is that we used subjec- tive measures that are strongly anchored in the commuting stress literature. Our outcome measure, perceived commuting stress, has been used alongside physiological measures in several studies and has proven to be sensitive to effects of stress- ors, when physiological measures were (Evans & Wener, 2006; Evans et al., 2002; Wener & Evans, 2011; Wener, Evans, & Lutin, 2006; Wener et al., 2003). This should add weight to our findings. Our results with respect to predictability have to be interpreted carefully due to the weak reliability of the predictability on the commute-scale. The original reliability of .65 for the scale by Evans et al. (2002) did not suffer a change for the worse when transferred to German by Sposato (2011). For future research the construction of a new scale with better test-theoretical properties should be considered.
Apart from implicating stronger research designs and objective measures, few other suggestions for future studies on commuting stress can be made. First, the model we present must be explicitly tested, preferably using stronger statistical analysis methods, such as structural equation modelling. Second, to allow an evaluation of our findings, researchers must
Commmuting Stress
Control Environmental Stressors
Duration of the Commute
Impedance
Predictability
Costs
Years of commuting
Commute Characteristics
Stressors
Moderating effect
Direct effect
Interaction effect
Fig. 1. Model of commuting stress.
R.G. Sposato et al. / Transportation Research Part F 15 (2012) 581–587 587
seek to collect comprehensive data on different commute characteristics. Research treating only single factors at a time would be inhibitory to a thorough understanding of the complex determinants and dynamics affecting commuters’ health and well-being. Third, new findings uncovered in this study (effect of costs) should be tested for consistency and placed into a theoretical framework.
5. Conclusion
Summarising our findings, it is important to note that we do not claim causal evidence but see our work as a first step to a more complete understanding of commuting stress. The primary purpose of this study was to generate a framework of hypotheses and to offer reference points for future investigations in this research field. Based on our results and building on previous studies, we propose a model of commuting stress (Fig. 1).
One of the most interesting features of this study pertains to the wide range of variables that were investigated, allowing us to further examine the exact nature of effects reported in the commuting stress literature. The implications of this study are both of theoretical and practical importance. On the theoretical side, we shed light on several previously unstudied facts, and as a consequence we propose a revision of the concept of control and predictability and present an amended Commuting stress model. On a practical note this study delivers valuable scientific input to decision making processes in the mobility sector. The daily tribulations commuters encounter and the possible consequences to mental and physical health of a stress- ful travel to work deserve this attention and will become a key issue to address in a labour market that increasingly demands mobility from employees.
Acknowledgements
The authors would like to thank the reviewers for their thoughtful comments and Dr. Lukas K.J. Stadler for his valuable scientific advice.
References
Bollini, A. M., Walker, E. F., Hamann, S., & Kestler, L. (2004). The influence of perceived control and locus of control on the cortisol and subjective responses to stress. Biological Psychology, 67, 245–260.
Cohen, S. (1980). Aftereffects of stress on human performance and social behavior: A review of research and theory. Psychological Bulletin, 88, 82–108. Evans, G. W., & Carrere, S. (1991). Traffic congestion, perceived control, and psychophysiological stress among urban bus drivers. Journal of Applied
Psychology, 76, 658–663. Evans, G. W., & Cohen, S. (1987). Environmental stress. In D. Stokols & I. Altman (Eds.). Handbook of environmental psychology (Vol. I). New York: Wiley. Evans, G. W., & Wener, R. E. (2006). Rail commuting duration and passenger stress. Health Psychology, 25, 408–412. Evans, G. W., & Wener, R. E. (2007). Crowding and personal space invasion on the train: Please don’t make me sit in the middle. Journal of Environmental
Psychology, 27, 90–94. Evans, G. W., Wener, R. E., & Phillips, D. (2002). The morning rush hour: Predictability and commuter stress. Environment and Behavior, 34, 521–530. Glass, D. C., & Singer, J. E. (1972). Behavioral aftereffects of unpredictable and uncontrollable aversive events. American Scientist, 60, 457–465. Herry Consult GmbH. (2007). Traffic in Numbers. Vienna: Federal Ministry for Traffic, Innovation and Technology Department V/Infra 5. Kluger, A. N. (1998). Commute variability and strain. Journal of Organizational Behavior, 19, 147–165. Koslowsky, M., Kluger, A. N., & Reich, M. (1995). Commuting stress causes, effects and methods of coping. New York: Plenum Press. Little, T. D., Bovaird, J. A., & Widaman, K. F. (2006). On the merits of orthogonalizing powered and product terms: Implications for modeling interactions
among latent variables. Structural Equation Modeling, 13, 497–519. Novaco, R. W., & Gonzalez, O. (2009). Commuting and well-being. In Y. Amichai-Hamburger (Ed.), Technology and well-being. Cambridge: Cambridge
University Press. Novaco, R. W., Stokols, D., Campbell, J., & Stokols, J. (1979). Transportation, Stress, and Community Psychology. American Journal of Community Psychology, 7,
361–380. Novaco, R. W., Stokols, D., & Milanesi, L. (1990). Objective and subjective dimensions of travel impedance as determinants of commuting stress. American
Journal of Community Psychology, 18, 231–257. Schaeffer, M. H., Street, S. W., Singer, J. E., & Baum, A. (1988). Effects of control on the stress reactions of commuters1. Journal of Applied Social Psychology, 18,
944–957. Seligman, M. E. P., & Miller, S. M. (1979). The psychology of power: Concluding comments. In L. C. Perlmuter & R. A. Monty (Eds.), Choices and Perceived
Control. Hillside, NJ: Lawrence Erlbaum. Sposato, R. G. (2011). Stress and quality of life of commuters. Saarbrücken: VDM Verlag Dr. Müller. Statistik Austria (2004). Census 2001, Commuters. Vienna: Verlag Österreich. Stokols, D., Novaco, R. W., Stokols, J., & Campbell, J. (1978). Traffic congestion, type A behavior, and stress. Journal of Applied Psychology, 63, 467–480. Taylor, P. J., & Pocock, S. J. (1972). Commuter travel and sickness absence of London office workers. British Journal of Preventive and Social Medicine, 26,
165–172. Troup, C., & Dewe, P. (2002). Exploring the nature of control and its role in the appraisal of workplace stress. Work and Stress, 16, 335–355. Wener, R. E., Evans, G. W., & Lutin, J. (2006). Leave the driving to them: comparing the stress of car and train commuters. In 5th global congress on engineering
education, congress proceedings (pp. 199–203). Wener, R. E., & Evans, G. W. (2011). Comparing stress of car and train commuters. Transportation Research Part F: Traffic Psychology and Behaviour, 14,
111–116. Wener, R. E., Evans, G. W., Phillips, D., & Nadler, N. (2003). Running for the 7:45: The effects of public transit improvements on commuter stress.
Transportation, 30, 203–220.
- The influence of control and related variables on commuting stress
- 1 Introduction
- 2 Method
- 2.1 Participants
- 2.2 Procedure
- 2.3 Measures
- 3 Results
- 4 Discussion
- 5 Conclusion
- Acknowledgements
- References
memo3/wenerandevans.pdf
Transportation Research Part F 14 (2011) 111–116
Contents lists available at ScienceDirect
Transportation Research Part F
journal homepage: www.elsevier .com/locate / t r f
Comparing stress of car and train commuters
Richard E. Wener a,⇑, Gary W. Evans b,1
a Department of Humanities and Social Sciences, Polytechnic Institute of New York University, 6 Metrotech Center, Brooklyn, NY 11201, USA b Departments of Design & Environmental Analysis and of Human Development, Cornell University, Ithaca, NY 14853, USA
a r t i c l e i n f o
Article history: Received 5 October 2010 Accepted 11 November 2010
Keywords: Behavior Commuting Transportation mode Stress Mass transit Job strain Automobile
1369-8478/$ - see front matter � 2010 Elsevier Ltd doi:10.1016/j.trf.2010.11.008
⇑ Corresponding author. Tel.: +1 718 260 3585; fa E-mail addresses: [email protected] (R.E. Wener)
1 Tel.: +1 697 255 4775; fax: +1 607 255 0305.
a b s t r a c t
Commuting times and distances continue to increase in the United States with potential impacts to the environment as well as possible health consequences for the travelers, because of stress from the commuting trip. There is very little empirical information, how- ever, on the differences between various modes of commuting on commuter stress. This study provides a cross-sectional comparison of car and train commuters with multiple indicators of stress, including statistical controls for group characteristics. We compared commuters in the same geographic region, Metropolitan New York City, who had compa- rable starting and destination points, and were from homogeneous socioeconomic back- grounds. We also explored potential underlying psychological processes (i.e., control, effort, predictability) to help explain stress differences related to commuting mode. There were statistically significant differences for perceived commuting stress and mood. Car commuters showed significantly higher levels of reported stress and, more negative mood. Mediational analyses indicated that effort and predictability largely account for the ele- vated stress associated with car commuting.
� 2010 Elsevier Ltd. All rights reserved.
1. Introduction
Commuting long times and distances has become a regular part of the daily routine for most workers. The average trip to work for Americans in most urban areas has reached 30 min (American Community Survey, 2005). The most recent US census reveals that the average American worker spends more time each year commuting than on vacation (American Community Survey, 2005) nearly 100 h per year. Moreover, workers making ‘‘extreme commutes’’ (over an hour in each direction) represent the fastest growing segment of commuters in America (Reschovsky, 2004).
How people travel to work is in part a function of personal preference, which has been discussed in terms of comfort in the vehicle, addressing issues such temperature, air quality, noise, vibration, light, and ergonomics (da Silva, 2002). Mode choice, however, also reflects contextual factors including economics – the cost and acceptability of different commuting modes (Schade & Schlag, 2003) – and the relative ease and availability of various modes of transport. Most American workers live in spread-out cities or suburbs with little access to economical or efficient mass transit. Not surprisingly their commut- ing choices reflect these geographic and demographic constraints: the vast majority of American workers commute by car, and typically alone. Almost 88% of all workers get to their jobs by car, with 77.7% driving individual vehicles and 10.1% using carpools, while only 4.6% use some form of public transit. Only a few major cities are significantly different in the use of tran- sit modes – New York, Chicago, and San Francisco have more people commuting by train than by car (American Community Survey, 2005). By comparison, more than 10% of workers commute by public transportation in the United Kingdom (Dargay
. All rights reserved.
x: +1 718 260 3136. , [email protected] (G.W. Evans).
112 R.E. Wener, G.W. Evans / Transportation Research Part F 14 (2011) 111–116
& Hanly, 2004), and over half of Italians and Dutch commuters used alternatives to cars in a 1984 survey (Costa, Pickup, & Martino, 1988). In one Japanese study, over 50% of workers indicated they commuted by train (Commuting, 2005).
The continued expansion of commuting distance and time, overwhelmingly in single occupancy and car-pool vehicles, has obvious environmental consequences as it deepens reliance on fossil fuels. Pollution generated by internal combustion en- gines has health consequences for travelers (Frumkin, Frank, et al., 2004; Wener & Evans, 2011). Commuting can also be stressful, and the length of trip contributes to the stress experienced by workers (Evans & Wener, 2006; Wener, Evans, Phillips, & Nadler, 2003). Commuting by car elevates physiological markers of stress like blood pressure and neuroendocrine hormone levels compared to baseline measures (Bellet, Roman, & Kostis, 1969; Robinson, 1991; Simonson et al., 1968). Highway con- gestion increases blood pressure among car or bus drivers (Evans & Carrere, 1991; Novaco, Stokols, & Campbell, 1979; Schaeffer, Street, Singer, & Baum, 1988; Stokols, Novaco, Stokols, & Campbell, 1978; White & Rotton, 1998). It can also lead to increased absenteeism (Knox, 1961; Novaco, Stokols, & Milanesi, 1990), reduced job satisfaction (Koslowsky, Kluger, & Reich, 1995), and decreased task motivation (Novaco et al., 1979; Schaeffer et al., 1988; Stokols et al., 1978; White & Rotton, 1998).
There are fewer studies that have looked at stress for mass transit commuters. Singer, Lundberg, and Frankenhauser (1978) found increased physiological stress on crowded trains. White and Rotton (1998), in a field study with college stu- dents, found greater stress in car than bus commuters. Multimethodological indices of stress (i.e., neuroendocrine hormones, task motivation, self-report) were reduced when train commutes were improved by route changes that shortened commut- ing time and enhanced predictability of the trip (Wener et al., 2003). A review of studies concluded that although little empirical data were available, large numbers of rail commuters were unlikely to cause serious problems unless crowding interfered with passengers’ sense of predictability and control, although discomfort from poor design could negatively affect passengers (Cox, Houdmont, & Griffiths, 2006). Similarly, it has been suggested that stress from crowding on Japanese com- muter trains is mitigated by the reliability and predictability of service there (Meyer & Dauby, 2002).
Few studies have directly compared riders across modalities, such as train versus car commuting, on stress measures. Based on available information one might predict that car commuting is more stressful than train commuting, particularly because of differences in predictability and effort, both of which have been linked to environmental stress (Evans, Lercher, Meis, et al., 2001; Kluger, 1998). For example, the vagaries of traffic and sudden onset of accidents or other kinds of traffic jams make driving times for the commute to and from work unpredictable, especially in densely populated major metropol- itan areas. Driving also requires constant attention and effort – more so as conditions worsen. Trains are likely to be more predictable and less effortful as a mode of travel.
On the other hand, driving may afford a higher level of control for the driver. The driver has more ability to influence time of departure, route, and road speed. Williams, Murphy, and Hill (2008) found that drivers in the UK had higher levels of per- ceived control than those using other transit modes, and consequently lower levels of stress. Past research in other situations also indicates that control may be an important factor in reducing stress (Glass & Singer, 1972). Car commuting also affords a greater degree of control over social interaction, a critical aspect of privacy. If drivers do, indeed, have higher levels of per- ceived control than do train commuters, driving may be a less stressful mode of travel.
This cross-sectional study was designed to compare commuters who drive to work with those who use mass transit. We compared commuters in the same geographic region, Metropolitan New York, from homogeneous socioeconomic back- grounds. We also explored potential underlying psychological processes (i.e., control, effort, predictability) to help explain stress differences related to commuting mode.
2. Method
2.1. Participants
This was a cross-sectional study comparing car and train commuters who had roughly the same potential commute, but chose different modes for their travel. All were traveling from the same communities in Northern New Jersey to work in New York City (a trip of approximately 20–30 miles). Car commuters (n = 122) were recruited in several ways, through flyers left on parked cars, radio and newspaper ads, and an advertisement on the EZ Pass Web site. Train commuters (n = 164) were primarily recruited with flyers handed out at train terminals. Each participant was offered an incentive that equaled a sig- nificant portion of their commuting costs for 1 month (a monthly ticket for rail passengers, $100 cash for car commuters) for their participation.
Those wishing to participate phoned or sent an email to the study site at which point they completed a brief screening sur- vey to see if they qualified as subjects for the study. To qualify commuters had to be from the identified geographical area, make the same trip to work at least 4 days a week, have been commuting on this route for at least 1 year (M = 58 months), and expect to continue their commute for at least another year. The average duration of the commute to work was 75 min. The median income level was between $85,000 and $95,000 and 80% had a college degree. All subjects reviewed and signed a consent form approved by the Cornell University Committee on Human Subjects that provided full disclosure of procedures and risks.
2.2. Procedure
Commuters were sent a packet of materials containing survey forms prior to the day of data collection. Rating scales assessing perceived stress and effort were completed by the commuters immediately at the end of the commute to work.
R.E. Wener, G.W. Evans / Transportation Research Part F 14 (2011) 111–116 113
All scales were five point Likert scales derived from prior work on subjective evaluations of commuting (Kluger, 1998; Koslowsky et al., 1995; Novaco, Kliewer, & Broquet, 1991; Novaco et al., 1990). A sample stress item was ‘‘Overall, commut- ing is stressful for me.’’ This scale had good reliability (alpha = .80). Perceived effort (a sample item was ‘‘Commuting to work takes effort’’) was also reliable (alpha = .89).
2.2.1. Control factors Basic sociodemographic information was collected of all participants for use as statistical controls when necessary in the
cross sectional, mode shift, and travel mode comparisons. Gender, ethnicity, education, income, family composition, and information about residential location were collected.
3. Results
Basic descriptive statistics and zero order correlation matrix for the principal variables examined are included in Table 1. With the exception of commuting time to work and months on the route, the two samples of car and train commuters are similar. For the sociodemographic variables, only race is related to commuting mode with more white (83%) train passengers than car commuters (67%), v2 = 10.74, p < .001. Gender and marital status were unrelated to commuting mode. Because of the longer commuting time for train passengers (M = 83 min vs. 61 min) and longer time on the route for car passengers (M = 71 months vs. 55 months), all inferential analyses comparing car to train commuting incorporate statistical controls for commuting time, months on the route, and race.
With these statistical controls, there are statistically significant differences for perceived commuting stress and for mood. Table 2 includes the means, standard deviations and inferential results for each of these variables. Multiple regression with the three control variables was used for the inferential analyses. Car commuters showed significantly higher levels of re- ported stress and had more negative mood than did train commuters.
Table 1 Descriptive information and zero order correlations.
Variable Mean SD 1 2 3 4 5 6 7 8 9 10 11
1. Education 5.58 1.31 .28** 0.08 0.07 0.01 .13* .13* 0.06 �0.01 �0.05 2. Income 7.48 2.19 .12* .12* 0.07 �0.01 0.00 0.06 �0.08 �0.06 �0.01 3. Job hours 43.88 10 0.06 �0.01 �0.05 �0.06 0.06 �0.09 �0.03 0.03 4. Months on route 58.64 51.68 �0.11 �0.02 0.01 0.10 �0.01 �0.03 .16**
5. Commuting time 74.55 26.28 .17** .14* �.16** �0.06 0.08 �.40**
6. Commuting stress 3.28 0.74 .90** .27** .37** .49** 0.11 7. Effort 3.28 0.9 .32** .35** .45** .16**
8. Unpredictability 2.47 0.72 0.11 0.06 .50**
9. Mood 2.79 0.62 .19** .15*
10. Uncontrollability 3.36 0.74 �0.01 11. Sample 56% Train 12. Gender 58% Male 13. Race 70% White 14. Marital status 81% Married
* p < .05. ** p < .01.
Table 2 Effect of commuting mode on commuter stress and mood.
Variable Train Car b t Value Significance
Commuting stress 3.21 (.68) 3.37 (.81) 0.31 3.13 (270) 0.002 Mood 2.70 (.60) 2.88 (.62) 20 2.38 (269) 0.018
Table 3 Mediators of transport mode.
Train Car
Uncontrollability 3.37 (.72) 3.36 (.78) Unpredictability 2.16 (.48) 2.88 (.77)*
Effort 3.15 (.86) 3.44 (.93) *
* p < .01.
Fig. 1. Mediational model.
114 R.E. Wener, G.W. Evans / Transportation Research Part F 14 (2011) 111–116
In addition to examining the direct association of commuting mode with commuting stress and mood, we also explored whether mediation by effort and predictability might help explain the mode of commuting effects. The multiple regression, with controls, indicated that car commuters found that their trip required significantly more effort, and was significantly less predictable than did train commuters. Perceived control over the commute was not significantly associated with commuting mode and so we did not test this variable as a third possible mediating mechanism (see Table 3).
In order to test the mediational model, shown in Fig. 1, we repeated the original regression equations, including the sta- tistical controls (race, commuting time, months on the commute) but forced into each equation the hypothetical mediators prior to the mode of transportation term. If the impact of commuting mode on each of the two respective outcome variables (i.e., commuting stress, mood) is mediated by one or both of these underlying processes (see Fig. 1), then the previously sig- nificant link between commuting mode and the respective outcomes will become non-significant once they are partialled out of the equation.
For perceived commuting stress, both effort and predictability were significant mediators of the effects of commuting mode. Effort reduced the previously significant b weight from .31 to �.03 which is no longer significant. For predictability the b weight shrank to .11 which is also non-significant. For mood the beta weight was reduced from .20 to a non-significant .09 after the inclusion of the effort term and to .17 with the inclusion of the predictability term. In the latter case, the b went from being statistically significant (p < .02) to marginally significant (p < .08), suggesting partial rather than full mediation.
4. Discussion
For metropolitan New York residents, train commuting is less stressful and creates less negative mood than commuting by car. These cross-sectional findings incorporate statistical controls for differences in time of commute and race in other- wise homogeneous samples of car and train commuters. Further, as suggested in Fig. 1, part of the reason why commuting by private automobile is more stressful than by train is because car drivers experience the commute as more effortful and unpredictable than do train commuters. Contrary to expectation, however, both transit and auto modes have similar, rela- tively high levels of perceived control.
These results were found even though both car and train commuters used the commuting mode of their own choice. All subjects lived within reasonable proximity of both highways and rail access to the urban core. This is not to say that options were unconstrained. For many train riders, the cost of parking in New York City made driving to work prohibitively expen- sive. In addition, anecdotal data suggest that many car commuters felt that train commuting was difficult because of the dis- tance of their workplace from a station, or because they needed to use their car during the day for their work.
Limitations to validity and generalizability of the results of this study come from several sources. First, this study was a cross-sectional comparison using volunteer samples. Even though the populations are very similar, come from the same communities and have similar end points of their commute, and in spite of statistical controls for remaining background dif- ferences, there could be possible differences in populations that could affect results. Keep in mind, however, that such dif- ferences would need to be unrelated to income, education, race, or geographic location.
Future research should seek to compare modes using longitudinal designs. This approach, while inferentially superior poses major logistical challenges. Our intent in designing the present study was to collect longitudinal data so that we could compare stress among commuters who switched travel modes, but there were not a sufficient number of drivers switching to train routes to allow for this analysis. Future studies might, for instance, target organizations that are moving headquarter locations, causing employees to change their commute, in which case some are likely to have longer and other shorter trips, and some are likely to switch commuting modes. In addition, it could be important to look at the long term health conse- quences of different kinds of commutes, as exposure to different levels of daily commuting stress for the many years of com- muting careers could have significant effects on stress-related diseases.
These data come exclusively from self-report of psychological states. Data from multiple sources, including physiological and behavioral measures, might provide additional insights about commuting, stress, and well being.
Taking these concerns into account, the results herein are supported by other studies that suggest that the commute to work can be stressful (Bellet et al., 1969; Novaco et al., 1979; Schaeffer et al., 1988; Stokols et al., 1978; Wener et al., 2003; White & Rotton, 1998), and extend the findings by: (a) indicating differences in levels of stress between commuting modes and (b) revealing that effort and predictability of the commute largely account for these travel mode effects. Our results dif- fer from those of Williams et al. (2008) who found driving to provide increase perceived control and reduced stress. We found no differences in perceived control in relation to commuting mode. Future research should also investigate the param- eters of different kinds and levels of car commutes, to determine which factors affect the level of perceived control for driv- ers. For example, levels of traffic congestion could well influence the relative control enhancing benefits of automobile
R.E. Wener, G.W. Evans / Transportation Research Part F 14 (2011) 111–116 115
modes of commuting vis a vis different forms of public transit. Evans and Carrere (1991) found that the negative impacts of traffic congestion on urban bus drivers’ psychophysiological stress was largely mediated by perceived control.
The results herein are for data obtained from commuters in one metropolitan area, which may in some ways be unique. Differences in the way commuters in other regions experience their commute are possible. This might be especially true, our model suggests, if rail travel systems are less well-run and, hence, offer less predictable trips. Our data, and the mediational model, indicate that mass transit is not, in and of itself, less stressful, but, rather, only likely to be so when it offers a greater level of predictability and a trip that involves less physical and/or cognitive effort.
Exposure to significant levels of stress from an event that is experienced twice daily over extended periods of time is not a trivial occurrence given the extensive evidence that stress has important physiological, psychological and health conse- quences (Barling, Kelloway, & Frone, 2004; Cohen, Kessler, & Gordon, 1995). Indeed, some recent work in Germany has shown direct links between acute myocardial infarction and congestion experienced by car commuters (Peters et al., 2004). There is also evidence that commuting stress may spill over into the workplace, possibly affecting productivity, sat- isfaction, etc. (Wener, Evans, & Boately, 2005). Moreover, current demographic trends and housing economics do not bode well for the future commuting situation in many American metropolitan areas (American Community Survey, 2005).
For many commuters mode choice is related to economic and infrastructure conditions created by planning, policy, and funding decisions. This study suggests that one element that might be taken into account by transportation and planning agencies in making these decisions is the impact of commuting mode on stress. Choices that reduce travel time and effort and/or increase predictability may have important benefits to the worker and to the public at large. Reduction in stress could be another benefit of enhanced public transit infrastructure in addition to well documented environmental benefits. Efforts to provide ‘‘one-seat rides’’ for example (Lu, Shi, & Martland, 2002; Warsh, 2001), to lure car commuters to public transpor- tation, or to provide more or better information and reliable (hence more predictable) on-time performance (Wunderlich, Hardy, Larkin, & Shah, 2001) may have benefits to the commuter not previously considered.
Although there is a considerable literature on the contributions of occupational stress to worker health and well being (Barling et al., 2004), remarkably little scholarly attention has been devoted to commuting as a stressor. This is curious when one considers that for many workers commuting may be among the most stressful components of their work environment. How long people commute, under what traffic conditions, and as we suggest herein, their mode of commuting may all con- tribute to the health costs associated with commuting.
Acknowledgements
This research would not have been possible without the generous support of the Robert Wood Johnson Foundation and the New Jersey Department of Transportation. We also appreciate the support of New Jersey Transit, and the University Transportation Research Center. The ideas expressed in this paper are those of the authors and do not necessarily represent those of the Foundation or any other supporting organization. A version of this paper was originally presented at the ‘‘5th Global Congress of Engineering Education’’ organized by the UNESCO International Center for Engineering Education held in Brooklyn, NY, from 17–21 July 2006.
References
American Community Survey (2005). <http://www.census.gov/acs/www/>. Retrieved 10.01. Barling, J., Kelloway, K., & Frone, M. (Eds.). (2004). Handbook of work stress. Los Angeles: Sage. Bellet, S., Roman, L., & Kostis, J. (1969). The effects of automobile driving on catecholamine and adrenocortical excretion. American Journal of Cardiology, 24,
365–368. Cohen, S., Kessler, R. C., & Gordon, L. U. (1995). Strategies for measuring stress in studies of psychiatric and physical disorders. In S. Cohen, R. C. Kessler, & L.
U. Gordon (Eds.), Measuring stress: A guide for health and social scientists. New York: Oxford University Press. Commuting (2005). <http://www.japan-guide.com/topic/0011.html>. Retrieved 12.02.05. Costa, G., Pickup, L., & Martino, V. (1988). Commuting—A further stress factor for working people: Evidence from the European Community. International
Archives of Occupational and Environmental Health, 60(5), 377–385. Cox, T., Houdmont, J., & Griffiths, A. (2006). Rail passenger crowding, stress, health and safety in Britain. Transportation Research Part A, 40(3), 244–258. Dargay, J., & Hanly, M. (2004). Volatility of car ownership, commuting mode and time in the UK. In World conference on transport research, Istanbul, Turkey. da Silva, M. C. G. (2002). Measurements of comfort in vehicles. Measurement Science and Technology, 13(6), 41–60. Evans, G. W., & Carrere, S. (1991). Traffic congestion, perceived control, and psychophysiological stress among urban bus drivers. Journal of Applied
Psychology, 76, 658–663. Evans, G. W., Lercher, P., Meis, M., et al (2001). Community noise exposure and stress in children. Journal of the Acoustical Society of America, 109, 1023–1027. Evans, G. W., & Wener, R. E. (2006). Rail commuting duration and passenger stress. Health Psychology, 25, 408–412. Frumkin, H., Frank, L., et al (2004). Urban sprawl and public health. Washington, DC: Island Press. Glass, D. C., & Singer, J. E. (1972). Urban stress. NY: Academic Press. Kluger, A. N. (1998). Commute variability and strain. Journal of Organizational Behavior, 19, 147–165. Knox, J. (1961). Absenteeism and turnover in an Argentine factory. American Sociological Review, 26, 424–428. Koslowsky, M., Kluger, A., & Reich, M. (1995). Commuting stress. New York: Plenum. Lu, A., Shi, D. S., & Martland, C. D. (2002). The vital role of metropolitan access in intercity passenger transportation from the traditional limited-stop express to the
21st century ring railroad. Cambridge, MA: MIT. Meyer, W., & Dauby, L. (2002). Why is rail transport so attractive? The general situation of suburban rail travel throughout the world (pp. 4–7). Public Transport
International, Special edition. Novaco, R., Stokols, D., & Campbell, J. (1979). Transportation, stress, and community psychology. American Journal of Community Psychology, 7, 361–380. Novaco, R., Stokols, D., & Milanesi, L. (1990). Objective and subjective dimensions of travel impedance as determinants of commuting stress. American
Journal of Community Psychology, 18, 231–257.
116 R.E. Wener, G.W. Evans / Transportation Research Part F 14 (2011) 111–116
Novaco, R. W., Kliewer, W., & Broquet, A. (1991). Home environmental consequences of commute travel impedance. American Journal of Community Psychology, 19, 881–909.
Novaco, R. W., Stokols, D., & Milanesi, L. (1990). Objective and subjective dimensions of travel impedance as determinants of commuting stress. American Journal of Community Psychology, 18, 231–257.
Peters, A., Von Klot, S., Heier, M., Trentinaglia, I., Hormann, A., Wichmann, H., et al (2004). Exposure to traffic and the onset of myocardial infarction. The New England Journal of Medicine, 351, 1721–1730.
Reschovsky, C. (2004). Journey to work: 2000. Census 2000 Brief. Robinson, A. (1991). Lung cancer, the motor vehicle, and its subtle influence on bodily functions. Medical Hypotheses, 28, 39–43. Schade, J., & Schlag, B. (2003). Acceptability of urban transport pricing strategies. Transportation Research Part F: Psychology and Behaviour, 6(1), 45–61. Schaeffer, M., Street, S., Singer, J., & Baum, A. (1988). Effects of control on the stress reactions of commuters. Journal of Applied Social Psychology, 11, 944–957. Simonson, E., Baker, C., Bruns, N., Kepier, C., Schmitt, O., & Stackhouse, S. (1968). Cardiovascular stress produced by driving an automobile. American Heart
Journal, 75, 125–135. Singer, J., Lundberg, U., & Frankenhauser, M. (1978). Stress on the train: A study of urban commuting. In A. Baum, J. Singer, & S. Valins (Eds.). Advances in
environmental psychology. The Urban Environment (vol. 1). Hillsdale, NJ: Lawrence Elrbaum Associates. Stokols, D., Novaco, R. W., Stokols, J., & Campbell, J. (1978). Traffic congestion, Type A behavior, and stress. Journal of Applied Psychology, 63, 467–480. Warsh, J. T. (2001). EA 21 success stories. Testimony to the highway and transit subcommittee. The United States house of representatives. <http://
www.house.gov/transportation/highway/11-01-01/warsh.html>. Retrieved 30.10.06. Wener, R. E., & Evans, G. W. (2011). Transportation and Health: The Impact of Commuting. In A. Diez Roux (Ed.), Encyclopedia of environmental health. New
York: Elsevier. Wener, R. E., Evans, G., & Boately, P. (2005). Commuting stress: Psychophysiological effects on the trip and spillover into the workplace. Transportation
Research Record, 1924, 112–117. Wener, R. E., Evans, G. W., Phillips, D., & Nadler, N. (2003). The effects of public transit improvements on commuter stress. Transportation, 30, 203–220. White, S., & Rotton, J. (1998). Type of commute, behavioral aftereffects, and cardiovascular activity. Environment and Behavior, 30, 763–780. Williams, G., Murphy, J., & Hill, R. (2008). A latent class analysis of commuters’ transportation mode and relationships with commuter stress. In Fourth
international conference on traffic and transport psychology, Washington, DC. Wunderlich, K. E., Hardy, M. H., Larkin, J. J., & Shah, V. P. (2001). On-time reliability impacts of advanced traveler information services (ATIS). Case study (no.
0900610D-01), Federal Highway Administration, Washington, DC.
- Comparing stress of car and train commuters
- Introduction
- Method
- Participants
- Procedure
- Control factors
- Results
- Discussion
- Acknowledgements
- References