BA

FRAMING HEALTH MÄHERS

Wage Theft as a Neglected Public Health Problem: An Overview and Case Study From San Francisco's Chinatown District

Meredith Minkler, DrPH, Alicia L. Salvatore, DrPH, Charlotte Chang, DrPH, Megan Gaydos, MPH, Shaw San Liu, Pam Tau Lee, Alex Tom, Rajiv Bhatia, MD, MPH, and Niklas Krause, MD, PhD, MPH

Wage theft, or nonpayment of wages to which workers are legally entitled, is a major contributor to low income, which in turn has adverse health effects. We describe a participatory research study of wage theft among immigrant Chinatown restaurant workers. We conducted surveys of 433 workers, and developed and used a health department observational tool in 106 restau- rants. Close to 60% of workers reported 1 or more forms of wage theft (e.g., receiving less than minimum wage [50%], no overtime pay [>65%j, and pay deductions when sick [42%]). Almost two thirds of restaurants lacked required minimum wage law signage. We discuss the dissemination and use of findings to help secure and enforce a wage theft ordinance, along with implications for practice. (Am J Public Health. 2014;104;1010-1020. doi;10. 2105/AJPH.2013.301813)

Growing appredation of the sodal determi- nants of health has helped reconned public health with its roots in concern for social justice and the interdependence of fadors such as poverty, residence in poor neighborhoods, racism, and adverse health outcomes.'^* Yet, despite a growing body of literature documenting numerous adverse impacts of low income status on health,''̂ '̂ the issue of wage theft has been largely neglected in public health. Wage theft—or the nonpayment of wages and benefits to which workers are legally entitled—takes many forms, among them nonpayment of earned overtime, tmdeipayment of Üie minimtim wage, delayed or nonpayment of back wages, misdassification of employees as independent contradors, confisca- tion of tip money, and failure to grant mandated breaks or paid sick leave.

Although found across many industries and recently described as "a defining trend of the 21st century job market"^ wage theft is par- ticularly prevalent in low-wage jobs, in which a full-time, full-year worker does not earn enough to bring a family of 4 above the federal poverty line of $21 200.*̂ Between a fifth and a third of all workers are estimated to be in low-wage jobs, where wage theft is at epidemic proportions. ®~'°

Following a brief review of wage theft and its cormection to public health, we describe the

methods, findings, and policy and related out- comes of a community-based partidpatory re- search (CBPR) study of wage theft among immigrant Chinese restaurant workers. Using Kingdon's' ' model of the policymaking pro- cess, partictilar attention is paid to the trans- lation, dissemination, and use of findings to help develop and pass a dtywide ordinance banning wage theft, and follow-up action leading to landmark levels of enforcement We discuss implications for public health re- search, practice, and policy.

A GROWING EPIDEMIC OF WAGE THEFT

Across the nation, millions of workers are experiencing wage theft, the ripple effects of which also have an impad on the health of families, businesses, communities, and local economies. FtiUy 4.36 million workers re- ceived less than rrunimtmi wage in 2010.'^ Of these, almost two thirds were women and youths, African Americans, and Latinos.''' Low-wage immigrant workers are particularly likely to be victims of wage theft, in part because of their vulnerability to exploitation as a result of language, education, and dtizenship status.'*-*^

Pa3íment of less than minimtmi wage is most common in agricultural work, difld and home care.

clothing mantifacturing, repair services, day labor, and venues such as restaurants.®'®'*''̂ Violations of legal paid overtime are also prevalent, particularly among low-wage workers. A representative survey of 4387 such workers in New York, New York; Chicago, Illinois; and Los Angeles, California, fotmd that more than 76% had not received earned overtime pay the preceding week, and 69% had failed to receive meal breaks or had them reduced in time—all violations of labor law.® Workers in the hospitality industry, naü salons, and other venues where tipping is common also frequently face confiscation of tip money.®'®'"^ Finally, misdassification of employees as independent contractors is an increasingly common form of wage theft. State-level data suggest that between 10% and 30% of employers misclassify workers, in part to avoid having to comply with minimum wage and overtime laws, paid leave, and other basic accommodations.*^

REDISCOVERING AN IMPORTANT CONNECTION BETWEEN WAGE THEFT AND PUBLIC HEALTH

Although few studies have examined low wages separate from education or family in- come as contributors to health problems,® wages are indeed the largest category within income status, accotmting for more than 60% of income for all workers and 80% for those in the paid labor force.'®'® In one of the few studies on low wages versus low-income status as a contributor to adverse health out- comes, Leigh and Du examined nationally representative longitudinal US data over 4 waves from 1999 to 2005 and found strong negative correlations between wages and hypertension, particularly among younger workers (aged 25—44 years) and women.® Other studies have sho'wn links between low

1010 I Framing Health Matters \ Peer Reviewed 1 Minkler et al. American Journal of Public Health | June 2014, Vol 104, No. 6

FRAMING HEALTH MAÏÏERS

wages and diabetes, obesity, and breast

Wage theft—a major contributor to low wages and low-income status—has not been the focus of much public health research. Logic and reason suggest, however, that wage theft may exacerbate adverse health impacts of low wages and low-income status by generating income insecurity. Individuals not paid for hours worked, or paid less than what they earned, may not reliably be able to pay rent or heating, buy groceries, or access transit This in turn may result in increased crowding or homelessness, hunger, decreased mobility, and decreased ability to pay for childcare or med- ical care—all having an adverse impact on health. Wage theft may also increase the number of hours or jobs worked, which may in turn decrease time spent with famüy, leisure time for physical activity, and sleep or rest.'^

But wage theft also negatively affects society, creating unfair competition with law-abiding businesses, increasing demand on safety-net programs, and reducing needed tax revenues. According to one recent study,^^ wage theft is a $30 billion problem nationally. Employers that violate wage laws are also less Ukely to offer health insurance and paid sick days.® A üttle-studied aspect of wage theft—lack of en- forcement of mandated paid sick leave—has particularly clear implications for public health. Several studies have shown a relationship be- tween lack of paid sick leave and serious heart events, undertreated or exacerbated injuries or ülnesses,^* increased spread of influenza,̂ ^ and failure to get mammograms and other cancer screenings.̂ ® At the time of this writing, only 3 cities, the District of Columbia, and Connecticut, have enacted paid sick leave legislation^^; even in these jurisdictions, the extent of compliance is difficult to discern as monitoring and compliance tend to be reactive, not proactive in nature.

To further explore wage theft as a public health issue, we turn to a study conducted in San Erancisco, CaUfomia, which has the highest minimum wage rate in the nation as well as some of the nation's only local employer mandates for health insurance cover^e and paid sick leave. Even in this progressive labor environment, our research in the dty's Chinatown District illustrates that wage theft is rampant and undermining healthy working

and living conditions for many workers. This study provides new local-level data, and we illustrate how the findings were used as part of the evidence base for municipal-level policy change.

A CASE STUDY OF IMMIGRANT RESTAURANT WORKERS IN SAN FRANCISCO'S CHINATOWN

We used a CBPR approach to study wage theft as part of a larger investigation of the health and safety of immigrant restaurant workers in San Erandsco's Chinatown District̂ ® An orientation to research, raûier than a re- search method, CBPR is "systematic investiga- tion, with the collaboration of those affected by the issue, for the purposes of education and action or effecting change."^®'''̂ ' The topic of wage theft quickly emerged from worker and community partners as the single most impor- tant issue on which they wanted to collect data The focus of the study therefore broadened to include a heavier emphasis on wage theft along with unsafe physical conditions and other, more traditional occupational health concerns.

San Erandsco's Chinatown District is home to approximately 9400 people, most of them first- or second-generation immigrants whose median household income is $17 630, roughly one quarter the dtywide average of $70 040.^° As in other parts of the countiy, restaurants are the largest employer of Chinese immigrants in Chinatown, vnth a third of the neighborhood's workers in this industry. With a few notable exceptions,'^'^'"'''' little research has been conducted with low-wage restaurant workers in general, and even less on wage theft among immigrant workers in restaurants. Yet, in addition to having among the highest num- bers of nonfatal work-related physical injuries of any workers in the country^^ and high levels of psychosodal stressors,''̂ '̂ ^ low-wage res- taurant workers appear to be disproportion- ately subject to wage theft (e-mail communica- tion, D. Levitt, director. Office of Labor Standards Enforcement, July 3, 2013).^'^'^'

DATA COLLECTION

We collected data from restaurant workers and from restaurants through (1) a cross- sectional survey of Chinatown restaurant

workers (n = 433) and (2) an observational survey of restaurant working conditions con- ducted by health department staff (n= 106).

Community-Based Participatory Research

Described by Horowiti et al.̂ * as "the ultimate form of translational research," CBPR may employ any of a wide range of research designs and methods. It engages community and other partners throughout the research process, from selecting and refining the re- search questions through interpretation, dis- semination, and use of findings to help effect change.^^ A strong accent also is placed on recognizing and building on the assets of all partners and planning for sustainability."^ '̂''®

To conduct this multimethod study, a part- nership was formed in 2007 consisting of tiie Chinese Progressive Association of San Erancisco (CPA); the University of California, Berkeley School of Public Health and its Labor Occupational Health Program; a professor of occupational medidne at the University of California, San Erancisco; and the Environ- mental Health Section of the city's Departinent of Public Health (SEDPH). In addition, we hired and trained a core group of Chinatown res- taurant workers who served as a key partner in the project. Previous collaborations among several of the partners on other CBPR projects and the project coordinator's deep roots in Chinato'wn greatiy facilitated partnership crea- tion and initial trust among partners. The partnership aimed to follow CBPR prind- pjgg35.36 gjjj gjgQ induded an ongoing partic- ipatory evaluation of processes and out- comes."̂ '̂ A 12-member steering committee with members from each partner organization served as the project's primary coordinating and decision-making body.

The CPA hired 9 current or recent restau- rant workers as worker coordinators. Predom- inately women in their 30s and 40s, they were monolingual Cantonese speakers or had very limited English-language skills. An initial group of 6 partidpated in an early 8-week training on topics ranging from workplace health and workers' rights to human partidpants protec- tions and survey administration. An expanded group then met weekly or biweekly to provide feedback on study instruments and help de- velop a survey püoting and recruitment plan.

June 2014, Vol 104, No. 6 | American Journal of Public Health ivlinMer et al. \ Peer Reviewed | Framing Health Matters | 1011

FRAMING HEALTH MÄHERS

Barriers to participation included time con- straints, with 2 coordinators dropping out because of work scheduling and family reasons and other current restaurant workers lacking the time to participate at the coordinator level.

Restaurant Worker Survey

The partners developed a 103-item ques- tionnaire that included numerous questions about participants' experiences in restaurant work, validated scales to measure variables such as depression, and sododemographic questions.''®

Survey questions determined the usual number of work hours per day, work days per week, overtime, hourly wages, monthly per- sonal and family income, household size, and prevalence of paid overtime, vacation, holidays, sick leave, and uninterrupted 30-minute lunch breaks. Additional questions suggested by worker partners assessed whether the restau- rants 'withheld initial paychecks, delayed or owed back wages, took part of workers' tip money, or called nonsmokers in earlier from

We calculated yearly incomes on the basis of weekly work hours. We estimated yearly per- sonal income as reported hourly wage times 2000 hours, assuming fuE-time work of 40 hours per week, 50 weeks annually. This assumption overestimates regular work hours (by ignoring part-time work) and underesti- mates total work hours (by ignoring overtime work) but comes dose to the average nimiber of 38.2 regular weekly work hours reported by our sample.

We assessed poverty in several ways: in relative terms as hourly wages below 60% ($ 10.09) of the national median hourly wage in private industry ($16.81) as reported by the Bureau of Labor Statistics*"; in absolute terms by comparing famüy income with federal pov- erty levels published for 2008*'; and by comparing personal income with the respective poverty cut point of $21 200 for a 4-person household.

We assessed several indicators of wage theft. We compared hourly wages with the city's minimum wage ordinance for 2006 through 2008, respectively, depending on the last year of Chinatown restaurant employment for which wages were reporied. We assumed that there was a violation of the state law requiring

a 30-minute uninterrupted lunch break if the worker denied getting such a break, or reported a shoriened break or employer in- terruptions during break time. We assessed compliance with the city's sick leave ordinance by asking employees "If you stay home sick one day, are you paid wages for that day?" Finally, several questions assessed the prevalence of potential tip theft induding: "boss or company gets part of tips," "boss takes all tips," "boss takes a deduction from credit card tips," "credit card sups taken out of tip pool."

We quantified work hours, wages, wage theft, and poverty level for each individual worker and averaged for all workers, by gender and by 5 main occupational groups (cooks; kitchen workers, dishwashers; waiters, dim sum sellers, bus persons; cashiers; others). We quantified wage loss for violations of legal minimum wage, overtime, lunch break, and sick leave. We calculated overtime wage losses by assigning 1.5 times the hourly wage to the number of extra hours reported per week. For workers who did not get a 30-minute lunch break, we assigned a half-hour wage loss per workday. For workers not receiving paid sick leave we assumed the national average of 3 used sick days per year*^ at their actual wages. We averaged all wage losses over the total sample including workers with no losses. Fi- nally, we estimated the total amount of wage theft among Chinatown restaurant workers by adding the wage loss averages and multiplying this sum by 2500, the CPA's estimate of the total Chinatown restaurant workforce.

In addition to the worker coordinators, the CPA hired and trained 17 restaurant workers and community members to conduct the res- taurant worker survey in the winter of 2 0 0 8 - 2009. Adults aged at least 21 years who were currently working in or had worked in a Chinatown restaurant within the past 24 months were eligible to partidpate. Recruit- ment took place in venues identified as safe and appropriate by the CPA and the worker co- ordinators (e.g., popular coffee shops and a lo- cal park, as well as in the single-room occu- pancy hotels where about one third of the workers uved). No recruitment was conducted in restaurants. The questionnaire was written in Chinese and was administered by surveyors in Cantonese, Mandarin, or Toishanese. A total of 433 restaurant workers took the survey. We

excluded 28 from analysis because of incom- plete data or if the partidpant's work in a Chinatown restaurant had ended more than 2 years ago.

Restaurant Observation Checklist

Health departments' routine food safety in- spections have historically focused solely on consumer protection. As part of our study, SFDPH drafted, sought study partners' and restaurant inspectors' input on, and informally tested a new worker safety-focused observa- tional tool. The restaurant worker safety checklist enabled the collection of information about occupational hazards, safety equipment, and 3 required labor law postings (for workers' compensation and the dty's minimum wage and paid sick leave laws). Worker coordinators' recommendations resulted in important im- provements to the final tool—for example, in- suring that the checklist not only assessed whether required signage on labor laws was visible, but also whether these postings were in Chinese.

Health department staff members of the project team conducted the completed and pretested checklist in 106 of Chinatown's 108 restaurants during routine food safety inspec- tions. It took approximately 15 minutes to complete.*''

DATA ANALYSIS AND INTERPRETATION

University partners deaned and led the analysis of the survey data, as did health department partners for the restaurant data. Concurrently with data interpretation sessions invol'ving all members of the partnership, uni- versity and health department partners con- ducted descriptive data analyses and produced reports detailing results''^'*'' (see Data Collec- tion). During this period, the lead analysts regularly shared emerging results with aU partners "in real time" and sought their feed- back at steering committee, worker coordina- tor, and other project meetings. Indeed, a critical component of the analysis phase was the ongoing involvement of restaurant workers in data interpretation.

Six data interpretation workshops were held at CPA with the worker coordinators. Conducted in Cantonese with interpretation for

1012 I Framing Health Matiers | Peer Reviewed | Minkler et al. American Journal of Public Health | June 2014, Vol 104, No. 6

FRAMING HEALTH MAÏÏERS

university and health department partners as needed, these workshops employed hands-on learning to teach worker coordinators to speak "data language" and to facilitate interpretation of findings. For example, worker coordinators would first be asked to formulate their own "guesstimates" of the survey findings, such as the percentage of respondents who reported receiving less than minimum wage. The actual findings from the data analysis would then be revealed and discussion and interpretation would follow. The CPA and worker coordina- tors provided many insights into the data not originally apparent to universify and health department partners, and some of which added important new information to findings (e.g., concerning the common tinderstanding of "paid sick leave" in this community) that oth- erwise might have been misinterpreted.

RESEARCH FINDINGS

Table 1 provides an overview of the sodo- demographic characteristics of the workers, excluding those with incomplete data (n = 405). The majority was female (69%) and aged 40 years or older (72%), with 4 3 % aged 40 to 49 years. Nearly 80% reported speaking little or no English, with 50% having less than a high-school education and just 5% having any college or vocational training. Although 55% reported being in the country 6 years or more, more than half (56%) reported being nondtizens. Fifty-two percent were native Cantonese speakers with the second largest group (41 %) naming Toishanese as their native language. Only 1% were native Fngbsh or other non-Chinese language speakers. Study participants were roughly divided between those working in the "back of the house" (cooks and kitchen staff) and the "front of the house" (waiters, cashiers, bussers, etc.).

Table 2 displays regular work hours, over- time, wages, and several forms of wage theft for all workers and by gender and occupational group. Total weekly work hours, including usual overtime hours, exceeded 40 hours for most workers, with cooks, kitchen workers, and dishwashers often working 6 or 7 days and on average more than 50 hours per week.

Average hourly wages were below minimum wage and ranged from $7.11 among kitchen workers to $8.68 for cooks. Average yearly

personal income from restaurant work ranged from $11 092 for cashiers to $21 175 for cooks. These wages fall below relative poverty levels for 91.5% of restaurant workers and below absolute levels based on personal in- come for 75.3% of all workers. Taking into accotmt income from other household mem- bers wotild indicate absolute poverty for 23.7% of restaurant worker households. Most workers (80%) did not get any paid vacation or holidays.

Fifty-eight percent (n = 235) of workers reported experiencing wage theft. Minimimi wage law violations were experienced by 50% of all workers and by 70% of kitchen workers and dishwashers, resulting in an average wage losses of $3342 per year and nearly double this amotint ($6187) for kitchen workers and dishwashers. About 4 5 % of all workers failed to receive an tminterrupted lunch break, equivalent to an additional $106 wage loss per year for all workers. Although 40% of study participants reported working extra hours, 6 5 % did not get paid overtime, result- ing in $763 yearly wage losses for workers with overtime, or an average of $565 for all workers.

Mandatory paid sick leave was self-reported as not available for 42% of workers. Yet, as worker coordinators pointed out, even this figure may tmderestimate the extent of the problem, as there is a prevalent and erroneous belief that making up an extra day of work for one that was taken off without pay constituted "paid sick leave.""*"*

About a quarter of all workers reported that their boss withheld initial paychecks when they were hired, or delayed or owed wages at some time. A third of aU workers reported that the boss kept some or all tips and 59% responded that they did not know what happened to tips, suggesting that fewer than 8% of workers observed appropriate distribution of tips.

(No workplaces are exempt from the federal minimum wage on the basis of the number of employees. Most states apply the federal rule for tipped workers, which requires a minimum cash wage of $2.13 and a combined cash wage plus tips of $7.25. Employers in these states can pay a lower wage than minimum wage based on the amount of tips [i.e., a tip credit]. Ten states disallow tip credits, requiring em- ployers to pay tipped workers cash wages

equivalent to the applicable minimum wage irrespective of tips. Employers generally cannot take tips from workers who received tips from customers. Employers can apply a service charge [instead of accepting tips] and dedde how to use or distribute that service charge. Eor example, they could use the service charge to pay wages. Employers in some states can re- quire tip pooling. The pooled tips could be shared among service employees and be ap- plied as a tip credit.)

The estimated total wage loss for 2500 Chinatown restaurant workers in 2008 was $ 10 450 000, not induding loss of state or federal income taxes, Sodal Security or other retirement benefits, or health care, disability, or unemployment insurance premiums. The total estimated amount of wage theft in Chinatown restaurants attributable to minimum wage vi- olations alone was $8 605 965.

Results of the SFDPH partner's restaurant- level observations are reporied elsewhere.'*'' Two of the findings, however, were highly relevant to the problem of wage theft. First, only 3 5 % of the 106 restaurants observed had any of the 3 labor law signs of concern posted, including minimum wage law and paid sick leave signage. Second, of the few restaurants that had the postings, 10 (27%) had them solely in English in this largely monolingual Chinese neighborhood, even though such postings are available in multiple languages induding Chinese."*'̂

DISSEMINATION AND ACTION

The dissemination and action phases of this projed took many forms including bimonthly educational teas for workers on occupational health, safety, and worker rights, and bringing pressure to bear on individual restaurant owners to pay back wages. The health de- partment's Environmental Health Section also took several tmprecedented administrative steps to sanction wage theft violators, legiti- mizing these actions primarily as health pro- tective responses to demands from workers' rights groups. Although the department had the authority to sanction restaurants for violations of other dty and state laws, the study findings and community interest group pressure helped the department defend using the health code to suspend restaurants that had committed

June 2014, Vol 104, No. 6 I American Journai of Public Heaith Minkler et al. | Peer Reviewed | Framing Heaith Matters | 1013

FRAMING HEALTH MÄHERS

TABLE l-Sociodemographic Characteristics, Chinatown Restaurant Workers: San Francisco, CA, 2 0 0 8

Characteristic

Gender

Male

Female

Age, y

21-29

30-39

40-49

50-59

>60

Marital status

Married or living together

Single

Divorced, separated, or widowed

Highest level of education completed

None

Elementaiy school

Junior high or middle school

High-school diploma

College or graduate school

Vocational or technical training

Tinae lived in United States, y

< 1

1-5

6-10

>10

US citizenship

Yes

No

Country of birth

China

Did not report

Level of English spoken

None

A little

Basic conversation

Complex conversation

Ruent

Household size

1-3 people

4-6 people

7-9 people

Children living in household

Yes

No

No. (%)

126 (31.1)

279 (68.9)

34 (8.5)

80 (20.0)

174 (43.4)

91 (22.7)

22 (5.5)

347 (85.7)

38 (9.4)

20 (4.9)

2 (0.5)

57 (14.2)

143 (35.7)

180 (44.9)

17 (4.2)

2 (0.5)

10 (2,5)

175 (43.2)

91 (22.5)

129 (31.9)

174 (43.5)

226 (56.5)

398 (98.3)

7 (1.7)

144 (35.7)

173 (42.9)

62 (15.4)

18 (4.5)

6 (1.5)

202 (50.4)

193 (48.1)

6 (15.0)

295 (72.8)

110 (27.2)

Continued

wage theft. The health department also sent letters to the heads of regulatory agencies, such as the California Division of Occupational Safety and Health and the local Office of Labor Standards and Enforcement, sharing findings that related directly to areas for which these bodies had enforcement responsibility and offering help in promoting compliance. De- scriptive reports of findings firom the worker survey and restaurant observations were pre- pared and made available online, with the observational checklist shared through the SFDPH's Web site (http://www.sfphes.org/ component/jdownloads/finish/Sé-chinatown- restaurant-workers/65-restaurant-health-and- safety-checklist/O?Itemid=O).

Several peer-reviewed presentations and publications were used to disseminate some of the findings from the larger study to profes- sional audiences.̂ ®'̂ ^""'®'̂ ''*"' We focus here, however, on how study findings specifically concerning wage theft were disseminated and used to help bring about broader policy-level change, under the leadership of the study's community partner. As a framework for anal- ysis, we use Kingdon's" model of the 3 "streams" in the policymaldng process. Briefly, the problem stream involves gaining visibility for an issue and convincing policymakers that a problem exists. The policy stream then re- quires proposing feasible, politically attractive proposals to address the problem. In times of fiscal austerity, cost-saving, cost-neutral, or cost-efîective proposals may cany special weight. The politics stream then involves ne- gotiating the politics that influence whether a proposal succeeds in the political arena Having strong relationships with key polit- ical figures who can help sponsor or support the proposed measure is especially critical at this stage, as is evidence of growing popular and media support for its passage. Finally, a policy window opens when favor- able developments in all 3 streams create a special opportunity for policy change to take place."

Problem Stream

To help gain visibility and move the issue of wage theft onto the policymakers' agenda, study findings concerning wage theft, as well as poor and unsafe working conditions, were translated into media and other user-friendly

formats. The CPA produced a colorful report, entitled Check Please.' Health and Working

Conditions in San Francisco Chinatown Restau-

rants'^^ in September 2010, with assistance from the other partners and writing support from the Data Center, a local nonprofit. Printed in Chinese, English, and Spanish, and also made available on CPA's Web site (http:// www.cpasf org), the report highlighted key findings, but also workers' stories, which helped contextuaHze the findings and added "political punch" to the numbers and frequen- cies. A page with statistics on wage theft, for example, also included a poignant quote from a worker who had remarked, "We don't get minimum wage, maybe four dollars an hour. Think about it, $1,200 for an entire month, working 10 hours a day, and six days a week."̂ 5<P">

A diverse group of 170 people attended the well-publicized report launch and press event, including approximately 20 members of the mainstream and ethnic media and 4 of the 11 city supervisors. Several members of the part- nership, including worker coordinators, spoke at the event, as did 2 supervisors. The report release was widely covered in the nightly television news, the San Francisco Chronicle'*^

(the major local newspaper with 1.8 million readers), stories in major Chinese news outlets, and videos on YouTube.

To gain the attention of policymakers as part of the problem stream, broadening the base of support is aitical.''®'̂ ^ The CPA therefore helped create a dtywide coalition, the Pro- gressive Workers Alliance (PWA), with orga- nizations representing diverse radal/ethnic, occupational, and other groups among the dty's large low-wage worker population. The PWA in turn helped craft an action agenda— a "low-wage workers' bill of rights"—that was prominently featured in the report. Although much of the data on which Check, Please! was based came from the Chinatown study, illus- trative data on other worker groups also was induded, such as news artides on wage theft among domestic workers, day laborers, care- givers, and restaurant workers in other geo- graphic areas. Finally, and to broaden its base to include employers who took "the high road" in their treatment of workers, the report's bul of rights included supporting and rewarding "responsible businesses."'*

1014 I Framing Health Matters | Peer Reviewed | Minkler et el. American Journal of Public Healtti | June 2014, Vol 104, No. 6

FRAMING HEALTH MAÏÏERS

TABLE l-Continued

Monthly family income, $

< 7 5 0

751-1500

1501-2000

> 2 0 0 1

Employment status

Not currently working

Working part-time (< 40 h/wk)

Working full time (> 40 h/wk)

Employed in restaurant at time of survey

Yes

No

Restaurant position'

Back of the house

Front of the house

Other

13 (3.5)

80 (21.5)

104 (27.9)

176 (47.2)

58 (14.3)

190 (46.9)

157 (38.8)

313 (77.3)

92 (22.7)

196 (48.4)

196 (48.4)

13 (3.2)

Note. The sample size was n = 405.

"Back of the house positions include kitchen work

such as cooking, food preparation, and dishwashing.

Front of the house positions include serving and

selling of food (i.e., cashier and take-out counter

work) as well as waiting and bussing positions. Other

includes janitorial work, food bagging, food delivery,

and sidewalk promotional activities such as handing

out leaflets.

Policy Stream

Particularly in a time of severe economic downttim, a wide-ranging proposal addressing all of the planks covered in the original bill of rights was unlikely to be feasible or attractive to policymakers. The topic of wage theft, how- ever, had caught the imagination of both the media and poKcymakers. In the words of a community partner, "Wage theft is sexy; it's what gets tis in the door." As a consequence, CPA and the PWA worked with 2 dty super- visors to craft a more spedfic Wage Theft Ordinance, based in part on data from the Chinatown study, that was introduced at a rally and hearing at City Hall in May 2010. Spon- sored by the PWA and the CPA, the rally featured a range of speakers, induding worker coordinators, as well as political theater and remarks by the bill's cosponsors. This event, too, received widespread media coverage, in- cluding a lengthy artide in the San Frandsco Chronide^^ and video footage on YouTube (http ://www.y0utube.com/user/q3asf).

Several speakers at the rally dted the report findings in their presentations, with

a supervisor highlighting the pivotal role of the Chinatown projed's "action research and bot- tom up grassroots organizing." Consistent with Kingdon's' ' emphasis on the importance of proposing feasible solutions that, in tough eco- nomic times, ideally are cost-saving or cost- neutral, speakers and segments of the report. Check Please.' emphasized the ways in which curtailing wage theft could save the dty money. For example, the study's finding that in 2008 alone Chinatovm restaurant workers lost $10 450 000, not counting the loss of state and federal income taxes, etc., was tised to illustrate how wage theft negatively afFected both workers and the revenue base of the dty. Check, PZease.'further cited recent studies demonstrating that, contrary to industry fears and daims, in- creasing wages and providing health care and sick leave benefits resulted in few adverse effects on business, with the provision of benefits sometimes also increasing employee retention.*^

Politics stream

By partnering early on with several sympa- thetic members of the board of supervisors, the CPA and its partners had laid critical grotmd- work that can be of real value "in the belly of the beast" of political negotiations. ̂ ° The con- tinued leadership role of 2 dty supervisors working closely with the CPA and PWA also was pivotal during the policy stream negotia- tions, as was the continuing attention to Check, Please! Although the report generated strong support from many quarters, it also brought active opposition from a major Chinatown "power broker." Ironically, however, the widely publicized opposition of this powerful but di- visive local figure brought much additional attention to the proposed ordinance, and ended up working to the advantage of the bill's pro- ponents in this liberal dty. As a commtmity partner commented.

She did us a favor by drawing a line in the sand.

Although some agreed with her [that adverse

publicity would hurt the restaurant industry

in Chinatown], many saw the need for a

change.

Subsequent community mobilization, and further base building to embrace Latino, Fili- pino, and other immigrant and nonimmigrant low-wage earners, including domestic workers, day laborers, and hotel room cleaners, further helped in the politics stream.

Policy Window

Favorable developments in all 3 policy streams (problems, polides, and politics) helped open a "policy window" through which the diverse partners and stakeholders cotild then jump to help secure the desired change. The continued support of dty supervisors, the ascendance of the dty's first Chinese American mayor, and continued positive media attention, were among these developments. In addition, an economic climate that preduded costly new initiatives made an ordinance showing a commitment to "doing something" for the hardest-hit workers politically attractive in this dty known for its progressive poKtics. The continued successful base-building of PWA and CPA also was critical to the measure's unanimous passage and signing into law in spring 2011.

Ftuther taking advantage of the policy win- dow that remained open during this period, a second piece of legislation was introduced and quickly passed that created a wage theft implementation task force. Comprised of members of the Board of Supervisors, the PWA and CPA, the health department, and other stakeholder groups, the resultant task force met monthly in City Hall and worked to help ensure that the wage theft ordinance was enforced.

Policy Enforcement

Passage of a second bul to promote en- forcement was critical in making stire that the original ordinance "had teeth" and could result in positive change. The Anti-Wage Theft Task Force saw several vidories in its first year. Key among them was the biggest anti-wage theft vidory in the dty's history, with more than $525 000 recovered from a single pastry shop in Chinatown. This development, too, was prominentiy featured in the mainstream and ethnic press^' and on television news stations and the Internet It was followed by a second major vidory in a settlement for Filipino workers, with others following.

Capitalizing on these events, the PWA and CPA arranged a spedal event inside City Hall in the spring of 2013, at which workers from a wide range of groups, such as Mujeres Unidas y Activas (United and Engaged Women) and Young Workers United, as well as CPA, were honored by the dty for their work, and worker representatives in turn thanked the City

June 2014, Vol 104, No. 6 | American Journal of Public Health Minkler et al. | Peer Reviewed 1 Framing Health Matters | 1015

FRAMING HEALTH MAÏÏERS

c3

g II 3 3

s(S "5 e '.B (B

a.3 u ë B (B S

eft

|5

^ M

a i

i II

Re ste

er s

(n

g -g

2 m

LO Oí Lf)

CO CN <O CO t o i r i CO - ^

LTJ t o en CO OÖ i r i od CO

r o CO

g g s

îîî S

S as

LO

CD

Si -e

E

tI *. I ^ -s ^ ce:

S.

I I

CO uTî

OJ .—

i -B

8

se

â i

e i s

s ä -

s, s.

Q . CD

Ä i ^ 35 g

"^

'S

.11 2 ^

.i g 1 .i

i 1

i s

S - -^ -2 a S2

^ o o .= o .£

2 ii I Ä o « o

1016 I Framing Health Matters [ Peer Reviewed | Minkler et al. American Joumai of Pubiic Health | June 2014, Vol 104, No. 6

FRAMING HEALTH MÄHERS

CO o CO o o r-̂ iri cd o uS

oj CN ro r-- o CO r-̂ •Ï-Î

to CO en •<-« oo

CO 1-H o CO

T-H t o oq oq

I

i

III S.

f

to

I

.g

.g

ra I E Oj S;

- 1 1

Supervisors and enforcement officers for their support. Finally, and responding to the new law promoting enforcement, the SFDPH took steps such as collaborating with the local Office of Labor Standards Enforcement to temporarily revoke health permits for businesses with un- resolved labor law violations. This action has already resulted in the payment of thousands of dollars owed to workers in back wages, in one case prompting action within 4 months after nonpayment for more than 4 years.^^

REVIEW

We have attempted to draw attention to, and augment, a small but gro'wing body of literature suggesting that wage theft is a serious and largely overlooked problem with imporiant public health implications. We further demon- strate the utility of a study combining individ- ual and restaurant-level data to provide a mul- tilevel look at the problem of wage theft, induding findings with utüity for health de- partments and other enforcement agendes.

Our research, however, suffered several limitations. First, the survey was a cross- sectional study based on a convenience sample of 405 workers. Although the sample was a suffident size to enable the statistical calcu- lations tmdertaken, the findings are not gener- alizable. Because of the need for extreme caution in inviting workers, and particularly immigrant workers, to participate in such a study without putting themselves at risk for employer retaliation, however, convenience sampling appeared the most ethical and realistic choice. Second, and relatedly, on a few ques- tions regarding issues such as what happened to workers' tip money, a number of partidpants responded "don't know." Although this may have resulted from a genuine lack of knowledge, worker partners suggested that it may also have reflected reluctance to give a response that made one's employer look bad, even in an anonymous survey. If this was the case, it may suggest that our findings on these questions likely erred in a more conservative direction.

The results of our survey largely supported key findings (e.g., lower hourly wages among dishwashers, table cleaners, and female workers, plus very high poverty rates) from the few other studies of low-wage restaurant workers conducted to date.'^'°'"'^'"^^ A 2005

survey of 530 restaurant workers in New York City thus showed that 59% had been denied overtime pay, compared with 6 5 % of those in our study who worked overtime.^' Our study, however, provided more detailed information on wages and relative as well as absolute poverty, and more in-depth calculations of estimated wages lost per year attributable to a number of different forms of wage theft. Finally, the survey data provided policy- relevant cost estimates (e.g., suggesting that Chinatown restaurant workers lost close to $10 500 000 in 2008 alone because of wage theft). However, these figures do not include lost tax revenue or local economic stimulus (i.e., the costs to the dty, local businesses, and sodety at large), and further research is needed in this area.

We also examined a little-studied aspect of wage theft, lack of enforcement of mandated paid sick leave, which is of spedal relevance to public health. Several empirical studies have shown a relationship between lack of paid sick leave and adverse health outcomes.̂ *" '̂̂ Furthermore, failure to provide paid sick may lead to

increased reliance on unemployment insurance and social assistance programs resulting in an indirect subsidy to employers engaging in "low road" practices and fewer such public resources available to all those in need.^"P^'

Yet an important distinction exists between lack of paid sick leave, and lack of enforcement of mandated paid sick leave, with the latter more accurately included as a component of wage theft. Not only do few jurisdictions cur- rently require paid sick leave, but also state- level measures under re'view increasingly are being pressured to propose weak coverage, either forbidding the passage of stronger mu- nicipal polides or negating those that exist.^'' In this cKmate, it is important to conduct studies in jurisdictions that have mandated paid sick leave on its health and economic impacts. Although technically aU San Frandsco workers have this benefit, for example, employers of low-wage immigrant workers, particularly in areas like Chinatown, are kno'wn to frequently ignore this dty mandate. Furthermore, as our study demonstrated, it may also be important to mount educational efforts explaining, particu- larly to covered immigrant and other low-wage workers, the actual meaning of paid sick leave.

June 2014, Vol 104, No. 6 1 American Journal of Public Health Minkler et al. [ Peer Reviewed | Framing Heaith Matters | 1017

FRAMING HEALTH MÄHERS

We focused much of this article on the dissemination-to-action components of our partnership's work, and spedfically its contri- butions to helping bring about one of the nation's first mimidpal wage theft ordinances. Yet much caution must be used in such as- sessments, because as Sterman^"* and others*'"^^ point out, numerous adors and contextual fadors are at play in affecting policy, and teasing apart the roles of particular groups or variables is often impossible. As discussed previously, fadors such as the election of the dty's first Chinese American mayor and a pro- gressive board of supervisors wishing to take a low-cost action during a major recession that would show concern about wage theft were among the fadors that contributed to policy wins. Their roles should not be minimized.

We designed the literature review for this article to highlight key recent articles on wage theft. A thorough and systematic review is needed, however, paying spedal attention to potential pathways between various forms of wage theft and public health outcomes.

It is also important to consider whether the research approach and community and policy outcomes described in the Chinatown study could likely be replicated elsewhere. In terms of the orientation to research, several strong examples exist of CBPR with workers, induding indigenous and Latino agricultural workers in Oregon,̂ ® hotel room cleaners in San Francisco and Las Vegas, Nevada,̂ '̂̂ ® and custodial workers in Iowa®^ The Restaurant Opportu- nities Center also has been successful in orga- nizing restaurant workers of diverse races and ethnicities for living wages and against wage theft in dozens of dties across the country.®''̂ '̂ '

From the perspective of the "politics of the possible," San Frandsco, with the nation's highest minimum wage rate ($ 10.55), as well as some of the only local employer mandates for health insurance coverage and paid sick leave, was indeed well ahead of the curve in being ripe for such data-supported action. However, other states and dties have demonstrated similar commitments to labor rights, and there have been recent notable labor movement successes in securing and enforcing labor rights nationally. Several dties, states, and territories have wages substantially above the federal minimum wage of $7.25, among them Santa Fe, New Mexico ($10.51); Washington State

($9.19); and Caüfomia, which recently in- creased its minimum wage to $9.00 in 2014 and will increase it to $10.00 in 2016.

More state and local governments also are moving toward greater enforcement of existing labor laws, with New York collecting $28.8 million in 2009 alone in unpaid wages for nearly 18 000 workers (http;//www.cpasforg). Seattle, Washington; Portland, Oregon; New York City; Washington, DC; and Connecticut have all now passed sick leave laws, and the Affordable Care Ad includes an employer health care mandate. Federal rules also re- cently have applied wage and overtime pro- tections to home care workers®" with California passing, in 2013, a domestic workers' bill of rights, including the right to overtime pay.^' Regardless of this progress, however, enforce- ment of a legal mandate, as noted earlier, is very different from the legal mandate itself. We believe that public support for enforcement of an adopted legal minimum wage, particularly when framed as "wage theft," is likely to be substantial regardless of the level of wage.

Policy efforts to prevent wage theft are distinct from efforts to achieve other health- supportive labor rights, such as sick leave and health insurance benefits. State and federal laws already make nonpayment of wages ille- gal. The local efforts in San Francisco, including polity and legislation, were taken as actions to address gaps in local, state, and federal assur- ance of wage laws. If federal and state regula- tion of the existing laws were adequate and effective (e.g., including proactive and targeted monitoring, and suffident staff and enforce- ment), additional local wage theft efforts might not be needed. Although local efforts could and should be replicated, the ultimate goal is achieving compliance with existing wage laws.

As Baron et al. point out, "many seeking to improve health equity view employment as part of the solution, and are reluctant to acknowledge that it may also be part of the problem."®^*P'̂ ' As this artide suggests, wage theft, in part through its substantial contribu- tions to low-income status, is a health equity problem of major proportions. Epidemiology, health economics, medical anthropology, and other disdplines all have a role to play in increasing our understanding of wage theft and its implications for the public's health. In a simüar way, as Bhatia et al.̂ ^ suggest, local

and state public health departments can play a number of roles to prevent wage theft, as part of their core public health functions, including

• Monitoring information about various forms of wage theft in population health assess- ments;

• Analyzing relationships between health outcomes and wage theft, including lack of enforcement of mandated paid sick leave and receipt of less than minimum wage;

• Educating workers and employers about wage theft and what they can do about it;

• Conducting health impad assessments of proposed health-protective labor poUdes, induding those related to wage theft;

• Monitoring compliance with labor laws re- lated to wage theft in routine agency activ- ities, referring potential violations to labor enforcement agencies; and

• Engaging in partidpatoiy research with employees to identity and improve working conditions, induding fair wage practices.^^

Broadening the role of health departments in a time of badly constrained resources is diffi- cult, and may sometimes lead to push back,̂ ^ as when the New York City Department of Public Health argued against new legislation that would have broadened its mandate to include wage and hourly violations in decisions concerning permitting renewal.^^ Although the New York City Department of Public Health argued that wages and hourly violations lacked public health significance®^ we have attempted to demonstrate that by causing or exacerbating low income or income insecurity, wage theft is indeed a public health problem, and one with which health departments and other public health organizations must increasingly be concerned.

To our knowledge, community advocacy against wage theft has engaged with only 2 local health departments—San Francisco and New York City. In San Francisco, the depart- ment worked coUaboratively with advocates to explore tising their regulatory capadty to sanction wage violators. In New York City, as noted previously, the health department op- posed a mandate introduced at dty coundl. Differences in the responsiveness of these agencies may be attributable to differences in local political fadors or to differences in the process—collaboration versus mandate. How- ever, communily organizations have successiuBy

1018 ¡ Framing Health Matters | Peer Reviewed | Minkler et al. American Journai of Public Health 1 June 2014, Voi 104, No. 6

FRAMING HEALTH MAÏÏERS

mobilized public health support for paid sick leave laws in other dties and their partnership efforts are instructive. Early community demand for health department action and the develop- ment of relationships among community wage theft advocates and department staff may have enabled or influenced a more supportive position in New York City. We are not aware of any negative community impacts of health agency support of wage law enforcement

With increasing support from the commu- nities they serve, broadening the mandate of health departments in particular could help us better understand the extent and nattire of wage theft and related health problems, while also ofFering potential new avenues for dissem- ination and evidence-based action. As national momentum grows to build local and state health department capadty to address the sodal de- terminants of health, wage theft is one concrete avenue to address the well-documented rela- tionship between income and health.

CONCLUSIONS

Affecting, as it does, millions of workers, their families, and communities, and dispro- portionately affecting low-wage workers, wage theft is an epidemic in the United States. An increasing body of evidence from government data sets, as well as studies in economics, labor, business, sociology, and industrial medicine, has underscored the importance of recognizing and dealing with this "defining feature" of today's job market.^ Yet, despite public health's strong emphasis on the link between low in- come and poor health outcomes, our profes- sion has been slower to recognize and address wage theft as an epidemic requiring our con- certed attention.

The Chinatovm restaurant worker study illustrates how community-based organizations and workers, health departments, and aca- demic partners can coUaboratively study and help address wage theft on the munidpal level. But, as with other epidemics, wage theft and its public health implications can only be addressed if we have "all hands on deck." According to Kingdon's' ' model, this means collecting the data needed to illustrate that wage theft is a significant problem, identifying and proposing promising strategies, and work- ing to get new polides adopted and enforced.

With such a commitment, the field of public health can become a major player in studying and addressing what is arguably among the most neglected epidemics of our times. •

About the Authors Meredith Minkler is with the Division of Community Health and Human Development, School of Public Health, Uni- versity of California, Berkeley. Alicia L. Salvatore is with the Stanford University School of Medicine, Stanford, CA. Charlotte Chang is with the Labor Occupational Health Program, School of Public Health, University of California, Berkeley. At the time of the study, Pam Tau Lee was also with the Labor Occupational Health Program, and Megan Gaydos with the Program on Health, Equity, and Sustain- ability, San Francisco Department of Public Health, San Francisco, CA. Shaw San Liu and Alex Tom are with the Chinese Progressive Association, San Francisco. Rajiv Bhatia was with the San Francisco Department of Public Health, and Niklas Krause was with the School of Medicine, University of California, San Francisco.

Correspondence should be sent to Meredith Minkler, DrPH, MPH, School of Public Health, University of California, Berkeley, CA 94720-7360 (e-mail: mink@ berkeley.edu). Reprints can be ordered at http://www.ajph. org by clicking the "Reprints" link.

This article was accepted on November 15, 2013.

Contributors All authors served as members of the CBPR partnership, actively participated in all phases of the study, and contributed to and commented on the article. Spedfi- cally, M. Minkler served as principal investigator, super- vised the study, and took the lead in writing this article. A. Salvatore helped conceptualize and design the study, analyzed data, and participated in the writing of this article. C. Chang conducted the evaluation of the partnership and participated in writing. P. T. Lee served as the study coordinator and participated in dissemina- tion and policy action. S. Liu and A. Tom served as the lead community partners, and led dissemination and policy action. M. Gaydos and R. Bhatia served as the lead health department partners at the time of the study, analyzing the restaurant-level data, and conducting dissemination and policy action at the health department and with other regulatory bodies. Both also made major contributions to the writing and editing of this article. N. Krause helped conceptualize the project, led the survey development, analyzed wage and wage theft data, and helped substantially in the writing of this article.

Acknowledgments This study was funded by grants from the National Institute for Occupational Safety and Health/Centers for Disease Control and Prevention (R219081), and the California Endowment, with additional support from the Occupational Health Internship Program.

The authors acknowledge the partners and partici- pants in the Chinatown Restaurant Worker Health and Safety Project, and pariicularly worker partners Rong Wen Lan, Li Li Shuang, Li Zhen He, Hu Li Nong, Michelle Xiong, Zhu Bing Shu, Gan Lin, Christy Wu, Huang Pei Yu, and organizer Feiyi Chen. Special thanks also are due Robin Baker and Alvaro Morales for their contributions.

Note. The contents of this article are solely the responsibility of the authors and do not necessarily represent the official views of the funding agendes.

Human Participant Projection This study was approved by the Committee for the Protection of Human Subjects at the University of California, Berkeley.

References 1. Braveman P, Egerter S, Williams DR. The sodal determinants of health: coming of age. Annu Rev Public Health. 2011,32:38l-39a.

2. Iton A. Tackling the root causes of health disparities through community capacity building. In: Hofrichter R, Bhatia R, eds. Tackling Health Inequities Through Public Health Practice. New York, NY: Oxford University Press; 2 0 1 0 : 3 7 0 - 4 0 3 .

3. Krieger N. Epidemiology and the People's Health: Theory and Context. New York, NY: Oxford University Press; 2 0 1 1 .

4. Marmot M, Friel S, Bell R, Houweling TAJ, Taylor S, for the Commission on Sodal Determinants of Health. Closing the gap in a generation: health equity through action on the sodal determinants of health. Lancet. 2008;372(9650):1661-1669.

5. Lipscomb HJ, Loomis D, McDonald MA, Argue RA, Wing S. A conceptual model of work and health dispar- ities in the United States. Int] Health Serv. 2006;36 (l):25-50.

6. Leigh JP, Du J. Are low wages risk factors for hypertension. Eur f Public Health. 2 0 1 2 ; 2 2 ( 6 ) : 8 5 4 - 859.

7. National Employment Law Project. Winning wage justice: a summary of research on wage and hour violations in the United States. Avañable at: http://nelp. 3cdn.net/509a6e8alb8f2a64f0 J'2m6bhlf6.pdf Accessed February 20, 2012.

8. Bobo KA. Wage Theft in America: Why Millions of Working Americans Are Not Getting Paid-And What We Can Do About It. New York, NY: The New Press; 2 0 1 1 .

9. Bernhardt A, Milkman R, Theodore N, et al. Broken Laws, Unprotected Workers: Violations of Employment and Labor Laws in America's Cities. New York, NY: Center for Urban Economic Development, UlC National Employ- ment Law Project, UCLA Institute for Research on Labor and Employment; 2009.

10. Winning Wagefustice: An Advocate's Guide to State and City Policies to Fight Wage Theft. New York, NY: National Employment Law Project; 2 0 1 1 .

11. Kingdon JW. Agendas, Alternatives, and Public Polides. 2nd ed. New York, NY: Addison-Wesley Edu- cational Publishers Inc; 1995.

12. Labor Force Statistics From the Current Population Survey: Characteristics of Minimum Wage Workers, 2010. Washington, DC: Bureau of Labor Statistics, US De- partment of Labor; 2010.

13. Labor force characteristics by race and ethnidty, 2010. Washington, DC: Bureau of Labor Statistics, US Department of Labor; 2 0 1 1 . Report 1032.

14. Lashuay N, Harrison R. Barriers to Occupational Health Services for Low-Wage Workers in California. San Frandsco, CA: Commission on Health and Safety and Workers' Compensation, California Department of In- dustrial Relations; 2006.

15. Siqueira CE, Gaydos M, Monforton C, et al. Effects of sodal, economic, and labor polides on occupational health disparities. AmJ IndMed. 2013;Epub ahead of print.

June 2014, Vol 104, No. 6 | American Journal of Public Health Minkler et al. | Peer Reviewed | Framing Health Matters | 1019

FRAMING HEALTH MAÏÏERS

16. Cleaning up the carwash industry; empowering workers and protecting communities. A report by the Carwash Workers Organizing Committee of the United Steelworkers. Los Angeles, CA: United Steelworkers; 2008.

17. Behind the Kitchen Door: A Multi-Site Study of the Restaurant Industry. New York, NY: Restaurant Oppor- tunities Center United; 2 0 1 1 .

18. Economic Report of the President. 2 0 1 1 . Available at: http ://www.whitehouse.gov/admini5tration/eop/ cea/economic-report-of-the-President/ZOl 1. Accessed February 2 0 , 2 0 1 2 .

19. US Bureau of the Census. Statistical Abstract of the United States. 2 0 1 1 . Available at: http://www.census. gov/compendia/statab. Accessed February 20, 2012.

20. Brown HS III, Perez A, Yamell LM, et al. Diabetes and employment productivity: does diabetes manage- ment matter? Diabetes. 2 0 n ; 1 7 : 5 6 9 - 5 7 6 .

2 1 . Yach D, Stuckler D, Brownell KD. Epidemiologie and economic consequences of the global epidemics of obesity and diabetes. Nat Med. 2006;12(l):62-66.

22. Gerend MA, Pai M. Sodal determinants of Black- White disparities in breast cancer mortality: a review. Cancer Epidemiol Biomarkers Prev. 2008;17(l 1):2913- 2923.

23. Interfaith Worker Justice Study on Wage Theft. Cited by Bobo KA. Wage Theft in America: Why Millions of Working Americans Are Not Getting Paid—And What We Can DoylAowi A New York, NY: The New Press; 2011,

24. Asfaw A, Pana-Ciyan R, Rosa R. Paid sick leave and nonfatal occupational injuries. Am J Public Health. 2012;102(9):e59-e64.

25. Kumar S, Quinn SC, Kim KH, Daniel LH, Freimuth VS. The impact of workplace policies and other social factors on self-reported influenza-like illness incidence during the 2009 H l N l pandemic. AmJ Public Health. 2012;102(l):134-140.

26. Peipins LA, Soman A, Berkowiti Z. The lack of paid sick leave as a barrier to cancer screening and medical care-seeking: results from the National Health Interview Survey. BMC Public Health. 2012;12(l):520.

27. GambreU S, Russen LLP, Hill PJ, Clarke MW. Mandated paid sick leave now and in the future. Lexology. March 22, 2 0 1 3 . Available at: http://www.lexology. com/library/detail.aspx?g=a7194482-52ff-45cb-9b68- le70c6f94077. Accessed April 1, 2013.

28. Minkler M, Lee PT, Tom A et al. Using commtmity-based participatory research to design and initiate a study on immigrant worker health and safety in San Francisco's Chinatown restaurants. AmJ Ind Med. 2010;53(4):361-371.

29. Green L, George M, Daniel M, et al. Study of Participatory Research in Health Promotion. Ottawa, ON: The Royal Society of Canada; 1994.

30. Sustainable Communities Index. Healthy Cities, Healthy People. San Francisco Demographic: D.3 per capita and household income. Available at: http://www. sustainablecommunitiesindex.org/city_indicators/view/ 63. Accessed June 10, 2013.

3 1 . Behind the Kitchen Door: Pervasive Inequality in New York City's Thriving Restaurant Industiy. New York, NY: Restaurant Opportunities Center of New York; 2005.

32. Woo SC, Krause N. Working Conditions and Health of Hotel Kitchen Workers in San Francisco. San Frandsco, CA: US Bureau of Labor Statistics; 2003.

33. Webster T. Occupational hazards in eating and drinking places. Compensation Working Conditions. 2001 ;2001 (Summer):27-33.

34. Horowitz CR, Robinson M, Seifer S. Community- based partidpatory research from the margin to the mainstream: are researchers prepared? Circulation. 2009;119(10):2633-2642.

35. O'Fallon LR, Deany A Community-based partidpatoiy research as a tool to advance enviromnental health sdences. Environ Health PerspecL 2002;l 10(suppl 2):155-159.

36. Israel BA, Eng E, Schulz AJ, Parker E. An introduc- tion to methods in community based partidpatory re- search for health. In: Methods for Community Based Partiapatory Research for Health. 2nd ed. San Francisco, CA: Jossey-Bass; 2012: 4 - 4 2 .

37. Chang C, Salvatore AL, Lee PT, et al. Adapting to context in community-based partidpatory research: "partidpatoiy starting points" in a Chinese immigrant worker community. AmJ Comm Psych. 2013;51(3-4):480-491.

38. Salvatore AL, Krause N. Health and Working Conditions of Restaurant Workers in San Francisco's Chinatown: Report of Survey Findings. Berkeley and San Francisco, CA: UC Berkeley and UC San Francisco; 2010.

39. Chang C, Salvatore A, Lee PT, Liu SS, Minkler M. Popular education, partidpatory research, and commu- nity organizing with immigrant restaurant workers in San Francisco's Chinatown: a case study. In: Minkler M, ed. Community Organizing and Community Building for Health and Welfare. 3rd ed. Piscataway, NJ: Rutgers University Press; 2012.

40. US Bureau of Labor Statistics. Full-time private industry workers: mean and median hourly, weekly, and annual earnings and mean weekly and annual hours December 2007-January 2008. National Compensation Survey, Table 4. Available at: http://www.bls.gov/ncs/ ncswage2008.htm. Accessed March 10, 2014.

4 1 . 2 0 0 8 HHS poverty guidelines for the 4 8 contiguous states and the District of Columbia. Federal Register, 2008;73(15):3971-3972.

42. Drago R, Lovell V. San Francisco's Paid Sick Leave Ordinance: Outcomes for Employers and Employees. Washington, DC: Institute for Women's Policy Research; 2 0 1 1 .

4 3 . Gaydos HBhaMaR, Morales A, et al. Promoting health equity and safety in San Francisco's Chinatown restaurants: findings and lessons learned from a pilot observational survey. Public Health Rep. 201 l;126(suppl 3):62-69.

44. Minkler M, Salvatore A. Partidpatory approaches for study design and analysis in dissemination and imple- mentation research. In: Bronson R, Coldditz G, Proctor E, eds. Dissemination and Implementation Research in Health: Translating Science to Practice. New York, NY: Oxford University Press; 2012.

45. Chinese Progressive Assodation. Check, Please! Health and Working Conditions in San Francisco China- town Restaurants. 2010. Available at: http://www.cpasf org. Accessed February 20, 2012.

46. Abate T. "Wage theft" alleged at Chinatown restaurants. San Francisco Ommick September 17, 2010: Dl, D4.

47. Staples L. Analyze, strategize, and catalyze: issues and strategies, hi: Staples L, ed. Roots to Power: A Manual of Grassroots Organizing. 2nd ed. Westport, CT: Praeger; 2 0 0 4 : 9 9 - 1 3 7 .

48. Coté J. Low-wage workers cheated. San Francisco Chronicle. ]u\y 18, 2 0 1 1 : A l , A7.

49. Boots, S, Martinson, K, Danziger, A. Employers' perspectives on San Francisco's paid sick leave policy. Washington, DC: The Urban Institute, Research of Re- cord; 2009.

50. Themba Nixon M, Minkler M, Freudenberg N. The role of CBPR in policy advocacy. In: Minkler M, Wallerstein N, eds. Community-Based Participatory Re- search for Health: From Process to Outcomes. 2nd ed. San Frandsco, GA: Jossey-Bass; 2 0 0 8 : 3 0 7 - 3 2 2 .

5 1 . Coté J. Wage theft payment sets S.F. record. San Francisco Chronicle. February 13, 2 0 1 3 : E l .

52. Bhatia R, Gaydos M, Yu K, Weintraub J. Protecting labor rights: roles for public health. Public Health Rep. 20I3;128(suppI3):39-47.

53. Zomick G. The paid sick leave battle widens in the states. Nation. March 8 , 2 0 1 3 . Available at: http://www. thenation.com/blog/173265/paid-sick-leave-battie- widens-states#. Accessed February 20, 2012.

54. Sterman JD. Learning from evidence in a complex world. AmJ Public Health. 2006;96(3):505-514.

55. Minkler M. Linking sdence and policy through community-based partidpatory research to eliminate health disparities. AmJ Public Health 2010;100(suppl 1):S81-S87.

56. Farquhar S, Shadbeh N, Samples J, Ventura S, Goff N. Occupational conditions and well-being of indigenous farmv/oñier^, AmJ Public Health. 2008;98(l 1):1956-1959.

57. Lee PT, Krause N, Goetchius C, Agriesti JM, Baker R. Partidpatory action research with hotel room deaners: from collaborative study to the bargaining table. In: Minkler M, Wallerstein N, eds. Community-Based Partic- ipatory Research for Health: From Process to Outcomes. 2nd ed. San Frandsco, CA: Jossey-Bass; 2 0 0 8 : 3 3 5 - 3 5 3 .

58. Krause N, Scherzer T, Rugulies R. Physical work- load, work intensification, and prevalence of pain in low wage workers: results from a partidpatory research project with hotel room cleaners in Las Vegas. Am find Med. 2005;48(5):326-337.

59. Flum MR, Siquedra CE, DeCaro A, Redway S. Photovoice in the workplace: a partidpatory method to give voice to workers to identify health and safety hazards and promote workplace change—a study of university custodians. AmJ Ind Med. 2010;53 (11):1150-1158.

60. Greenhouse S. US to indude home care aides in wage and overtime law. New York Times. September 17, 2013. Available at: http://vTOW.nytimes.com/2013/09/ 18/business/us-to-indude-home-care-workers-in-wage- and-overtime-law.html. Accessed February 20, 2012.

6 1 . Miles K. California minimum wage increase to $ 10 and hour passes both hotises. Governor Brovm expected to sign. Hufßngton Post. September 13, 2 0 1 3 . Available at: http://www.huffingtonpost.com/2013/09/12/ califomia-minimum-wage-increase_nJ908620.html. Accessed February 20, 2012.

62. Baron SL, Beard S, Davis LK, et al. Promoting integrated approaches to reducing health inequities among low-income workers: applying a sodal ecological framework. AmJ Ind Med 2013;Epub ahead of print

63. New York City Council, Committee on Health. Hearing transcript March 3 1 , 2008, for Int 0569-2007. Available at: http://legistar.coundljiyc.gov/ LegislationDetail.aspx?ID=451330&GUID=A90C8C 12- 5590-49C0-8672-FC97812EA690&tnqh_x26; Options=&Search=. Accessed February 20, 2012.

1020 I Framing Health Matters | Peer Reviewed | Minkler et al. American Journal ot Public Health | June 2014, Vol 104, No. 6

Copyright of American Journal of Public Health is the property of American Public Health Association and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.