FOR A-PLUS WRITER ONLY

56 ProfessionalSafety MAY 2016 www.asse.org

T he construction industry continues to ex- perience a high number of workplace inju- ries and fatalities as compared to other U.S.

industrial sectors. Although this number has been declining over the past 20 years, the rate of decrease has been slowing, and is nearly stagnant in recent years (ILO, 2003). As an industry, construction has averaged 1,010 fatalities per year, indicating that much improvement is still needed to achieve zero injuries, illnesses and fatalities (BLS, 2013a). One such improvement can be found in the collection and measurement of safety data.

Historically, the construction industry has defined safety performance through the mea- surement and assessment of lagging indicators including injuries, illnesses and fatalities. These lagging indicators are required by OSHA to assess the state of construction safety (BLS, 2013a). One major limitation of assessing safety performance using lagging indicators is that incidents must oc- cur before hazards or unsafe behavior can be iden- tified and mitigated.

Leading indicators are an alternative form of safety metrics that proactively assess safety per- formance by gauging processes, activities and con- ditions that define performance and can predict future results (Hinze, Thurman & Wehle, 2013). One such leading indicator is a near-hit, defined as an incident in which no property damage or per- sonal injury occur, but could have occurred given a slight shift in time or position (BLS, 2013a). The major advantage of measuring leading indicators such as near-hits is that data can be collected and analyzed without requiring an injury to occur.

This article presents research products in the development, deployment and effectiveness of using a near-hit management program on con- struction sites. The authors gathered the informa- tion through personal experience, formal research in the Construction Industry Institutes Research Team 301: Using Near Misses to Enhance Safety Performance, and through secondary research and literature review. The goals of this article are to present the near-hit management program and

Eric Marks, Ph.D., P.E., is an assistant professor in the Department of Civil, Construction and Environmental Engineering at the University of Alabama. His research focuses on innovation and automation in construc- tion safety, including hazard mitigation strategies and real-time data collection. Marks is a professional member of ASSE’s Alabama Chapter.

Ibukun G. Awolusi is a Ph.D. student in the Department of Civil, Con- struction and Environmental Engineering at The University of Alabama. He holds an M.Sc. in Construction Management from the University of Lagos.

He has both industrial and teaching experience in construction and occupa- tional safety, and he is actively involved in several research projects related to construction safety and technology/innovation in construction.

Brian McKay, M.P.H., CSP, CIH, is director of quality, health, safety and the environment for Fairweather LLC. His research inter- ests include human error modeling, near-hit reporting and behavioral economics. McKay is a professional member of ASSE’s Alaska Chap- ter and a member of the Society’s Construction Practice Specialty.

Near-Hit Reporting

Reducing Construction Industry Injuries

Program Development Peer-Reviewed

By Eric Marks, Ibukun G. Awolusi and Brian McKay

056_062_F2Mar_0516.indd 56 4/22/16 1:22 PM

www.asse.org MAY 2016 ProfessionalSafety 57

demonstrate its quantitative effect and proof of ef- fectiveness when applied to a multibillion dollar construction project, to encourage the use of this methodology in the field.

Construction Incident Statistics In the U.S., construction companies are required

to report all fatalities, injuries and illnesses that oc- cur during or as a result of the work environment (OSHA, 2011). OSHA categorizes reported inci- dents as 1) occupational fatality, 2) nonfatal injury or 3) nonfatal illness, and further categorizes them as to severity: OSHA recordable injuries and lost time/days away from work cases.

U.S. Bureau of Labor Statistics (BLS) data show 117 recordable incident cases for every 10,000 workers in the U.S. in which the injury or illness was nonfatal but required days away from work (BLS, 2012). Construction workers experienced 179,100 nonfatal injuries in 2012 (6% of cases when compared to the total nonfatal injuries experience by the U.S. private sector that year), a decrease compared to the 184,700 injuries reported by the industry in 2011 (8.6% of cases) and 3,153,701 in 1992 through 2010 (10.6% of cases) (BLS, 2013b).

Leading Indicators As noted, construction companies are required

to document work-related incidents (OSHA, 2013). These metrics, termed lagging indicators, cannot reflect whether a hazard, the event sever- ity or causation has been mitigated (Flin, Mearns, O’Connor, et al., 2000; Lindsay, 1992). According

to Hallowell, Hinze, Baud, et al. (2013), leading indicators are measures of processes, activities and conditions that define performance and that can predict future results. Unmitigated high-risk situ- ations, including near-hits, will result in a serious or fatal injury if allowed to continu- ally exist (Krause, Groover & Martin, 2010).

Linear causation mod- els (e.g., domino theory, loss causation models) sug- gest that incidents are the end result of a sequence of events and provide a sound motivation to collect and analyze near-hit data. Ear- lier researchers have also found that most serious inju- ries can be successfully pre- vented (Hecker, Gambatese & Weinstein, 2005; Hinze, 2002; Hinze & Wilson, 2000; Huang & Hinze, 2006).

Near-Hit Reporting Across Industries Near-hit reporting has been widely used in vari-

ous industries throughout the world for some time. A company in the offshore drilling business real- ized exceptional decreases in lost-time incident

Results of a construction site

case study for the implementation of the created

program indicate that near-

hit reporting and analysis can improve

the safety performance of workers on construction

sites.

IN BRIEF •The construction industry continues to rank as one of the most hazardous work environments, experiencing a high num- ber of workplace injuries and fatalities. •Safety performance improvement is needed to achieve zero injuries, illnesses and fatalities on construc- tion sites. One systematic method of achieving this improvement is through the collection and analysis of safety data such as near-hits. •This article highlights best practices for collecting and analyzing near-hit information. A near-hit management program for assessing collected data is created so that lessons learned from re- ported events can be applied to mitigate future hazards on construction sites. •Results of a construction site case study of the implementation of the created program indicate that near-hit reporting and analysis can improve the safety performance of workers on construction sites.

© IS

t O

c K

P H

O t

O .c

O M

/R a

In E

R P

L E

n D

L

056_062_F2Mar_0516.indd 57 4/22/16 1:22 PM

58 ProfessionalSafety MAY 2016 www.asse.org

rates when it implemented a near-hit program; the company found that a reporting rate of 0.5 near- hits/person/year correlated with a 75% reduction in lost-time injury rates (Phimister, Oktem, Klein- dorfer, et al., 2003).

The process of collecting and analyzing near-hit data has been studied in the chemical process in- dustry (van der Schaaf & Kanse, 2004). The study also investigated barriers to and human behavior associated with reporting near-hits. Within the chemical processing industry, the U.S. Nuclear Regulatory Commission has collected and re- viewed near-hit reports for nuclear reactors since 2000 (Donovan, 2011).

The aviation industry also benefits from near-hit reporting practices. Aircraft proximity hazard (air- prox) is an aviation industry term for a near-hit. An airprox is a situation in which the distance between aircraft, as well as their relative positions and speed, have been such that the safety of the aircraft involved was compromised (CAA, 2013). Safety recommendations are focused at limiting the risk of recurrence of a specific airprox event. The pri- mary objective is to improve flight safety with re- gard to identified hazards and lessons learned from near-hit occurrences.

The firefighter near-hit reporting system is an- other distinct industry adopting near-hits as an op- portunity to learn. The near-hit database (http:// fire.nationalnearmiss.org/reports) is managed by the International Association of Fire Chiefs (IAFC) and funded by Federal Emergency Man- agement Agency’s Assistance to Firefighters grant. This anonymous reporting database is designed to accept near-hit reports from fire departments throughout the U.S. The database is open for re- view, and shares lessons learned and experiences from the firefighting community. A similar database is also available for law enforcement officials (http:// leo.nationalnearmiss.org/browse-reports).

A study by Callum, Kaplan, Merkley, et al. (2001), on near-hit reporting in the medical field concerning transfusion medicine collected data on human errors and near-hits at a blood bank. Three of the most concerning events were 1) samples collected from the wrong patient; 2) mislabeled samples; and 3) requests for blood for the wrong patient (Callum, et al., 2001). Similar studies were conducted in nursing home environ- ments (Wagner, Capezuti & Ouslander, 2006).

The construction industry has been slower to adopt near-hit reporting when compared to other industries in the U.S. private sector (Cambraia, Saurin & Formoso, 2010), with some notable ex- ceptions. For example, a large U.S. manufactur- ing company uses a system called ARTTS-NMA: Autonomous Real-Time Tracking System of Near Miss Accidents on construction sites (Caterpillar, 2013). This system uses ultrasonic technology for outdoor and indoor real-time location tracking, sensors for environmental surveillance, radio fre- quency identification for access control and worker information, and wireless sensor networks for data transmission (Wu, Huanjia, Chew, et al., 2010). The goal is to automatically identify a specific type of hazard as a near-hit event and alert safety per- sonnel before a similar situation occurs.

In summary, BLS maintains a database for lag- ging indicator data including workplace fatali- ties, injuries and illnesses, but does not require near-hit reporting. Several industrial sectors col- lect and analyze near-hit data for potential safety improvement. Many of these industries maintain an industry-wide near-hit reporting database so that other industry personnel can learn from each other’s near-hit information. The construction in- dustry has been slower to adopt near-hit reporting for reasons such as fear of retaliation, anticipated barriers, and miscommunication that the more near-hits that are reported, the poorer safety per- formance can be expected on a project.

Near-Hit Data Collection & Analysis Framework This article presents a high-level model for a

near-hit management program as the basic meth- odology for site safety managers and construction management personnel to collect, analyze and use safety data effectively. This framework implements a management system for near-hit data and can be a vital component in the data flow within a near-hit reporting program. Figure 1 presents the five steps for this framework of transitioning near-hit data to information and ultimately knowledge for dissemi- nation. The five steps are further described here.

Step 1: Identification The first step occurs when construction site per-

sonnel recognize an unsafe event or set of conditions on a construction site. Employees should be trained to identify near-hits and how they differ from lag-

Figure 1

Framework for Near-Hit Data Collection & Analysis

1) Identification 2) Reporting 3) Root-cause analysis 4) Solution

determination 5) Dissemination

and resolution

www.asse.org MAY 2016 ProfessionalSafety 59

ging indicators (e.g., injuries, illnesses). If the near-hit is of high severity or danger is im- minent, the worker should execute the stop work au- thority and mitigate any hazards immediately. Simi- larly to hazard identification, construction workers should be trained as an extension of existing safety training pro- grams to identify and report near-hit events. For example, when workers are educated about proper PPE for work- ing at heights, they should also be instructed on how to identify and report cases in which coworkers are not wearing PPE while working at heights. The success of a near-hit reporting program largely depends on the ability and motivation of individuals to identify and report near- hits on construction sites.

Step 2: Reporting Construction site person-

nel who identify near-hits must report those events to their immediate supervisor through a near-hit reporting system. Depending on the site constraints, this report- ing system can use either electronic- or paper-based re- porting. Both systems should maintain employee anonym- ity, and both options must have the database capabil- ity to house the collected data.

Although required near-hit report criteria may vary between companies, a set of standard criteria is essential for each report (e.g., company name, event date, time, location, description). Automated near-hit reporting programs allow for photos with the report. Additional information might include record supervisor name, job/craft, possible conse- quences, corrective measures taken, whether fur- ther action is required, and whether the event was reported to the observer’s supervisor.

Step 3: Root-Cause Analysis Determining the factors that contributed to the

near-hit occurrence is the next step. When a near- hit is reported, use a consistent measure of cat- egorization so that similar events are categorized accordingly, regardless of who is taking in the re- port. Van der Schaaf (1992) used one such categori- zation scheme in his doctoral thesis work.

McKay (2013) later developed a construction- specific Eindhoven Classification Model (ECM); he used it to categorize more than 3,000 near-hits. The categories, defined in Table 1, are classified as either a skill-based, rule-based or knowledge-based factor.

Step 4: Solution Determination Once near-hits have been categorized, the next step

is to present solutions, taking into account the sever- ity and consequences of the preceding near-hit events. Simple, noncomplex or life-threatening events are treated as an exchange of information. More signifi- cant events are treated differently, but only after threats to life safety are removed and the site is rendered safe. These more complex events may involve changes in strategy on site and may involve the use of systematic root-cause analysis methods or work groups in order to find resolution. In many cases, a simple human er- ror determination using the ECM will direct the type of remedial actions needed in the field to prevent recur- rence of the unsafe condition or behaviors.

Table 1

ECM for Human Errors in Construction Factor   Category   Definition   Skill-­‐based   Slips   Failure  in  highly  developed  motor  skills  such  as  using  a  

hammer  but  missing  the  nail.   Tripping   Failure  in  whole-­‐body  movements  such  as  climbing  a  ladder,  

tripping  on  even  ground,  swinging  arm  or  kicking  something.   Rule-­‐based   Qualifications   Asking  someone  to  do  something  in  which  s/he  has  limited  

experience  or  knowledge.   Coordination   A  lack  of  coordination  between  two  construction  groups  such  

as  walking  into  a  barricaded  area  or  groups  not  coordinating   with  each  other  on  work  assignments.  

Verification   The  incomplete  assessment  of  something  on  the  worksite   such  as  using  equipment  which  hasn’t  been  inspected  or  using   the  wrong  materials  at  the  wrong  time  

Identification   Failures  that  result  from  faulty  task  planning  such  as  hazards   not  identified  on  the  job  safety  analysis  or  hazardous   conditions  that  remain  unrecognized.  

Monitoring   Improper  identification  controls  such  as  checks  or  calibration.   Compliance   Procedures  that  are  not  followed,  off  task  or  shortcuts.   Construction   Correct  design  that  was  not  constructed  properly  or  was  set  

up  in  inaccessible  areas  and  not  constructed  to  plan.   Protocol   Failures  relating  to  the  quality  and  availability  of  department  

protocols  (e.g.,  too  complicated,  inaccurate,  absent  or  poorly   presented).  

Knowledge-­‐   based  

Knowledge   Inability  of  a  person  to  apply  his/her  existing  knowledge  to  a   new  situation  (e.g.,  the  worker  was  unaware  of  a  rule).  

Other   External   Technical  failures  beyond  the  control  and  responsibility  of  the   investigating  organization.  

Mechanical   Failures  involved  with  mechanical  issues  beyond  the  control  of   field  personnel.  

Culture   Failures  resulting  from  collective  approach  and  its  attendant   modes  of  behavior  to  risks  in  the  investigating  organization.  

 

60 ProfessionalSafety MAY 2016 www.asse.org

Step 5: Dissemination & Resolution Ideally, corrective actions will have been em-

ployed in the field following the near-hit events and the work area will have been left in a safe state. In many cases, the reported near-hit may not have required a stop-work or other lifesaving mea- sure. The incident may have occurred, been cor- rected and workers in the area will have continued their jobs. If a near-hit occurs but is not reported, then the lesson learned is only of consequence for those in the immediate area. The broader audience (including all other site personnel) should be in- formed of the reported near-hit and corrective ac- tions taken, and should be communicated as soon as possible (e.g., the next day’s toolbox talks).

Safety managers will integrate learned lessons from the reported near-hit into existing safety train- ing. This step allows for the worker who reported the near-hit to receive feedback on how the situa- tion was corrected. By educating construction site personnel from other projects on lessons learned from near-hits, safety performance of workers can be enhanced. Figure 2 depicts the flow of informa- tion for a single reported near-hit.

Case Study A novel near-hit data collection and analysis sys-

tem was implemented on a large-scale liquefied nat- ural gas construction project located outside the U.S. The multibillion-dollar engineer-procure-construct project had a sophisticated, mature safety program, and recordable and lost-time rates that were stable and low compared to other construction projects in the same NAICS, but nonetheless stagnant.

The rates of near-hit reporting, first-aid cases and other recordable injuries as defined by OSHA were tracked for 15 weeks before and after imple- mentation of the near-hit data collection and anal-

ysis system. Researchers found the rates of near-hit reporting increased significantly after implementa- tion of the system (McKay, 2013). No statistically significant change was experienced between the values of first aids experienced before and after implementation. However, the number of OSHA- recordable injuries differed after the implementa- tion of the near-hit collection and analysis system (p = 0.026). A correlation study using Kendall rank correlation coefficient (Sen, 1968) identified the connection between the number of near-hits re- ported and OSHA-defined recordable injuries af- ter the system was implemented (Table 2).

The ECM adopted for construction safety was used to categorize near-hits reported. Types and frequencies of near-hits reported after implemen- tation of the near-hit data collection and analysis system are shown in Table 3. Table 3 also presents the number of incidents per category of preimple- mentation and postimplementation of the near-hit data collection and analysis system.

Significant differences were identified between the number of first-aid reportable cases before and after implementation of the system. The signifi- cance of the increase in near-hits reported allowed for a favorable testing situation in which research- ers theorized that an increase in near-hit reporting would affect the rates of first-aid cases and record- able injuries in an inverse relationship. Significant differences were also identified between the mea- sures of near-hit reporting, with an overall increase of 966% reported after system implementation on the construction site. The project experienced a 100% decrease in OSHA recordable cases during the time of the near-hit intervention.

The increased reporting of near-hits was attrib- uted to management investment and ownership of the implemented program, as well as a significant effort to educate employees about the near-hit program, including training on identifying near- hits, the reporting process and benefits of report- ing. No incentives were provided to employees for quantity or quality of near-hit reports.

The Mann-Whitney statistics (Ruxton, 2006) were used to correlate collected safety data. The rates of near-hits reported were inversely corre- lated with the number of recordable first-aid cases [r(30) = -0.281; p < 0.05] and the counts of record- able injury cases [r(30) = -0.373; p < 0.01]. The near-hit data collection and analysis system was found to be less correlated with recordable first-aid cases than with recordable injury cases.

One possible explanation for this is that first-aid cases seem to have a more random distribution. A first-aid case could include dust blown into the eye, treatment for an insect sting, heat rash or other con- ceivable event that could befall a person while at work. Recordable injuries are more action-oriented events that are typically the result of a larger release of energy such as a slip and fall, hitting one’s thumb

Figure 2

Flow of Near-Hit Information 1) Worker observes

a near-hit.

2) Worker reports the observed near-hit.

3) Safety manager compiles near-hits

in database.

4) Safety manager and investigative team

analyze near-hit.

5) Determined corrective actions are implemented.

6) Lessons learned from near-hits are

integrated into training.

www.asse.org MAY 2016 ProfessionalSafety 61

with a hammer or suffering a laceration while at work. This could be further investigated by research stemming from this initial attempt.

The top five near-hit cat- egories were identical across the preintervention and post intervention samples, and are reported in order from highest to lowest: com- pliance, identification, slips, trips and verification (Mc- Kay, 2013). In view of this, an OSH management program would have a target-rich en- vironment when consider- ing where to apply limited resources given that the top five near-hit types are related to human error and are ac- tive errors, but this should be tested on other projects.

Near-Hit Reporting for Other Industries

Many industrial sec- tors have benefitted from the collection and analysis of near-hits, including en- ergy production (Fabiano & Curro, 2012), medicine (Cal- lum, Kaplan, Merkley, et al., 2001) and manufacturing (Lander, Eisen, Stentz, et al., 2011). The chemical process- ing industry has experienced many benefits and improved safety from implementing near-hit reporting programs (Phimister, Oktem, Klein- dorfer, et al., 2003). When interviewing more than 100 chemical processing and management personnel from 20 chemical and phar- maceutical facilities, researchers identified a decrease in traditional mistakes that can contribute to inju- ries or illnesses. Furthermore, the study identified a qualitative improvement in the proactive approach of interviewees in terms of safety and mitigating unsafe situations (Phimister, et al., 2003). Consequently, an overall improvement in safety management among most interviewees was identified after implementing a near-hit reporting program. The benefits of near- hit reporting can be realized through industries.

Conclusion The strength of the near-hit reporting data col-

lection and analysis system lies in its ability to generate useful safety information for a given con- struction site. During the evaluation period, near-

hit information was consistently presented to the entire workforce in the form of plan-of-the-day meetings, toolbox talks or similar pre-work task planning sessions. The ability to collect, analyze and disseminate safety information allows em- ployers to mitigate hazardous events and condi- tions before an incident occurs.

The primary contribution of this research is the correlated link between the number of near-hits collected and the decreased number of first-aid cases and recordable injuries on a construction site. The ECM was modified to categorize near- hit events specific to construction sites for the first time, at least at the time of this publication. This initial research step provides a foundation for fu- ture research in near-hit reporting on construction site injuries. Future research could include correlat- ing near-hit reporting to expected severity and risk

Table 2

Correlation Coefficient Analysis

Note. Values marked in italics denote a correlation value that is statistically significant at the 0.05 level (1-tailed test). Numbers marked in bold show a correlation value that is statistically significant at the 0.01 level (1-tailed test).

  Near-­‐hits   reported  

First-­‐aid   count  

OSHA  injury   recordable   count  

First-­‐ aid  rate  

OSHA  injury   recordable   rate  

Near-­‐hits   reported  

Correlation   coefficient  

1.000   -­‐0.281   -­‐0.373   -­‐0.207   -­‐0.320  

Significance   (1-­‐tailed)  

  0.019   0.008   0.056   0.014  

 

Table 3

Near-Hit Categorization Using ECM        

Frequency   Total   frequency   Percent  

Cumulative   percent  

Pre-­‐ implementation  

Post-­‐ implementation  

Slips   9   41   50   4.5   4.5   Tripping   7   17   24   2.2   6.6   Coordination   5   118   123   11.0   17.7   Verification   10   56   66   5.9   23.6   Identification   21   264   285   25.6   49.2   Monitoring   0   4   4   0.4   49.6   Compliance   37   478   515   46.2   95.8   Construction   5   6   11   1.0   96.8   Protocol   0   1   1   0.1   96.9   Knowledge   3   8   11   1.0   97.8   External   2   1   3   0.3   98.1   Mechanical   5   15   20   1.8   99.9   Culture   0   1   1   0.1   100.0   Total           1,114   100.0      

 

62 ProfessionalSafety MAY 2016 www.asse.org

exposure and the generation of predicted variables or outcomes. PS

References

Bureau of Labor Statistics (BLS). (2012). Nonfatal occupational injuries and illnesses requiring days away from work, 2011. Retrieved from www.bls.gov/news .release/osh2.nr0.htm

BLS. (2013a). Accident/incident investigation. Re- trieved from www.osha.gov/SLTC/etools/safetyhealth/ mod4_factsheets_accinvest.html

BLS. (2013b). Industry at a glance: Construction. Retrieved from www.bls.gov/iag/tgs/iag23.htm

Callum, J.L., Kaplan, H.S., Merkley, L.L., et al. (2001). Reporting of near-miss events for transfusion medicine: Improving transfusion safety. Transfusion, 41(10), 1204- 1211. doi:10.1046/j.1537-2995.2001.41101204.x

Cambraia, F.B., Saurin, T.A. & Formoso, C.T. (2010). Identification, analysis and dissemination of information on near misses: A case study in the construction indus- try. Safety Science, 48(1), 91-99. doi:10.1016/j.ssci.2009 .06.006

Caterpillar. (2013). Object detection system. Caterpil- lar. Retrieved from www.cat.com/detect

Civil Aviation Authority (CAA). (2013). Safety management: Incident reporting and investigation. Retrieved from www.caa.co.uk/default.aspx?catid=1779 &pageid=9864

Donovan, T. (2011). U.S. nuclear power plants had 14 “near-miss” problems in 2010: UCS report. Retrieved from www.huffingtonpost.com/2011/03/17/us-nuclear -power-near-misses-2010_n_837176.html#s254851 &title=HB_Robinson

Fabiano, B. & Curro, F. (2012). From a survey on accidents in the downstream oil industry to the develop- ment of a detailed near-miss reporting system. Process Safety and Environmental Protection, 90(5), 357-367.

Flin, R., Mearns, K., O’Connor, P., et al. (2000). Mea- suring safety climate: Identifying the common features. Safety Science, 34(1), 177-192.

Goldenhar, L., Moran, S. & Colligan, M. (2001). Health and safety training in a sample of open-shop construction companies. Journal of Safety Research, 32(2), 237-252.

Hallowell, M., Hinze, J., Baud, K., et al. (2013). Proac- tive construction safety control: Measuring, monitoring and responding to safety leading indicators. Journal of Construction Engineering and Management, 139(10).

Hecker, S., Gambatese, J. & Weinstein, M. (2005, Sept.). Designing for worker safety: Moving the con- struction safety process upstream. Professional Safety, 50(9), 23-44.

Hinze, J. (2002). Safety plus: Making zero accidents a reality. Construction Industry Institute, 160-11, The University of Texas at Austin.

Hinze, J. & Godfrey, R. (2003). An evaluation of safety performance measures for construction projects. Journal of Construction Research, 4(1), 5-15.

Hinze, J., Madariaga, M. & Pizarro, D. (2006). Target safety: Programs focused on preventing specific hazards. Construction Industry Institute, 216-11, The University of Texas at Austin.

Hinze, J., Thurman, S. & Wehle, A. (2013). Leading indicators of construction safety performance. Safety Science, 51(1), 23-28.

Hinze, J. & Wilson, G. (2000). Moving toward a zero injury objective. Journal of Construction Engineering and Management, 126(5), 399-403.

Ho, C. & Dzeng, R. (2010). Construction safety training via e-learning: Learning effectiveness and user satisfaction. Computers and Education, 55(1), 858-867.

Huang, X. & Hinze, J. (2006). Owner’s role in con- struction safety. Journal of Construction Engineering and Management, 132(2), 164-173.

International Association of Fire Chiefs (IAFC). Fire near-miss. Retrieved http://fire.nationalnearmiss.org

International Labor Organization (ILO). (2003). Safety in numbers: Pointers for a global safety culture at work (Report No. 61), Geneva, Switzerland.

Krause, T., Groover, D. & Martin, D. (2010) Prevent- ing incidents and fatalities: Eight questions every senior leader should ask. Professional Safety, 55(6), 46-53.

Lander, L., Eisen, E., Stentz, T., et al. (2011). Near- miss reporting system as an occupational injury preven- tive intervention in manufacturing. American Journal of Industrial Medicine, 54(1), 40-48.

Lindsay, F. (1992). Successful health and safety man- agement: The contribution of management audit. Safety Science, 15(1), 387-402.

Martin, S., Etchegaray, J., Simmons, D., et al. (2005). Development and implementation of The University of Texas close call reporting system. Information for the Defense Community, 1-13.

McKay, B. (2013). Measures of effect: Near-miss reporting and construction site injuries (Master’s practi- cum thesis). Retrieved from Scholarworks@UA: http:// hdl.handle.net/11122/4828

OSHA. (2011). OSHA injury and illness recordkeep- ing and reporting requirements. Retrieved from www .osha.gov/recordkeeping

OSHA. (2013). Commonly used statistics. Retrieved from www.osha.gov/oshstats/commonstats.html

Phimister, J., Oktem, U., Kleindorfer, P., et al. (2000). Near-miss system analysis: Phase I. Wharton School, Center for Risk Management and Decision Processes, University of Pennsylvania.

Phimister, J., Oktem, U., Kleindorfer, P., et al. (2003). Near-miss incident management in the chemical pro- cess industry. Risk Analysis, 23(1), 445-459.

Rozenfield, O., Sacks, R., Rosenfeld, Y., et al. (2010). Construction job safety analysis. Safety Science, 48(4), 491-498.

Ruxton, G. (2006). The unequal variance t-test is an underused alternative to student’s t-test and the Mann- Whitney U test. Behavioral Ecology, 17(4), 688-690.

Sen, P.K. (1968). Estimates of the regression coef- ficient based on Kendall’s tau. Journal of the American Statistical Association, 63(324), 1379-1389.

van der Schaaf, T.W. (1992). Near-miss reporting in the chemical process industry (Ph.D. thesis). Eindhoven University of Technology.

van der Schaaf, T.W. & Kanse, L. (2004). Biases in incident reporting databases: An empirical study in the chemical process industry. Safety Science, 42(1), 57-67.

Wagner, L., Capezuti, E. & Ouslander, J. (2006). Reporting near-miss events in nursing homes. Nursing Outlook, 54(1), 85-93.

Wu, W., Huanjia, Y., Chew, D., et al. (2010). Towards an autonomous real-time tracking system of near-miss accidents on construction sites. Automation in Construc- tion, 19(1), 134-141.

Copyright of Professional Safety is the property of American Society of Safety Engineers 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.