Utterback, David F. PhD; Charles, Luenda E. PhD; Schnorr, Teresa M. PhD; Tiesman, Hope M. PhD; Storey, Eileen MD; Vossenas, Pamela MPH
The National Occupational Research Agenda (NORA) services sector includes many industry groups with elevated numbers and rates of occupational injuries and traumatic injury fatalities. (NORA) uses a framework of industry sectors from the North American Industrial Classification System (NAICS), a system which categorizes U.S. industries based on the goods or services they provide.1 A total of eight NORA sectors, most with a combination of multiple industry sectors, were designated and national research and intervention agendas have been posted on the National Institute for Occupational Safety and Health (NIOSH) internet site. The purpose of this study was to identify priorities for research and interventions across the NORA services sector through the tabulation and evaluation of occupational injury and fatality data between the years 2003 and 2007.
The NORA services sector was formed by the NIOSH as a combination of the following 11 North American Industry Classification System (NAICS) industrial sectors: information (NAICS 51); finance and insurance (52); real estate and rental (53); professional, scientific, and technical (54); management of enterprises (55); administrative support and waste services (56); education (61); arts, entertainment, and recreation (71); traveler accommodation and food services (72); other services except public administration (81); and public administration (92). The services sector is the largest of the eight NORA sectors employing more than 68 million workers,2 which is more than 50% of the US workforce. Work environments and occupational hazards vary greatly within the services sector, ranging from many industry groups with relatively low annual injury and illness counts and rates to others with much higher injury and illness counts, rates of days-away-from-work (DAFW) injuries, and large fatality counts.
Data used for these analyses were collected and tabulated by the US Bureau of Labor Statistics (BLS),1–3 which are available to the public through BLS Web sites.4 They include establishment-based employment data collected by the occupational employment statistics (OES) program; occupational nonfatal injury and illness data from the annual Survey of Occupational Injuries and Illnesses (SOII), and fatal traumatic injury data from the Census of Fatal Occupational Injures (CFOI). The OES and SOII are population-based surveys and CFOI is a complete census of occupational fatalities.
The OES program is a Federal/State cooperative program that surveys approximately 200,000 establishments by mail every 6 months, taking 3 years to fully collect the sample of 1.2 million establishments.5,6 The OES survey covers all full-time and part-time wage and salary workers in nonfarm industries, yet it excludes self-employed, owners, and partners in unincorporated firms, household workers, and unpaid family workers. Data on wage and salary workers are collected to produce national estimates of employment and wage for about 800 occupations and more than 450 industries. Workers are categorized into 22 major occupational groups and in one of 801 detailed occupations according to the 2000 version of the Standard Occupational Classification (SOC). The SOC major group excluded from the OES survey consists of military-specific occupations.7 Industry classifications correspond to the two-, three-, four-, and five-digit NAICS industrial classes.3 The NAICS levels are defined as sectors, subsectors, industry groups, and industries, respectively.
Nonfatal Occupational Injury and Illness Data
Between 2003 and 2007, the Federal/State SOII program statistics were collected annually from about 176,000 private industry establishments.8 The program requires employers to report summary injury and illness information from their Occupational Safety and Health Administration record-keeping logs. These logs contain data for all disabling, serious, or significant injuries and illnesses, whether or not time-away-from-work was required. Nature, event, source, and part of body are also collected by SOII for cases resulting in at least 1 DAFW. These characteristics are coded according to the Occupational Injury and Illness Classification System.3 The SOII also collects data for the number of employee hours worked and average employment to calculate prevalence rates and to classify establishment size. The survey began using version 2002 NAICS codes to classify industry in 2003. The SOII excludes all work-related fatalities as well as nonfatal work injuries and illnesses involving self-employed workers; workers on farms with 10 or fewer employees; private household workers; and Federal, State, and local government workers.3 Days-away-from-work were used in this analysis because they represent more serious injuries and illnesses, are likely to be reported more consistently, and have additional characteristic information that is useful for identifying intervention needs.
Occupational Fatality Data
The CFOI is a Federal/State cooperative program in all 50 states and the District of Columbia, Puerto Rico, and the Virgin Islands. Data for CFOI have been collected since 1992 for all occupational traumatic injury fatalities regardless of age, industry, or workers' compensation coverage. The BLS defines fatal work-related injury as those fatalities occurring to noninstitutionalized persons, who were working at the time of the incident, and on the premises of their employer or other places while on active duty. Multiple information sources are used by BLS to identify, verify, and profile fatal worker injuries. Information about each workplace fatality—occupation and other worker characteristics, source of injury, and circumstances of the event—is obtained by cross-referencing source records, such as death certificates, workers' compensation reports, and Federal and State agency administrative reports. To ensure that fatalities are work related, cases are substantiated with two or more independent source documents, or a source document and a follow-up questionnaire.3
Counts are available for private and public sector fatalities and can be reported separately or combined. In addition, wage and salary worker fatalities are available for industry groups. Fatality rates are published by BLS only for those occupations and industries that had at least 30 fatalities and 40,000 employed in 2003, the base year.9 As a result, fatality rate estimates are published for only 14 of the three-, four-, and five-digit services sector NAICS codes.10 Instead of these limited data, we calculated rates for industry groups with wage and salary worker fatality counts as numerators and OES population estimates as denominators.
May 2007 employment estimates by two-digit NAICS sectors and major occupational groups of SOC within each industry code were copied from public use data sets of OES.2 In addition, the OES 2007 estimates for employment within four-digit NAICS industry groups were tabulated for those services sector industries with published annual SOII rates for at least 4 of 5 years, 2003 to 2007, or with noncensored wage and salary worker fatality values in each of the years.
Occupational Injury and Illness Data
The Prevention Index (PI), in which the count and rate rankings of injuries and illnesses within a data set are reduced to a combined score, can be used to guide priorities for occupational safety and health resources.11 First developed in Washington State with workers' compensation data,12 the method is duplicated here with industry-group-specific estimates from SOII. The data utilized in the analysis were the set of annual SOII DAFW counts and rates for NORA services sector industry groups that had estimates for at least 4 years between 2003 and 2007. Estimates for SOII were tabulated using BLS Profile Web site.4 To determine the PI, industry group DAFW rates and counts for each year were separately ranked from highest to lowest. The average of the industry group annual ranks was determined for both rates and counts. For each industry group, the PI was calculated as the average of the rank orders of both rates and counts and then rank ordered from lowest to highest. In cases where one-yearly SOII rate estimate was missing, the average rank from the other 4 years was used. Missing counts were imputed as the average of the other four counts for that industry group.
Occupational Fatality Data
The BLS CFOI annual data were tabulated for the 92 unique four-digit NAICS industry groups in the NORA services sector for 2003 to 2007. In addition, Justice, Public Order, and Safety Activity (also known as “Public Safety, NAICS 9221”) worker fatal injuries at Federal, State, and local levels were tabulated in each year. Fatal injuries in the remaining public administration groups at the three levels of government are the differences between their total fatalities and those experienced by the public safety workers. All together, fatality data for 98 industry classes were evaluated. Publicly available CFOI data from 2003 to 2007 were downloaded for each industry group using the BLS Profiles Web-based data system.4 Because CFOI does not report cells with fewer than three deaths, only those services industry groups with uncensored counts in each of the 5 years were tabulated and listed in order of frequency. Fatalities due to transportation events and homicides are prevalent in services sector industry groups, so we tabulated fatalities for the industry groups with higher counts or proportions of deaths due to these events as a way to identify intervention needs.
The CFOI fatality rates are published for only a few of the services industries, so they are not presented here.10 Instead, occupational fatality rates were calculated for industry groups as the average number of wage and salary worker fatalities for 2003 to 2007 divided by the estimated number of wage and salary workers from 2007 OES and expressed as the number of fatalities per 10,000 workers.
Table 1 displays 2007 OES estimates of employment by occupation class for the industries that make up the NORA services sector. Nearly 50% of the 68 million workers in the NORA services sector are employed in three industry sectors: education (n = 12,455,550, 18%) (NAICS 61), accommodations and food services (n = 11,341,810, 17%) (72), and public administration (n = 9,533,390, 14%) (92). All two-digit NAICS sectors include diverse populations across SOC classes with few null values in the cross tabulations. The largest SOC classes in the NORA services sector are administrative support (12.7 million, 19%), food (10.2 million, 15%), and education (7.8 million, 11%). Only education and food services sectors have greater than 50% of workers from a single major occupational class. Because OES counts the number of workers at establishments, multiple jobholders can be counted more than once.
Nonfatal Occupational Injury and Illness Data
Fifty of the 92 four-digit NAICS industry groups in the NORA services sector had SOII DAFW rate estimates tabulated for at least 4 of the 5 years between 2003 and 2007 (Table 2). Annual ranks for most industry groups were relatively stable from year to year; the average industry group range of ranks was less than 8. The greatest ranges of ranked annual values for industry groups were 25 for personal and household goods repair and maintenance (NAICS 8114) and 22 for commercial and industrial machinery and equipment rental (NAICS 5324).
TABLE 2-a. Four-Digi...Image Tools
The median DAFW for the highest-ranking NORA services sector industry groups (Table 2) ranged from 5 to 18 days in 2007. Over half of the ranked industry groups have median DAFW values of 7 or less. Those with median DAFW values greater than 11 are recreational vehicle parks and recreational camps (median = 18), dry cleaning and laundry services (15), death care services (15), waste treatment and disposal (12), and wired telecommunications carriers (12). The leading events or exposures listed by SOII for the injuries in these industry groups include contact with objects and equipment, falls on same level, and overexertion (data not shown).
TABLE 2-b. Four-Digi...Image Tools
The highest 25 PI ranks for the NORA services sector industry groups are listed in Table 3. The list includes seven large industry groups, each with greater than 1 million workers, and three relatively small industry groups with fewer than 100,000 workers each. The average population of the ranked industry group size is 1.26 million workers. The NAICS sectors with the largest number of ranked industry groups in Table 3 are as follows: 53—real estate and rental (N = 5); 56—administrative support and waste services (N = 5); 72—traveler accommodation and food services (N = 4); 71—arts, entertainment, and recreation (N = 3); and 81—other services except public administration (N = 3). The two industry groups with the highest PI rankings are NAICS 5617—services to building and dwellings and NAICS 7211—traveler accommodations.
Occupational Traumatic Injury Fatality Data
The total 2003 to 2007 fatality count for the NORA services sector at the two-digit NAICS level is 7740, which is more than 27% of all US occupational fatalities for those years (data not shown). The sum of all reported services sector fatalities at the industry group or four-digit level is 7614. The difference, 126 fatalities, is dispersed among the censored values in the more detailed industry group data sets. It is noteworthy that fatality counts for some industry groups include private and public sector employees, for example, Waste Collection and Elementary and Secondary Education.
Overall, 33 of the NORA services sector industry groups recorded 50 or more total fatalities for the 5-year period (Table 4). Twenty-two of these groups (67%) reported greater than 100 fatalities for the 5-year period. The greatest number of fatalities (n = 1047, 14%) occurred in the services to buildings and dwellings industry group (NAICS 5617). The majority of these fatalities (n = 847, 81%) were recorded in landscaping services (NAICS 56173) industry. Most of the fatalities that occurred in services to buildings and dwellings were due to transportation incidents (n = 297, 28%), falls from elevation (n = 268, 26%), contact with objects and equipment (n = 233, 22%), especially struck by falling objects (n = 147, 14%), and contact with electric current (n = 98, 9%).
Fatalities for public safety, which includes law enforcement, firefighting (may also operate emergency medical services), and corrections, had the second highest total with 839 deaths for 2003 to 2007 (Table 4). Approximately 70% of these fatalities (N = 585) were among local police protection (NAICS 92212) and 26% for fire protection (N = 215) (NAICS 92216). For both police protection and fire protection personnel, nearly one half of their fatalities were due to transportation events. About 40% of local law enforcement fatalities were attributed to homicides.
Automotive repair, solid waste collection, and limited-service eating establishments industry groups each experienced more than 260 fatalities, or an average of one death per week. Fatalities in automotive repair were primarily due to assaults and violent acts, contact with objects and equipment, and transportation. Solid waste collection fatalities mostly resulted from transportation incidents and contact with objects and equipment. Limited-service eating establishment fatalities were mostly homicides.
Estimated fatality rates for salary and wage workers by industry group also are presented in Table 4. This table includes only those industry groups with greater than 50 total fatalities and with fatalities tabulated in each year, 2003 to 2007. Rates were calculated while using the 2007 OES estimates of the salary and wage worker populations within industry groups as denominators. When using the OES denominators, our method produced an overall 2007 fatality rate of 0.33 per 10,000 salary and wage workers. The occupational fatality rate for all US wage and salary workers published by CFOI for 2007 was 0.34 per 10,000 workers10 (The published CFOI rates are calculated using Current Population Survey [CPS] denominator data).
Twelve services sector industry groups experienced more than 50 total fatalities and fatality rates greater than the 2007 US average rate for wage and salary workers of 0.34 per 10,000 workers, according to our calculations. The three highest fatality rates for wage and salary services sector workers were in the private sector waste management and remediation industry (NAICS 562). The majority of these fatalities occurred in transportation incidents. Religious organizations also experienced an elevated fatality rate. Most of these fatalities were due to highway incidents and falls to lower levels. Services to buildings and dwellings led all industry groups with 1047 fatalities and their fatality rate for wage and salary workers was about 2 times the US average.
Among most services sector industry groups that averaged more than 20 total fatalities per year, transportation events were the leading cause of fatalities (Table 5). The largest number of transportation-related fatalities occurred in services to buildings and dwellings (n = 297) and local police protection industries (n = 287). The proportion of transportation fatalities was particularly high in technical and trade schools (97%), newspaper, periodicals, book and directory publishers (81%), waste collection (71%), architectural, engineering and related services (67%), waste treatment and disposal (57%), and spectator sports (53%). Remarkably, 106 of 109 transportation fatalities in technical and trade schools were associated with aircraft crashes during flight training.13
Homicides were also a leading cause of occupational fatalities in a number of services sector industry groups during 2003 to 2007 (Table 6). Local law enforcement experienced the largest number of homicides (232), which accounted for approximately 40% of their fatalities. The industry group with the second largest number of homicides (172) was limited-services eating establishments. Three food and beverage industry groups combined account for a total of 399 homicides between 2003 and 2007. The largest proportion of all fatalities due to homicides (83%) was observed in the personal care services industry group. The proportion of homicides attributed to robbery was the highest in these industry groups: Nondepository credit intermediation (69%), limited-service eating places (56%), traveler accommodation (40%), and investigation and security services (40%).
This review, which relied on BLS data systems between 2003 and 2007, reveals elevated occupational injury and fatality risks for many US services sector workers, although data are limited for many of the industry groups. Fatality counts are quite high for landscape services workers and local government public safety workers. Automotive repair, solid waste collection, and limited-service eating establishments industry groups also totaled greater than 260 fatalities, or an average of one or more deaths per week. The PI method, as applied with BLS SOII DAFW data, identified elevated risks for nonfatal injuries and illnesses in many of the 92 industry groups in the services sector. The largest numbers of ranked industry groups were in real estate and rental (NAICS 53), administrative support and waste services (NAICS 56), and traveler accommodation and food services (NAICS 72). Many of the ranked industry groups from the PI analysis also experienced more frequent fatalities.
Combinations of rates and counts for occupational injuries and illnesses help to direct priorities in health and safety prevention programs.11,14,15 The PI-ranked industries are based on injury and illness rates and counts from Washington State workers' compensation data.12 A similar method is applied here with the SOII counts and rates for services sector industry groups but is limited to those that had published data for at least 4 years between 2003 and 2007. The PI gives equal weight to count and rate ranks and frequently these ranks differ. The different patterns of count and rate ranks have been suggested as guides for public health prevention strategies.16
Using the PI, the services sector industry groups with greater DAFW injury and illness risks are more prevalent in real estate and rental (NAICS 53); administrative support and waste services (56); traveler accommodation and food services (72); arts, entertainment, and recreation (71); and other services (81). Hazards across these industry sectors are highly diverse, yet the surveillance data for injuries are generally consistent with US industry as a whole. Contact with equipment and objects, falls on same level and from heights, transportation incidents, and assaults and violent acts are the most frequent events and exposures associated with the injuries, although the order of prevalence and severity differs among the groups. Functional limitations from the injuries can be severe with values for the median DAFW as high as 18.
The SOII data collection included a nationally representative sample of establishments of all sizes and industries in the private sector (before 2008) but excluded self-employed, farms with fewer than 11 employees, private households, US postal service, and public sector establishments.3 The SOII publishes estimates for total cases and cases that result in reassignment or transfer, plus DAFW cases with additional detailed nature, event, part of body, and source and demographic variables. We chose to use DAFW data in this analysis because they represent more serious injuries and illnesses and the additional information that is available helps identify research and intervention needs. The SOII also collects data on employment levels that are used for rate calculations and classifying establishment size.3
Cases reported to SOII are those that should appear on establishment Occupational Safety and Health Administration injury and illness record-keeping logs, that is, those injuries and illnesses that require medical treatment beyond first aid, result in job restrictions or the employee being reassigned or transferred to a different position, or cause the employee to miss work for at least 1 day after the date of injury or illness.3 Recent reports have expressed concerns about underreporting of occupational injuries and illnesses in the SOII data.8,17–21
Only 50 of the 92 NAICS industry groups in the services sector met the “4 of 5-year criterion” for inclusion in the PI-ranked data set. A few industry groups, which were excluded because of limited SOII data, namely cable and other program distribution (NAICS 5175) and spectator sports (NAICS 7112), also experienced rates and counts that were elevated in at least 2 of the 5 years, but the data were deemed insufficient for generalizability. Appropriate research and intervention needs may, nonetheless, be high priorities for those industry groups and additional surveillance with BLS SOII data for many higher-risk groups is warranted.
The BLS published NORA services sector CFOI rates for 2003 to 2007 for only seven subsectors (three-digit NAICS codes), five industry groups (four-digit), and two industries (five-digit). These groupings are consistent with Census Codes.10 Because of the small number of available rates, they are not presented here. Instead, we calculated fatality rates for industry groups as the ratio of the average annual number of CFOI wage and salary worker fatality counts to the 2007 OES population estimates of wage and salary workers in the industry groups. Nevertheless, OES excludes workers who may constitute significant portions in some services sector industry groups such as self-employed, unincorporated partnerships, household workers, and unpaid family members. To make the comparisons as accurate as possible, we used wage and salary worker fatality counts as numerators. The OES can differ substantially from CPS industry group estimates.3 The magnitude of differences in the population estimates can approach a factor of 2, with either system at times having a larger estimate. Because OES counts the number of workers at establishments, multiple jobholders can be counted more than once within an industry. For example, the CPS household survey estimate of employment for landscaping industry (NAICS 56173) is 1.276 million employees in 2007 whereas OES establishment-based survey estimates 681,010 employees. In contrast, CPS estimates 8.3 million employees in food services whereas OES estimates 9.5 million employees. Where OES might underestimate populations within groups, our rates would be higher than actual. Where OES overestimated populations, our estimates would be lower than actual. (CFOI has begun publishing rates on the basis of estimated hours worked for better comparisons among groups.3)
Primary causes of occupational traumatic injury fatalities among services sector employees are transportation incidents, assaults, and violent acts, contact with objects and equipment, and falls—similar to what is seen with DAFW injuries and illnesses. The numbers and rates of fatalities among many services sector industry groups were elevated over the 5-year period, especially for landscape services, public safety, automotive repair, and the solid waste management industry as a whole. Previous researchers, using data from the Fatality Analysis Reporting System between 1992 and 2001, reported that the services industry accounted for the second highest number of motor vehicle–related deaths in the United States (n = 1884).22 Transportation occupational fatalities and traumatic injuries can be prevented through effective comprehensive safety and health programs and interventions such as mandatory seat belt use policies, fatigue management, adequate allowance for transit times, and training for operation of specialized vehicles.23 Recent studies have identified best practice approaches with respect to vehicle selection, integrated training and education, and incentives for crash-free driving.24 We found that the highest number of transportation fatalities came from the services to buildings and dwellings industry group (n = 297), specifically, landscaping services. Transportation safety may be overlooked in the landscaping industry because many of these companies are family-owned and operate in a small business setting.25
The second highest number of transportation fatalities came from the local police protection industry group (n = 287). Traffic-related incidents have been the leading cause of occupational death among our nation's law enforcement officers for the last 13 years and the number of traffic-related deaths significantly increased between 2009 and 2010.26 The exact reasons for these findings are unknown, but several factors could play a role. First, training in high-speed and other risky driving situations is inconsistent for law enforcement nationwide.27 Second, research has demonstrated that there are situations where the wearing of a seat belt may interfere with the job responsibilities of a police officer, such as exiting a patrol car quickly or having the belt interfere with firearm access.28 These point to areas of future intervention research into effective seat belt redesign or various continuing education training programs.27
Fatality risks from assaults and violent acts are evident in the homicide data reported by CFOI. Homicides averaged more than 20 per year in local police protection (n = 232), limited-services eating places (172), drinking places (122), and full-service restaurants (105). Workplace violence is a pervasive problem—approximately 50% of all workplace homicides nationwide occurring within the services sector population. Robbery is frequently associated with homicides in limited-services eating establishments. Yet, some services industry group proportions of homicides associated with robbery are relatively low such as drinking places (<18%), full-service restaurants (37%), and automotive repair and maintenance (<24%). Reductions in fatalities in these industry groups will require effective interventions, which are dependent on the types of violence that are present.
The elevated incidence of homicide among services sector workers has been reported in other studies. Tiesman et al27 reported that 1072 law enforcement officers died as a result of homicide between 1992 and 2002, 5.6 officers per 100,000 workers. A significantly higher proportion of these deaths were among male officers. Women in the accommodation and food services industry have significantly higher workplace homicide rates than women in other industries.29 Between 1980 and 1992, the locations with the highest risk of homicide were grocery stores, eating and drinking establishments, hotels and motels, and justice and public order.30 The latter three of these groupings are in the services sector and homicide rates continue to be elevated in each. Factors such as working alone, working at night, and assisting customers have been shown to increase the risk for workplace homicide and these conditions are common in many services sector industry groups.31 Workplace violence is also common among young workers, minority workers, and immigrants.32–34
The full spectrum of occupational risks among services sector workers cannot be characterized with available BLS injury, illness, and fatality data. In the BLS data, acute disorders, such as skin diseases, respiratory conditions, and poisoning, are more likely reported than chronic illnesses such as cancers, which have long latencies between occupational exposure and clinical diagnosis.3 Nationally representative surveillance systems do not exist for most chronic outcomes.35–37
Several factors apart from occupational risks may contribute to some of our findings. For example, health disparities exist among workers in many services sector industries and numerous factors contribute to these disparities. Lack of health insurance may be the largest and the most important contributor.38 Even workers' compensation insurance is not universally available or utilized for occupational injuries and illnesses.21 Other studies show that blue-collar workers, older workers, and Hispanic workers, many of whom are employed in the services sector, have some of the lowest rates of preventive screenings for chronic diseases and highest rates of uncorrected visual and/or hearing impairment.39,40
The methods used in this analysis have at least four notable limitations. First, neither industry groups nor major occupation classes can be used as sensitive or specific indicators of exposure to occupational hazards. More detailed information is needed to match jobs with tasks that are associated with recognized hazards or to identify emerging hazards. Second, OES estimates of all full- and part-time employees used as fatality rate denominators exclude self-employed individuals, owners, and partners in unincorporated firms, household workers, or unpaid family workers.5 These exclusions may be significant because many small, unincorporated businesses exist in the services sector. Actual fatality rates may be lower than reported here for industry groups that employ a substantial number of workers through small, unincorporated entities. Third, SOII counts and rates are not available for public administration before data collection for calendar year 2008, so they are omitted from this analysis. An additional 48 four-digit industry groups in the services sector were omitted because of missing SOII estimates for more than 1 year. Additional SOII data on rates and counts for total occupational injuries and illnesses that result in reassignment or job transfer are available but are not reported here. Fourth, sufficient surveillance information does not exist for chronic occupational disease and illness.
The risks for occupational injuries, illnesses, and fatalities are elevated for many industry groups in the services sector and intervention strategies developed for these and other industries should be more widely utilized. Publically available data from BLS can be used to evaluate many risks even though the injury and illness data have been reported to undercount actual events 8,17–21 and, on the whole, the illness data are less useful than fatality and injury data. It is necessary to “drill down” in the data to best evaluate risks for specific industries and occupations and to identify the attendant intervention needs. Although aware of the many caveats, the NORA services sector used these data in conjunction with stakeholder input and expert knowledge to establish a set of priority occupational safety and health research and intervention goals. The sector will continue to monitor these data sources and others that are available to evaluate the appropriateness of the strategic and intermediate goals in the National Services Agenda.41
2. US Bureau of Labor Statistics. Occupational Employment Statistics, May 2007 National Industry-Specific Occupational Employment and Wage Estimates. Washington, DC: US Department of Labor; 2008. http://www.bls.gov/oes/2007/may/oessrci.htm
. Accessed September 13, 2011.
3. US Bureau of Labor Statistics. BLS Handbook of Methods. Occupational Safety and Health Statistics. Washington, DC: US Department of Labor; 2009. http://www.bls.gov/opub/hom/homch9.htm
. Accessed September 13, 2011.
4. US Bureau of Labor Statistics. Occupational Injuries/Illnesses and Fatal Injuries Profiles. Washington, DC: US Department of Labor; 2011. http://data.bls.gov/gqt/InitialPage
. Accessed September 13, 2011.
5. US Bureau of Labor Statistics. BLS Handbook of Methods. Occupational Employment Statistics. Washington, DC: US Department of Labor; 2008. http://www.bls.gov/opub/hom/pdf/homch3.pdf
. Accessed September 13, 2011.
6. US Bureau of Labor Statistics. Occupational Employment Statistics Overview. Washington, DC: US Department of Labor; 2011. http://www.bls.gov/oes/oes_emp.htm
. Accessed September 13, 2011.
8. Ruser J. Examining evidence on whether BLS undercounts workplace injuries and illnesses. Monthly Labor Rev. 2008;131:20–32.
11. Bonauto DK, Silverstein BA, Adams D, Foley MP. Prioritizing industries for occupational injury and illness prevention and research, Washington State workers' compensation claims, 1999–2003. J Occup Environ Med. 2006;48:840–851.
12. Silverstein B, Viikari-Juntura E, Kalat J. Use of a prevention index to identify industries at high risk for work-related musculoskeletal disorders of the neck, back, and upper extremity in Washington State, 1990–1998. Am J Ind Med. 2002;41:149–169.
13. US Bureau of Labor Statistics. Table A-1. Fatal Occupational Injuries by Industry and Event or Exposure, All United States, 2007.Washington, DC: US Department of Labor; 2009. http://www.bls.gov/iif/oshwc/cfoi/cftb0223.pdf
. Accessed September 13, 2011.
14. Anderson VP, Schulte PA, Sestito J, Linn H, Nguyen LS. Occupational fatalities, injuries, illnesses, and related economic loss in the wholesale and retail trade sector. Am J Ind Med. 2010;53:673–685.
15. Okun A, Lentz TJ, Schulte P, Stayner L. Identifying high-risk small business industries for occupational safety and health interventions. Am J Ind Med. 2001;39:301–311.
16. Silverstein B. Safety and Health Assessment and Research for Prevention (SHARP) Program. In:Utterback DF, Schnorr TM, eds. Use of Workers' Compensation Data for Occupational Injury and Illness Prevention, Proceedings from a September 2009 Workshop
. Cincinnati, OH: National Institute for Occupational Safety and Health; 2010:5–10. NIOSH Publication No 2010-152.
17. Boden LI, Ozonoff A. Capture-recapture estimates of nonfatal workplace injuries and illnesses. Ann Epidemiol. 2008;18:500–506.
18. Boden LI, Ozonoff A. Researcher judgment and study design: challenges of using administrative data. Am J Ind Med. 2010;53:37–41.
19. Oleinick A, Zaidman B. The law and incomplete database information as confounders in epidemiologic research on occupational injuries and illnesses. Am J Ind Med. 2010;53:23–36.
20. Rosenman KD, Kalush A, Reilly MJ, Gardiner JC, Reeves M, Luo Z. How much work-related injury and illness is missed by the current national surveillance system? J Occup Environ Med. 2006;48:357–365.
21. Azaroff LS, Levenstein C, Wegman DH. Occupational injury and illness surveillance: conceptual filters explain underreporting. Am J Public Health. 2002;92:1421–1429.
22. Work-related roadway crashes—United States, 1992–2002. MMWR Morb Mortal Wkly Rep. 2004;53:260–264.
23. National Institute for Occupational Safety and Health. Work-Related Roadway Crashes—Prevention Strategies for Employers. Cincinnati, OH: US Department of Health and Human Services, Centers for Disease Control and Prevention; 2004. NIOSH Publication No. 2004-136.
24. Haworth N, Tingvall C, Kowaldo N. Review of Best Practice Road Safety Initiatives in the Corporate and/or Business Environment. Report No. 166. Clayton, Victoria, Australia: Monash University Accident Research Centre; 2000.
25. Buckley JP, Sestito JP, Hunting KL. Fatalities in the landscape and horticultural services industry, 1992–2001. Am J Ind Med. 2008;51:701–713.
26. National Law Enforcement Officers Memorial Fund. Research Bulletin: Law Enforcement Officer Deaths: Preliminary 2010. Washington, DC: National Law Enforcement Officers Memorial Fund; 2011.
27. Tiesman HM, Hendricks SA, Bell JL, Amandus HA. Eleven years of occupational mortality in law enforcement: the census of fatal occupational injuries, 1992–2002. Am J Ind Med. 2010;53:940–949.
28. Oron-Gilad T, Szalma JL, Stafford SC, Hancock PA. Police officers seat belt use while on duty. Transport Res F. 2005;8:1–18.
29. Tiesman HM, Gurka KK, Coben J. fatal workplace violence among US women: the census of fatal occupational injuries, 2003–2008. Paper presented at: American Public Health Association; 2010; Denver, CO.
30. Jenkins EL. Homicide against women in the workplace. J Am Med Womens Assoc. 1996;51:118–122.
31. National Institute for Occupational Safety and Health. Workplace Violence Prevention Strategies and Research Needs. Cincinnati, OH: US Department of Health and Human Services, Centers for Disease Control and Prevention; 2006. NIOSH Publication No. 2006–144.
32. Rauscher KJ. Workplace violence against adolescent workers in the US. Am J Ind Med. 2008;51:539–544.
33. Wheeler K, Zhao W, Kelleher K, Stallones L, Xiang H. Immigrants as crime victims: experiences of personal nonfatal victimization. Am J Ind Med. 2010;53:435–442.
34. Shannon CA, Rospenda KM, Rickman JA, Minich LM. Race, racial discrimination, and the risk of work-related illness, injury, or assault: findings from a national study. J Occup Environ Med. 2009;51:441–448.
35. Leigh JP, Robbins JA. Occupational disease and worker' compensation: coverage, costs, and consequences. Milbank Q. 2004;82:689–721.
36. Schulte PA. Characterizing the burden of occupational injury and disease. J Occup Environ Med. 2005;47:607–622.
37. Steenland K, Burnett C, Lalich N, Ward E, Hurrell J. Dying for work: the magnitude of US mortality from selected causes of death associated with occupation. Am J Ind Med. 2003;43:461–482.
38. McCollister KE, Arheart KL, Lee DJ, et al. Declining health insurance access among US Hispanic workers: not all jobs are created equal. Am J Ind Med. 2010;53:163–170.
39. Vidal L, LeBlanc WG, McCollister KE, et al. Cancer screening in US workers. Am J Public Health. 2009;99:59–65.
40. Davila EP, Caban-Martinez AJ, Muennig P, et al. Sensory impairment among older US workers. Am J Public Health. 2009;99:1378–1385.
41. National Institute for Occupational Safety and Health. National Occupational Research Agenda. Washington, DC: US Department of Health and Human Services, Centers for Disease Control and Prevention; 2009. http://www.cdc.gov/niosh/nora/
. Accessed February 9, 2011.