Comparing fatal work injuries in the United States and the European Union

In July 2012, at the special session on statistics at the Seventh United States–European Union Joint Conference on Occupational Safety and Health in Brussels, Belgium, representatives from (1) the Bureau of Labor Statistics (BLS) and the Occupational Safety and Health Administration (OSHA), both part of the U.S. Department of Labor; (2) the National Institute for Occupational Safety and Health, part of the U.S. Department of Health and Human Services; and (3) Eurostat, the statistical office of the European Union (EU), discussed the potential for developing comparable data on workplace safety and health. Both the United States and the European Union capture and report data on worker safety, including information on fatal work injuries and nonfatal injuries and illnesses. However, differences in injury definitions, data sources, and collection techniques make data comparisons difficult. The representatives discussed these differences and recognized that the data are most comparable for fatal work injuries. At the close of the Conference, BLS and Eurostat agreed to produce a comparison of work-related fatal injury counts and rates, using data for calendar year 2010. This article summarizes the results of that comparison.¹

A comparative overview of U.S. and EU data

Fatal work injury data for U.S. workers come from the BLS Census of Fatal Occupational Injuries (CFOI) program, which provides an annual count of all fatal work injuries in the United States. Data are available from 1992 to the present. CFOI data are compiled with the use of multiple sources, including death certificates, coroner reports, OSHA reports, and media reports; each fatal work injury is substantiated by at least two different source documents. On average, four source documents are identified for each fatal work injury. CFOI data provide details about the fatally injured worker (such as age and gender), the worker’s employment (such as occupation and industry), and the circumstances of the fatality (such as the source of, and event resulting in, the fatal injury). BLS releases these details in preliminary form in the summer following the year of death and in revised and final form several months later.²

Data for EU workers come from the European Statistics on Accidents at Work (ESAW) program and are mainly derived from administrative sources.³ Data collection started in 1993, with data being produced and published annually. Employers in EU Member States are required to keep a list of occupational accidents resulting in a fatality or in a worker being unfit for work for more than 3 working days. Information collected includes characteristics of the worker (such as age and gender), characteristics of the enterprise (including economic activity and size of the enterprise), and characteristics of the injury (such as type and severity). Details on injury causes and circumstances—that is, the direct source of an injury and the sequence of events surrounding it—are similar to CFOI source and event data, but not yet available from all Member States. Of particular interest to the European Union is the concept of deviation, or the way in which the circumstances of an accident differ from a normal work practice.

Member States provide their occupational safety and health data to Eurostat no later than 18 months after the end of the reference year. Because each Member State gathers its data independently, the aggregate data for the European Union include some variation from one country to another. For example, the United Kingdom and Ireland do not supply data on road accidents. In addition, the ESAW data used in this article do not include data from Greece. With the 2011 adoption of a Commission Regulation on ESAW—a regulation that came into effect in 2013—these problems of comparability and lack of information will be gradually resolved.⁴

Data comparability adjustments

BLS used the available data to identify additional differences in the data sources and adjust the U.S. data to make them comparable to the EU data. The adjustments included the following:

• Excluding from the U.S. data fatal work injuries that occurred more than 1 year after the precipitating event or exposure. The ESAW data include only deaths that occurred within 1 year of the reference year. By contrast, the CFOI program includes all fatal injuries that meet the definitions of work relationship and are attributable to a work-related event or exposure, regardless of the amount of time between that event or exposure and the worker’s death. Although most fatalities occur instantly or within a few days of the event or exposure, the CFOI does include a small number of cases in which the death occurs months or years later.
• Excluding suicides from the U.S. data. CFOI data include workplace suicides, whereas EU data do not.

In addition, BLS limited the comparison to private-sector wage and salary workers, thus excluding public-sector workers and the self-employed. Also, the comparison is for the United States as a whole and the European Union as a whole and does not include information on individual U.S. states or individual EU Member States.

The fatality data in this article are presented for the so-called main NACE branches, which are industry aggregates used by the European Union in accordance with the Statistical Classification of Economic Activities in the European Community, an industrial classification system commonly referred to as NACE.⁵ Because the U.S. data are classified on the basis of the North American Industry Classification System (NAICS), BLS staff reclassified the U.S. data with the use of a NACE–NAICS crosswalk and supplemental material prepared by Eurostat.⁶ In cases with more than one NACE code for a given NAICS code, the supplemental material helped BLS match the descriptions of the industry of interest to an appropriate NACE code. Table 1 presents the main industry branches used in this comparison.⁷

Comparison of fatality counts and rates

Number of fatal occupational injuries. In 2010, the total number of fatal work injuries in the United States, as reported by the CFOI, was 4,690. After BLS implemented the data comparability adjustments described in the previous section, the number of fatalities dropped to 2,910, or 62 percent of the original count. About 60 percent of the cases excluded from the original count were self-employed workers; public-sector fatalities and suicides accounted for most of the remaining exclusions.

The comparisons for this study are limited to the main industry branches. The total number of fatal work injuries in these branches was 2,530 for the United States and 3,353 for the European Union. Table 2 shows the count of U.S. and EU fatalities, by NACE code. Notable differences in the data include (1) 19.3 percent of cases occurring in the manufacturing industry in the European Union, compared with 14.9 percent of cases in the United States, and (2) 1.4 percent of cases occurring in the “accommodation and food service activities” industry in the European Union, compared with 4.4 percent of cases in the United States.

Rates of fatal occupational injuries. Because the number of fatalities is influenced by the size of the labor force, both the United States and the European Union publish information on fatality rates. The European Union computes rates based on employment, and the EU rate for the main industry branches for 2010 was 2.8 fatalities per 100,000 employees. By contrast, BLS computes rates based on hours worked, with the intention to provide a closer measure of worker exposure to risk. For several years, BLS calculated both employment-based rates and hours-based rates and found that the results were often similar, although differences were evident among certain groups—such as younger and older workers—that tended not to work full-time schedules.⁸

Again, BLS had to adjust the U.S. data to match the concepts and industry classification used in the EU data. BLS matched the employment-based rates in the EU data by compiling employment data and using the NACE–NAICS crosswalk to develop employment denominators by NACE industry that are consistent with the fatality counts. These data resulted in a U.S. rate for the equivalent of the main industry branches of 3.1 fatalities per 100,000 employees.⁹ Rates for the main industry branches are included in table 2.

More work is needed

While the United States and the European Union were successful in producing a comparison of aggregate counts and rates for fatal work injuries, readers and data users are cautioned not to draw conclusions about relative workplace safety or risk. This comparison must be considered only as a first step in an attempt to identify and understand differences between U.S. and EU fatal work injuries.

For example, although the employment-based fatality rate for the United States in NACE A (Agriculture, forestry and fishing) is nearly twice the rate for the European Union (see table 2), this does not necessarily mean that all work in this industry branch is more hazardous in the United States than in the European Union. For one thing, it is not possible to compare the circumstances surrounding the fatalities. In addition, as noted earlier, the EU data exclude certain types of cases, such as road accidents in the United Kingdom and Ireland.

In light of these considerations, the main purpose of this effort is to show that a comparison is possible. This work begins a conversation and provides an opportunity to identify areas for further comparisons and research. Future work and collaboration will provide better information about fatal workplace hazards and, ideally, lead to improvements in worker safety and health in both the United States and the European Union.

ACKNOWLEDGMENT: The following individuals participated in discussions at the U.S.–EU Conference that led to this analysis: Dawn N. Castillo of the U.S. National Institute for Occupational Safety and Health; Miriam Schoenbaum of the U.S. Occupational Safety and Health Administration; William J. Wiatrowski of the Bureau of Labor Statistics; Malgorzata Stadnik, Ana Martinez Palou, and Antonio Cammarota of the European Commission; and Beate Mayer of the Austrian Workmen’s Compensation Board. Jill Janocha of the Bureau of Labor Statistics and Nivedita Bhushan, formerly of the Bureau of Labor Statistics, developed the data presented here. Bart De Norre of the European Commission provided final review.