How Far to the ER? Interpreting Work Injury Fatality Rates

How Far to the ER? Interpreting Work Injury Fatality Rates
by: Craig Radden Henderson, BLS Program Supervisor

“More important than using the fatality rate for state comparisons...individual states [can] gauge changes over time with the goal of making steady progress toward the reduction of work injury fatalities in all industries.”

Regional analysis of standard economic and statistical measures is important to employers and policy makers who want to gain insight into the broader context affecting workplace safety and health. Work injury fatality rates would seem to be a useful assessment measure, especially as safety and security issues are given higher priority in business and individual decision making. A state’s industry and occupational distributions, relative dependence on nonresident workers (e.g., interstate commuting patterns), prevalent types of fatal work injuries, and the distances of workplaces from urban trauma centers (whether located in-state or out-of-state), all provide the context for and impact the usefulness of statewide fatality statistics as a stand-alone measure of workplace health.

In comparison with other states, Wyoming’s historically significant but evolving relationship with both Agriculture and Mining (particularly oil & gas extraction), small resident workforce, the comparatively greater distances to cities with hospitals or out-of-state metropolitan areas with specialized trauma centers, and the state’s heavy dependence on highway transport for the provision of goods and services all contribute to placing Wyoming among states with the highest work injury fatality rates. Yet, in 2000 and 2001, with 36 and 40 work injury fatalities, respectively, do Wyoming and many other “high rate” states deserve any notoriety that may be attributed to this statistical measure?¹ Probably not.

Fatality Statistics Program

The Census of Fatal Occupational Injuries (CFOI) tracks workplace fatalities and primarily relies on “information found on death certificates, newspaper articles, and workers’ compensation reports.”² In unique circumstances employers may be contacted by mail for additional information related to a work fatality. All individually-identifiable information is kept confidential.

U.S. Fatalities in 2001

Nationwide, the annual counts of fatal work injuries and occupational fatality rates declined slowly between 1994 and 2001 (see Figure 1). After peaking in 1994 at 6,632, the number of fatalities gradually fell to 5,900 in 2001. Data for 2001 exclude the 2,886 work-related fatalities resulting from the events of September 11th, which the CFOI program reports separately.³ The fatality rate per 100,000 employed workers (a measure of the number of civilian worker fatalities, age 16 and older, divided by the number of employed civilians, age 16 and older) fell from 5.3 in 1994 to 4.3 in 2001.

2001 U.S. Fatalities by Industry, Occupation, and Type of Incident

In 2001, Construction continued to report the largest number of fatalities of any industry, reaching the highest reported level since the fatality census began in 1992. “From 2000 to 2001, decreases in [Construction] fatalities from transportation incidents and job-related homicides were offset by increases in fatalities from falls and electrocutions.”⁴

While Construction fatalities showed an annual increase of 6.0 percent in 2001, Manufacturing fatalities decreased by 10.0 percent to the lowest level in 10 years. Transportation, Communications, & Public Utilities (TCPU), Wholesale Trade, and Retail Trade also showed decreases between 2000 and 2001. The number of fatalities in Services remained unchanged, while all other industries showed increases. Excluding September 11th, fatalities in Government (including police, detectives, and aircraft pilots) increased by 10.0 percent.⁵

The highest occupational fatality rates in 2001 were reported in Mining; Agriculture; forestry and fishing; Construction; and Transportation. For the second year in a row, Mining (including oil & gas extraction) reported the highest fatality rate, 30.0 per 100,000 employed.

Occupations with the largest number of fatalities in 2001 (see Figure 2) included farm workers (499), construction laborers (349), timber cutters (92), airplane pilots (87), and roofers (78). The highest occupational fatality rates (per 100,000 employed) occurred among fishers (151.2), timber cutters (127.8), and mining machine operators (109.7).

For the third consecutive year, total fatalities resulting from transportation incidents decreased, from 2,573 in 2000 to 2,517 in 2001.⁶ However, the subcategory of highway incidents increased by 3.0 percent, remaining the leading cause of work injury fatalities. Non-highway fatal incidents (including tractor or forklift overturns) fell to their lowest level since the census began.⁷

Work-related homicides (excluding fatalities resulting from September 11th) also fell to a record-low level since 1992, down among technical, sales, and administrative workers though increasing significantly for workers in several services occupations (i.e., police, detectives, food preparation workers, barbers, and hairdressers).⁸ Work-related suicides increased in 2001 as did fatal assaults by animals.

State Fatality/Employment Rates

One important consideration in interpreting state fatality rates is that fatal work injuries (the numerator in computing the rate) are based on place of work injury, but employment (the denominator) is based on state resident employment as measured by the Current Population Survey (CPS), a survey of households. The same formula is used for calculating national fatality rates. While workers often live and work in the same state, a substantial number live and work in different states. Consequently, the use of an employment statistic based exclusively on state residency makes comparisons of state data problematic. For example, Table 1 shows that several New England and Mid-Atlantic states rank among states having comparatively low rates of work injury fatalities in 2001 [e.g., New Hampshire (1.4), Connecticut (2.4), and Delaware (2.5)]. In general, these states are urban, have fewer miles of open highway, are situated in close proximity to large cities along the Boston to Washington, D.C. metropolitan corridor, and presumably have many individuals who work out-of-state (see Map). Therefore, fatalities among these nonresident workers would be counted in states having greater resident workforces (e.g., Massachusetts, New York, Pennsylvania) if the nonresidents were injured while at work (not as a result of commuting). Higher resident workforces tend to dilute a state’s work injury fatality rate. On the other hand, higher numbers of nonresident work injury fatalities would drive up rates in states with smaller resident workforces. Two industries which contribute significantly to high fatality rates nationwide, Mining (particularly oil & gas extraction) and Agriculture (excluding fishing and forestry), are not as significantly represented in the workforces of many of these more economically diversified states in the Northeast.

At the other end of the scale (see Table 1 and Map), many states in the Rocky Mountains, Great Plains, and along the lower Mississippi River are part of or on the periphery of what commonly is referred to as the grain belt or the oil & gas belt. The Map illustrates a largely contiguous swath of the highest fatality rates occurring in states where Agriculture or oil & gas extraction prevail, and often in combination with extensive open-road trucking routes. These trucking networks connect urban hubs in the East or Central states with other urban areas in the South and West (e.g., Atlanta to Houston; Chicago to Los Angeles; or Minneapolis to Seattle), and from the Canadian to the Mexican borders. Untold by the geographical pattern, presumably a significant number of truck drivers who die of work injuries in generally rural states are residents of the most populous states (e.g., California, Texas, New York, and Florida). In other words, we cannot make universal assumptions that worker fatalities are a reflection of resident employer or employee safety in the states where the work injuries occur. For example, if a Californian truck driver suffers a work injury death in Wyoming, Wyoming counts the death in their CFOI reported data (i.e., the numerator). The worker is not counted as a resident worker of Wyoming (i.e., not in the denominator), though, so Wyoming’s fatal work injury rate appears higher per number of resident workers than it actually is. Weather conditions, employees working independently (removed from assistance if injured), distance from an emergency room or trauma center, or a host of other factors can also increase the incidence of a fatality.

Table 2 compares work injury fatality rates for Wyoming and its bordering states for 2001 and for the period 1996-2000. Although Wyoming shows an all industry fatality rate of 14.9 per 100,000 employed for 2001, the average rate (11.9) is considerably lower for the previous five-year period. Only Construction and TCPU had statistically reportable industry fatality rates (each industry at 31.2). The dashes in Table 2 show that other Wyoming industries each had fewer than five incidents of work injury fatality in 2001; therefore no fatality rates are computed for these industries.

South Dakota, Nebraska, and Utah’s fatality rates in 2001 were consistent with past five-year trends (8.4, 6.3, and 6.0 per 100,000 employed, respectively). South Dakota exceeded four injuries only among industry workers in Agriculture; fatality rates for other industries are not reported for that state. Interestingly, neither Wyoming nor any of its border states have reportable Mining fatality rates for 2001 because they experienced only four or fewer incidents. Finance, Insurance, & Real Estate (FIRE) also reported few recent incidents of fatal work injury in Wyoming and most bordering states. The exception is Montana, which reports a fatality rate of 6.8 in FIRE for 1996-2000. A single case of an airplane crash, a fire at a large business, or a major event such as the bombing of the Oklahoma City federal building in 1994 can redirect a state’s statistical trend line.

Like Wyoming, in 2001 Colorado’s work injury fatality rate of 6.2 exceeded its previous five-year rate of 4.6 per 100,000 employed. Montana’s 2001 rate of 12.6 also exceeded its five-year rate of 11.1. In contrast, Idaho’s work injury fatality rate decreased from 7.8 to 6.8. A slight variance in the number of fatal incidents for any industry can have a large effect on fatality rates of a small state.

States with larger resident workforces that report a significant number of industry-specific work injury fatalities but show no sharp rise in overall fatalities may, in fact, be shielded from the focus commonly associated with interstate comparisons, specifically the higher work injury fatality rates of small states like Wyoming or Alaska. Illinois or Texas have many miles of highly traveled open highways that serve as economic crossroads, as well as significant numbers of workers in Agriculture or oil & gas extraction employment. So, perhaps more important than using the fatality rate for state comparisons, data can be used by individual states to gauge changes over time with the goal of making steady progress toward the reduction of work injury fatalities in all industries.

The CFOI program has been instrumental in documenting the incidents including the causes of work injury fatalities in the U.S. for 10 years. The data collection has resulted in policy changes to improve occupational workplace safety and health. For example, researchers have used the data to study and bring attention to work environment issues related to injuries and illnesses involving electrical accidents.⁹ Others have studied fatality rates and illnesses and injuries in small businesses (fewer than 100 employees) to identify high-risk industries for occupational safety and health interventions.¹⁰ By studying documented work injury fatalities, the CFOI program aims to advance worker safety in general, without putting emphasis on extenuating factors (e.g., distance from work site to emergency room) that can lead to misinterpretations of the data, especially when making geographic comparisons. While the CFOI data are important in identifying causes of worker fatalities, they should be interpreted within the context of how rates are calculated.

The 2002 national work injury data are expected to be released by the U.S. Bureau of Labor Statistics in Fall 2003. R&P plans to publish updated state data in Wyoming Labor Force Trends as they become available.

¹U.S. Department of Labor, Bureau of Labor Statistics, “National Census of Fatal Occupational Injuries in 2001,” News, September 25, 2002, <http://stats.bls.gov/iif/oshcfoi1.htm> (August 6, 2003), p. 13.

²Krista L. Gerth, “Occupational Injuries and Illnesses: How Safe Are Wyoming’s Workplaces?” Wyoming Labor Force Trends, February 2002, p. 1.

³U.S. Bureau of Labor Statistics, Census of Fatal Occupational Injuries Summary, September 25, 2002, <http://www.bls.gov/news.release/cfoi.nr0.htm> (August 6, 2003), pp. 1-5.

⁴U.S. Bureau of Labor Statistics, p. 2.

⁵U.S. Bureau of Labor Statistics, p. 3.

⁶U.S. Bureau of Labor Statistics, p. 2.

⁷Ibid.

⁸Ibid.

⁹James C. Cawley, Abstract of “U.S. occupational electrical incidents, 1992-1998,” Journal of Safety Research, 32:3, November 1, 2001, p. 359.

¹⁰Andrea Okun, et al., “Identifying High-Risk Small Business Industries for Occupational Safety and Health Interventions,” American Journal of Industrial Medicine, 39:3, March 2001, pp. 301-311.

Table of Contents | Labor Market Information | Wyoming Job Network | Send Us Mail

These pages designed by Julie Barnish.
Last modified on by Susan J. Murray.