Fatal Injuries to Civilian Workers in the United States, 1980-1995  (National Profile)   Suzanne M. Marsh  Larry A. Layne DHHS/NIOSH Pub. No. 2001-129   PDF version

DISCLAIMER

Mention of any company or product does not constitute endorsement by the
National Institute for Occupational Safety and Health

Copies of this and other NIOSH documents are available from:

Publications Dissemination, EID
National Institute for Occupational Safety and Health
4676 Columbia Parkway
Cincinnati, OH 45226-1998

Fax number: (513) 533-8573
Telephone number: 1-800-35-NIOSH (1-800-356-4674)
E-mail: pubstaft@cdc.gov 

For further information about occupational safety and health
topics, call 1-800-35-NIOSH (1-800-356-4674), or
visit the NIOSH Website at www.cdc.gov/niosh.

DHHS (NIOSH) PUBLICATION No. 2001-129

Foreword 

Acknowledgments 

Public Health Summary 

Executive Summary 

Introduction

Methods

Selection Criteria

Why Death Certificates?

Data Utility

Data Coding: Industry/Occupation/Cause of Death 

Calculation of Rates 

Limitations of Death Certificate Data 

Comparison of Data to Previously Published Fatal Occupational Injury Data  

Additional Readings  

References 

National Data  - (All Tables) 

Table US-1. Number and Rate (per 100,000 workers)
    of Traumatic Occupational Fatalities by Year, US, 1980-1995

Table US-2. Number and Average Annual Rate (per 100,000 workers)
    of Traumatic Occupational Fatalities by State of Death, US, 1980-1995

Table US-3. Number and Average Annual Rate (per 100,000 workers)
    of Traumatic Occupational Fatalities by Sex, Race, and Age Group, US, 1980-1995

Table US-4. Number and Rate (per 100,000 workers) of Traumatic Occupational
    Fatalities for Sex, Race, and Age Group, by Year, US, 1980-1995

Table US-5. Number and Average Annual Rate (per 100,000 workers) of Traumatic
    Occupational Fatalities for Race and Age Group, by Sex, US, 1980-1995

Table US-6. Number and Average Annual Rate (per 100,000 workers) of Traumatic
    Occupational Fatalities by Age Group and Race, US, 1980-1995

Table US-7. Number and Average Annual Rate (per 100,000 workers) of Traumatic
    Occupational Fatalities by Cause of Death and Sex, US, 1980-1995

Table US-8. Number of Traumatic Occupational Fatalities
    by Cause of Death and Year, US, 1980-1995

Table US-9. Rate (per 100,000 workers) of Traumatic Occupational Fatalities
    by Cause of Death and Year, US, 1980-1995

Table US-10. Number and Average Annual Rate (per 100,000 workers) of Traumatic
    Occupational Fatalities by Cause of Death and Race, US, 1980-1995

Table US-11. Number of Traumatic Occupational Fatalities
    by Cause of Death and Age Group, US, 1980-1995

Table US-12. Average Annual Rate (per 100,000 workers) of Traumatic
    Occupational Fatalities by Cause of Death and Age Group, US, 1980-1995

Table US-13. Number and Average Annual Rate (per 100,000 workers) of Traumatic
    Occupational Fatalities by Industry Division, US, 1980-1995

Table US-14. Number and Rate (per 100,000 workers) of Traumatic Occupational
    Fatalities by Industry Division and Year, US, 1980-1995

Table US-15. Number of Traumatic Occupational Fatalities
    by Cause of Death and Industry Division, US, 1980-1995

Table US-16. Average Annual Rate (per 100,000 workers) of Traumatic Occupational
    Fatalities by Cause of Death and Industry Division, US, 1980-1995

Table US-17. Number and Average Annual Rate (per 100,000 workers) of Traumatic
    Occupational Fatalities by Industry Division and Age Group, US, 1983-1995

Table US-18. Number and Average Annual Rate (per 100,000 workers) of Traumatic
    Occupational Fatalities by Detailed Industry Groupings, US, 1983-1995

Table US-19. Number of Traumatic Occupational Fatalities
    by Selected Detailed Industry Groupings and Year, US, 1983-1995

Table US-20. Rate (per 100,000 workers) of Traumatic Occupational Fatalities
    by Selected Detailed Industry Groupings and Year, US, 1983-1995

Table US-21. Number of Traumatic Occupational Fatalities
    by Selected Detailed Industry Groupings and Cause of Death, US, 1983-1995

Table US-22. Average Annual Rate (per 100,000 workers) of Traumatic Occupational
    Fatalities by Selected Detailed Industry Groupings and Cause of Death, US,
    1983-1995

Table US-23. Number and Average Annual Rate (per 100,000 workers) of Traumatic
    Occupational Fatalities by Occupation Division, US, 1980-1995

Table US-24. Number and Rate (per 100,000 workers) of Traumatic Occupational
    Fatalities by Occupation Division and Year, US, 1980-1995

Table US-25. Number of Traumatic Occupational Fatalities
    by Cause of Death and Occupation Division, US, 1980-1995

Table US-26. Average Annual Rate (per 100,000 workers) of Traumatic Occupational
    Fatalities by Cause of Death and Occupation Division, US, 1980-1995

Table US-27. Number and Average Annual Rate (per 100,000 workers) of Traumatic
    Occupational Fatalities by Occupation Division and Age Group, US, 1983-1995

Table US-28. Number and Average Annual Rate (per 100,000 workers) of Traumatic
    Occupational Fatalities by Detailed Occupation Groupings, US, 1983-1995

Table US-29. Number and Rate (per 100,000 workers) of Traumatic Occupational
    Fatalities by Selected Detailed Occupation Groupings and Year, US, 1983-1995

Table US-30. Number and Average Annual Rate (per 100,000 workers) of Traumatic
    Occupational Fatalities by Selected Detailed Occupation Groupings and Cause of Death,
    US, 1983-1995

Table US-31. Number and Average Annual Rate (per 100,000 workers) of Traumatic
    Occupational Fatalities by Industry and Occupation Division, US, 1983-1995

Appendices

Appendix I: Abbreviations for Industry and Occupation Divisions 

Appendix II: Detailed Bureau of the Census (BOC) Industry Groupings 

Appendix III: Detailed Bureau of the Census (BOC) Occupation Groupings 

Appendix IV: ICD-9 E-code Rubrics for Cause of Death Categories 

Appendix V: Operational Guidelines for Determination of Injury at Work 

Appendix VI: Additional Readings 

The Occupational Safety and Health Act of 1970 was enacted "to assure so far as possible every working man and woman in the Nation safe and healthful working conditions." However, deaths from injuries at work continue to be a major public health problem. On average, 16 workers die each day in this country. These workers die simply trying to earn a living. Through the National Traumatic Occupational Fatalities surveillance system, NIOSH continues to help fill the gap in the knowledge of traumatic work-related injury deaths. 

This document provides an update to data published in the 1993 publication, Fatal Injuries to Workers in the United States, 1980-1989: A Decade of Surveillance. The current document includes 16 years of data from the National Traumatic Occupational Fatalities surveillance system for the years 1980 through 1995. Occupational injury mortality statistics on over 93,000 deaths are provided by demographic and injury characteristics. These data illuminate the nature and magnitude of work-related injury death for the United States and comprise the most comprehensive summary available in one document. 

Although fatal occupational injuries have decreased over the years, the burden remains high. The data presented in this report provide the basis for developing strategies to prevent traumatic work-related injury deaths by profiling high-risk industries, occupations, and causes of fatal injuries. It is our hope that the information contained in this document will serve as a comprehensive resource for federal, state and local agencies, safety and health professionals, researchers, and others who can affect the prevention of occupational fatalities.

Lawrence J. Fine, M.D., Dr.P.H.
Acting Director, National Institute for
    Occupational Safety and Health 

Acknowledgments

The authors are indebted to Joyce R. Spiker of the Division of Safety Research for all of her efforts in preparing and formatting this document for publication. 

The authors would also like to thank Susan P. Baker, M.P.H., Johns Hopkins Center for Injury Research and Policy; Lois A. Fingerhut, National Center for Health Statistics; James C. Helmkamp, Ph.D., West Virginia University Center for Rural Emergency Medicine; Edward L. Husting, Ph.D., National Center for Injury Prevention and Control; and John W. Ruser, Ph.D., Bureau of Labor Statistics for their helpful comments on this document. 

The authors are also grateful to the State vital registrars and their staffs for providing these valuable data. The success of the National Traumatic Occupational Fatalities surveillance system is due to the help and cooperation of these State vital records offices.

Public Health Summary

Over 93,000 workers were fatally injured while working in the United States from 1980 through 1995. Each day an average of 16 people died—simply by doing their jobs. These deaths result from exposure to many different hazards on the job. Leading causes of traumatic occupational fatalities include motor vehicles, homicides, machines, falls, electrocutions, and falling objects.

Workers are at risk of fatal injury in many different ways. Workers who operate motor vehicles or machines risk injury due to overturns or collisions. Taxicab drivers risk being killed during robbery attempts while construction workers risk fatal falls while working from heights. Electrical linemen risk electrocution while repairing power lines and loggers risk being struck by a falling tree during tree harvesting operations. These scenarios highlight some of the risks that are a daily presence in many of the industries and occupations in the United States. These jobs are frequently noted as being the jobs with the highest fatal injury rates. However, the potential for work-related injury exists in any job where injury risks are present and not controlled.

Several federal agencies promulgate standards and regulations for worker safety. Agencies with specific regulatory authority for worker protection include the Occupational Safety and Health Administration, Federal Railroad Administration, Mine Safety and Health Administration, Federal Highway Administration, and the Employment Standards Administration. NIOSH, however, is the only federal agency responsible for conducting research and making recommendations for the prevention of work-related illnesses and injuries. NIOSH conducts research programs that improve understanding of injury and disease risks, identify populations at risk, and evaluate prevention strategies. For example, prevention recommendations are developed for specific risks through the Fatality Assessment and Control Evaluation program and the Health Hazard Evaluation program. Additional recommendations are developed through the funding of individual research projects, both within NIOSH and with extramural partners. As a result of these efforts NIOSH creates and disseminates scientifically valid recommendations for preventing work-related disease, injury, and disability.

The references and additional readings contained in this document identify sources that provide additional information on fatal occupational injuries. Additional information on worker safety and health may be obtained from NIOSH through

The National Institute for Occupational Safety and Health collects and automates death certificates from the 52 vital statistics reporting units in the 50 States, New York City, and the District of Columbia for workers 16 years of age or older who die as a result of a work-related injury. Analysis of occupational injury deaths, such as those gathered through the National Traumatic Occupational Fatalities (NTOF) surveillance system, facilitates identification of high risk worker groups and potential injury risk factors by demographic, employment, and injury characteristics. This promotes the effective use of resources aimed at preventing injuries in the workplace. In reviewing these data, it is important to note the distinction between the number of deaths and fatality rates. The number of deaths indicates the magnitude of a problem and fatality rates depict the risk faced by workers. Fatal occupational injury data for 1980 through 1995 are provided for the U.S. and for each State.

Major findings from this study:

• There were 93,338 civilian workers who died from injuries sustained while working in the U.S., 1980 through 1995 (Table US-1).

• The average annual occupational fatality rate for the U.S. civilian workforce was 5.2 per 100,000 workers for 1980 through 1995 (Table US-1).

• Civilian fatal occupational injuries decreased 28%, from 7,343 fatalities in 1980 to 5,314 in 1995 (Table US-1).

• The average annual fatality rate per 100,000 civilian workers decreased, from 7.4 in 1980 to 4.3 in 1995 – a 42% decrease (Table US-1).

• The greatest number of fatal occupational injuries occurred in California (9,670), Texas (9,423), Florida (5,596), Illinois (4,169), and Pennsylvania (3,926) (Table US-2).

• The States with the highest occupational injury fatality rates per 100,000 workers were Alaska (24.3), Wyoming (16.7), Montana (12.4), Idaho (10.7), West Virginia (10.4), and Mississippi (10.1) (Table US-2).

• The fatality rate for males (8.8 per 100,000 workers) was 11 times higher than the rate for females (0.8 per 100,000 workers) (Table US-3).

• Eighty-five percent of civilian workers who died were white and 11% were black (Table US-3).

• Black workers had the highest fatality rate per 100,000 workers (5.8), followed by whites (5.1) (Table US-3).

• The age group with the largest number of occupational injury fatalities was the 25-34 year old age group (26%) followed closely by the 35-44 year old age group (22%) (Table US-3).

• Workers 65 years and older had the highest fatality rate of all age groups (13.6 deaths per 100,000 workers) in every industry and occupation division (Tables US-3, US-17, US-27).

• The leading causes of occupational injury death in the U.S. were motor vehicle crashes (23%), homicides (14%), machine-related incidents (13%), falls (10%), electrocutions (7%), and being struck by falling objects (6%) (Table US-7).

• The highest rates by cause of death varied by gender: the highest rate for females was homicide (0.3 per 100,000 workers), while motor vehicle crashes (2.0 per 100,000 workers) were the cause of death with the highest rate among males (Table US-7).

• While the rate of motor vehicle-related fatalities decreased 36% between 1980 and 1995 (from 1.7 per 100,000 workers to 1.1), motor vehicles continued to have the highest rate through 1995. Machines had the second highest rate per 100,000 workers until 1990, when they were surpassed by homicides (Table US-9).

• The highest rates by cause of death varied by race: the highest rate for whites was motor vehicle crashes (1.2 per 100,000 workers), while the highest rate by cause of death for blacks was homicide (1.4 per 100,000 workers) (Table US-10).

• The industry divisions with the greatest proportion of fatalities were construction (18%), transportation/communication/public utilities (17%), manufacturing (15%), and agriculture/forestry/fishing (12%) (Table US-13).

• The mining industry had the highest average annual fatality rate per 100,000 workers (30.4), followed by agriculture/forestry/fishing (19.6), construction (15.3), and transportation/communication/public utilities (12.6) (Table US-13).

• The highest rates by cause of death varied by industry: the highest rate in the agriculture/forestry/fishing industry was for machinery-related incidents (6.6 per 100,000 workers), while the highest rate by cause of death in the retail trade industry was for homicides (1.7) (Table US-16).

• The occupation divisions with the greatest proportion of fatalities were precision production/craft/repairers (21%), transportation/material movers (17%), farmers/foresters/fishers (13%), and laborers (11%) (Table US-23).

• The occupation division of farmers/foresters/fishers had the highest average annual fatality rate per 100,000 workers (21.9), followed by transportation/material movers (21.6), laborers (13.7), and precision production/craft/repairers (9.2) (Table US-23).

• The highest rates by cause of death varied by occupation: the highest rate among executives/administrators/managers was for homicides (0.8 per 100,000 workers), while machinery-related incident rates were highest among farmers/foresters/fishers (7.0) (Table US-26).

The National Institute for Occupational Safety and Health (NIOSH) began collecting death certificates from all 50 States and the District of Columbia in answer to the need for a comprehensive enumeration of workers who sustain a fatal work-related injury. The National Traumatic Occupational Fatalities (NTOF) surveillance system was developed to fill gaps in the knowledge of work-related injury deaths in the United States (U.S.) by providing a census of occupational injury deaths for all U.S. workers.

Surveillance data such as those gathered through NTOF allow analysis of demographic, employment, and injury characteristics and also enable the examination of trends over time. These data allow the description of the nature and magnitude of the occupational injury problem in the U.S., the identification of potential risk factors, the generation of hypotheses for further research, and the setting of research and prevention priorities.

In 1993, a comprehensive summary of fatal occupational injuries in the U.S. was published based on data collected through NTOF for the years 1980 through 1989 (Jenkins et al., 1993). The present document extends the period of analysis to 16 years, providing the most comprehensive summary of fatal occupational injuries in one document for the U.S. as a whole, and for every State. This current document however, is considerably different than the earlier publication with the primary goal of providing detailed data in tabular format. The data are being presented in this format to provide the occupational safety and health community with direct access to data from the NTOF surveillance system. This document will serve as a comprehensive resource to describe the magnitude and circumstances of occupational injury deaths in the U.S. from 1980 through 1995.

The NTOF surveillance system is composed of information obtained from death certificates from the U.S. vital statistics reporting units in the 50 States, New York City, and the District of Columbia. For a case to be included in NTOF, it must meet three criteria:

1. the decedent is 16 years of age or older;

2. the external cause of death is classified as E800-E999 (International Classification of Diseases, Ninth Revision (World Health Organization, 1977)); and

3. the "Injury at Work?" item is marked positive by the certifier.

While studies have shown that multiple source surveillance systems are the best approach in counting occupational fatalities, death certificates were chosen for NTOF because they are the single source that identifies the largest number of cases, and are fairly comparable between all vital statistics reporting units. Studies show that death certificates alone identify between 67% and 90% of all fatal work injuries among the various States (Baker et al., 1982; Karlson and Baker, 1978; Stout and Bell, 1991).

The NTOF system contains 30 variables useful for describing characteristics of victims as well as injury circumstances. Data elements include coded worker characteristics and injury circumstances such as sex, race, occupation, and cause of death. In addition, narrative text for industry, occupation, causes of death, and injury description is entered and maintained. Narrative data have been utilized in a number of focused analyses listed in the Additional Readings section of this document (Appendix VI).

Codes for ‘usual’ industry and occupation were assigned based on the narrative information contained on the death certificates. Two methods of coding were used for the periods 1980 through 1989, and 1990 through 1995. Prior to the availability of an automated coding system, narrative information from the occupation and industry items was manually hand-coded by an expert coder, starting with data year 1990. Earlier years of data had been coded using a crude software program with known problems. The Standardized Occupation and Industry Coding (SOIC) software,¹ a more sophisticated and accurate application released in 1998, was used to recode the data from 1980 through 1989. Cases that could not be assigned a code by the SOIC software were manually hand-coded if the death certificate contained adequate information. The industry and occupation narratives were coded according to the 1980 and 1990 Bureau of the Census classification schemes (Bureau of Census, 1982; Bureau of Census, 1992). Death certificates for which no occupation or industry entry was present or for which the entry was too vague were coded into the "not classified" category. Certificates which had entries such as "housewife" or "student" were also coded into the "not classified" category. Appendix I provides explanations of the abbreviations used in the tables for occupation and industry divisions. Appendices II and III provide codes for detailed industry and occupation groupings included in the tables.

Cause of death codes are based on the International Classification of Diseases, Ninth Revision (ICD-9) supplementary chapter for the classification of external causes of injury and poisoning (World Health Organization, 1977) as assigned by trained nosologists and compiled by the National Center for Health Statistics (NCHS) for the Vital Statistics Mortality data (National Center for Health Statistics, 1980-1995). Codes from this chapter, denoted with a preceding "E," cover the spectrum of unintentional and intentional causes of death, with the actual E-code rubrics of the aggregated categories shown in Appendix IV. E-codes in this report were aggregated in a manner believed to be most beneficial for demonstrating exposures specific to traumatic occupational death. Alternative methods for grouping E-codes have been proposed, including the use of a matrix to present mechanism of injury within the major intent categories of unintentional, intentional, and self-inflicted (Centers for Disease Control and Prevention, 1997). While our rubric does not directly match the intent categories laid out in the suggested matrix framework, the authors believe the categories presented in this report are more beneficial for occupational death analysis and provide for continuity of earlier research. A footnote on Table US-8 provides information that may be used to calculate totals based on major intent categories.

For this analysis, death certificates identified for inclusion in the NTOF data represent a complete count of traumatic occupational fatalities. These data therefore are not subject to sampling error, though they are subject to measurement error such as misclassification or failure to identify positive cases. The tables in this document include cells with counts of three or more deaths. Additionally, data for "Unknown" or "Not Classified" categories are only included for univariate tabulations.

Employment estimates for rate calculations were obtained from the Bureau of Labor Statistics’ (BLS) Current Population Survey (CPS), a population-based household sample survey of the civilian noninstitutionalized population. These data were extracted from the BLS Employment and Earnings and the CPS monthly micro data files (Bureau of Labor Statistics, 1981-1996; Bureau of Labor Statistics, 1992). Because the employment data are based on a sample survey, standard errors are associated with the workforce estimates. However, confidence intervals for fatality rates were not calculated for this publication but are addressed elsewhere.²

Fatality rates were calculated as deaths per 100,000 workers. Rates were not calculated for categories with less than three fatalities or less than 20,000 employees, due to the instability of rates based on small numbers. In addition to this safeguard, considerable caution should be exercised in the interpretation of rates based on less than 20 deaths due to the possibility of rate instability (National Center for Health Statistics, 1999). It should be noted that in some instances rates were calculated for 1983 through 1995 only (e.g., industry division by occupation division), due to the lack of comparably coded denominator data for the earlier period.

The employment data used for rate calculations were based on the number of workers, rather than hours of work (or full-time equivalents). Using the number of workers does not account for the difference in exposure for groups that commonly work less than a full-time schedule of 40 hours per week (e.g., youth and older workers). For most workers however, the injury rates are similar regardless of which type of employment measure is used (Ruser, 1998). In addition, crude rates are presented in this report rather than age-adjusted rates. Age adjustments made when calculating occupational injury death rates, in nearly all cases, have only a negligible effect as age is not the primary risk factor (Bailer et al., 1998). Lastly, there is a dissimilarity between death and employment data when State-specific rates are calculated. The death data indicate the State where the death occurred while the employment data indicate the State of residence. This should be kept in mind when reviewing State fatality rates as net commuter in- and outflows may artificially increase or decrease State-specific rates.

Limitations of death certificates used to ascertain work-related fatality information have previously been described (Bell et al., 1990; Jenkins et al., 1993; Russell and Conroy, 1991; Stout and Bell, 1991). Incomplete or unclear information on the death certificate and the lack of a national standard for the completion of the "Injury at Work?" item on the death certificate during this period are particular problems. Motor vehicle crashes and homicides accentuate the difficulty of attempting to identify occupational injuries (Colorado Dept. of Health, 1988; Russell and Conroy, 1991). Furthermore, because death certificates ask if the injury occurred at work, the death certificate may not be a true measure of work-relatedness for certain causes of death. For instance, a suicide that takes place at work may or may not be associated with work per se.

Standardized guidelines for coding the "Injury at Work?" item on the death certificates were not introduced until 1992.³ Certifier interpretation without the aid of standardized guidelines probably accentuated the problem of false positives being included in the data and false negatives not being identified. The improvements in the quality of the data, such as the sensitivity and positive predictive value, that resulted from these guidelines are unknown.

There are potential discrepancies in the collection of race and ethnicity in death data.⁴ NCHS made the first official recommendation to the States to include separate questions for race and ethnicity on their respective death certificates in January 1987 (Tolson et al., 1991). Variation in the collection methods employed by States is possible prior to the implementation of this recommendation. Additionally, studies have shown that race and ethnicity can be confused, and the manner in which the information is gathered affects the subsequent estimates (Tucker and Kojetin, 1996; Bureau of Census, 1996; Bureau of Census, 1997). As a result of the different approaches to collecting "race" and "ethnicity" information, rates should be interpreted with caution (Hahn, 1992).

Death certificates query for the "usual" occupation and industry of the decedent. Studies comparing the reliability of "usual" occupation and industry reported on death certificates to information derived from personal interviews prior to death reported agreement for "usual" occupation to be 48% to 76%, and 57% to 76% for "usual" industry (Gute and Fulton, 1985; Schade and Swanson, 1988; Schumacher, 1986; Swanson et al., 1984; Turner et al., 1987). Studies comparing death certificate entries for usual occupation and industry to employment information at the time of death reported agreement for occupation to be 64% to 74%, and 60% to 76% for industry (Davis, 1988; Illis et al., 1987; Schade and Swanson, 1988). Additionally, "most recent" occupation and industry were incorrectly entered in the "usual" occupation and industry fields on death certificates in about 6% to 11% of the cases (Davis, 1988; Schade and Swanson, 1988). For these reasons, there exists the possibility that for any surveillance system based on death certificates, cases may be misclassified with respect to industry and occupation.

Finally, this report only includes information on deaths of civilians who died at work in the U.S. Civilians who died at work while abroad, and military personnel, regardless of their duty station, are excluded. The number of U.S. civilian workers killed while abroad is not known. And while NTOF does contain military cases for those who died at work in the U.S., they were excluded from this report primarily because of difficulties in establishing a definition of work-relatedness comparable to that of civilians. Unlike civilians, whose death is generally considered work-related if the incident occurred (1) on an employer’s premise, or (2) off the worksite premises but while the worker was conducting legitimate work duties, active duty military personnel are considered to be on-duty 24 hours per day (Helmkamp and Kennedy, 1996).

The data contained in the current publication are comparable to those from the 1993 document with a few exceptions. In the current report, the industry and occupation narrative data from 1980 through 1989 were recoded, thus frequency counts and rates for these two variables may differ from what was published in the earlier document. Additionally, rates by industry, overall rates by state, and industry rates by state were calculated based on a different source of employment data. The fatality rates by industry, overall rates by state, and industry rates by state included in the earlier publication were calculated using County Business Patterns (CBP), an establishment-based census of employers, supplemented with data from the 1982 Agricultural Census and data for the public administration industry from the CPS. The CBP excludes agricultural production workers, domestic-service workers, railroad workers subject to the Railroad Retirement Act, most government workers, and the self-employed. These exclusions introduce a bias in the calculation of industry-specific incidence rates, resulting in artificially high rates. Because the CPS, which is a population-based survey, includes wage and salaried, self-employed, and all agricultural workers, it best matches the worker population included in NTOF. These differences are discussed in more detail elsewhere (Biddle and Kisner, 1998).

To address some of the limitations of death certificates and other existing data sources in the surveillance of fatal occupational injuries, in 1992 the BLS began collecting national work-related death data through the Census of Fatal Occupational Injuries (CFOI) system. The CFOI requirement that work-relatedness be substantiated by at least two sources has led to improvements in both the number of cases identified and the overall data accuracy. While CFOI and NTOF identified similar patterns from 1992 through 1995, NTOF captured 21,038 cases for this period compared to the 25,455 cases captured by CFOI (Toscano and Windau, 1998). Another difference between the two surveillance systems is that the coding systems used to specify cause of death differ. NTOF uses E-codes from the ICD-9 (World Health Organization, 1977) and CFOI uses the BLS-designed Occupational Injury and Illness Classification System (Toscano et al., 1996). Direct comparisons of the two systems are complicated, but broad results for cause of death are similar.

The last section in this document (Appendix VI) includes a list of articles and other publications that can be referred to for additional information. While not exhaustive, this list includes journal articles, NIOSH publications, and other articles that involved NTOF analyses.

¹The Standardized Occupation and Industry Coding (SOIC) System is a software application developed collaboratively by NIOSH and other agencies and organizations. The software system takes real-world literal descriptions of occupations and industries from death certificates and other documents, and translates them into the 1990 Bureau of the Census occupation and industry codes. Based on a comparison of cases that were coded both by the SOIC software and by hand, it was determined that 87% of the cases matched for both industry and occupation.

²Standard errors associated with the CPS are explained in BLS’ “Employment and Earnings” and “Geographic Profiles of Employment and Unemployment” (Bureau of Labor Statistics, 1981-1996; Bureau of Labor Statistics, 1982-1997).

⁴During the period from 1980 through 1995, categories for race and ethnicity were defined by the Office of Management and Budget’s Statistical Policy Directive 15. Racial and ethnic categories are not to be interpreted as biological or genetic, but are simply a social-political construct designed for collecting data. Race categories include: White, Black, Native American, and Asian/Pacific Islander. Ethnicity categories are: “Hispanic or Latino” and “Not Hispanic or Latino” (Office of Management and Budget, 1977).

Bailer AJ, Stayner LT, Stout NA, Reed LD, Gilbert SJ [1998]. Trends in rates of occupational fatal injuries in the United States (1983-92). Occup Environ Med 55:485-489.

Baker SP, Samkoff JS, Fisher RS, VanBuren CB [1982]. Fatal occupational injuries. JAMA 248: 692-697.

Bell CA, Stout NA, Bender TR, Conroy CS, Crouse WE, Myers JR [1990]. Fatal occupational injuries in the United States, 1980 through 1985. JAMA 236:3047-3050.

Biddle EA, Kisner SM [1998]. Denominator effects on traumatic occupational fatality incidence rates. MetLife Statistical Bulletin 79: 28-36.

Bureau of Census [1982]. 1980 census of population: alphabetic index of industries and occupations. Publication PHC80-R3. Washington, DC: US Department of Commerce, Bureau of Census.

Bureau of Census [1992]. 1990 census of population and housing: alphabetic index of industries and occupations. Publication CPH-R-4. Washington, DC: US Department of Commerce, Bureau of Census.

Bureau of Census [1996]. Findings on questions on race and hispanic origin tested in the 1996 national content survey. Population Division Working Paper No. 16. Washington, DC: US Department of Commerce, Bureau of Census.

Bureau of Census [1997]. Results of the 1996 race and ethnic targeted test. Population Division Working Paper No. 18. Washington, DC: US Department of Commerce, Bureau of Census.

Bureau of Labor Statistics [1981-1996]. Household data annual averages. Employment and earnings. Publications 28-40 (issue no. 1 for each year). Washington, DC: US Department of Labor, Bureau of Labor Statistics.

Bureau of Labor Statistics [1982-1997]. Geographic profile of employment and unemployment. Washington, DC: US Department of Labor, Bureau of Labor Statistics.

Bureau of Labor Statistics [1992 ]. Labor force, employment, and unemployment from the current population survey. In: BLS Handbook of Methods. Washington, DC: US Department of Labor, Bureau of Labor Statistics, pp. 3-12.

Centers for Disease Control and Prevention [1997]. Recommended framework for presenting injury mortality data. MMWR 46 (RR14):1-30.

Colorado Department of Health [1988]. Colorado population-based occupational injury and fatality surveillance system report, 1982-1984. Denver: Health Statistics Section, Colorado Department of Health.

Davis H [1988]. The accuracy of industry data from death certificates for workplace homicide victims. Am J Public Health 78: 1579-1581.

Gute DM, Fulton JP [1985]. Agreement of occupation and industry data on Rhode Island death certificates with two alternative sources of information. Pub Hlth Reports 100(1): 66-72.

Hahn RA [1992]. The state of federal health statistics on racial and ethnic groups. JAMA 267: 268-271.

Helmkamp JC, Kennedy RD [1996]. National mortality profile of active duty personnel in the U.S. armed forces: 1980-1993. (DHHS [NIOSH] Pub No. 96-103). Morgantown, WV: National Institute for Occupational Safety and Health.

Illis WR, Swanson GM, Satariano ER, Schwartz AG [1987]. Summary measures of occupational history. Am J Public Health 77: 1532-1534.

Jenkins EL, Kisner SM, Fosbroke DE, Layne LA, Stout NA, Castillo DN, Cutlip PM, Cianfrocco R [1993]. Fatal injuries to workers in the United States, 1980-1989: A decade of surveillance – National and state profiles. (DHHS [NIOSH] Pub No. 93-108S). Morgantown, WV: National Institute for Occupational Safety and Health.

Karlson TA, Baker SP [1978]. Fatal occupational injuries associated with motor vehicles. In: Proceedings of the 22nd Conference of the American Association for Automotive Medicine, Vol. 1. Arlington Heights, IL: American Association for Automotive Medicine: pp. 229-241.

National Center for Health Statistics [1999]. Vital statistics of United States: Mortality, 1995 Technical Appendix . Centers for Disease Control and Prevention: National Center for Health Statistics.

National Center for Health Statistics [1980-1995]. Vital statistics mortality public use data. Centers for Disease Control and Prevention: National Center for Health Statistics.

Office of Management and Budget [1977]. Race and ethnic standards for federal statistics and administrative reporting. Statistical Policy Directive No. 15. Washington, DC: Office of Management and Budget.

Ruser JW [1998]. Denominator choice in the calculation of workplace fatality rates. In: Fatal Workplace Injuries in 1996: A Collection of Data and Analysis. Report 922. Washington, DC: US Department of Labor, Bureau of Labor Statistics.

Russell J, Conroy C [1991]. Representativeness of deaths identified through the injury-at-work item on the death certificate: implications for surveillance. Am J Public Health 81: 1613-1618.

Schade WJ, Swanson GM [1988]. Comparison of death certificate occupation and industry data with life-time occupational histories obtained by interview. Am J Ind Med 14: 121-136.

Schumacher MC [1986]. Comparison of occupation and industry information from death certificates and interviews. Am J Pub Hlth 76(6): 635-637.

Stout NA, Bell CA [1991]. Effectiveness of source documents for identifying fatal occupational injuries: a synthesis of studies. Am J Public Health 81: 725-728.

Swanson GM, Schwartz AG, Burrows RW [1984]. An assessment of occupation and industry data from death certificates and hospital medical records for population-based cancer surveillance. Am J Pub Hlth 74(5): 464-467.

Tolson GC, Barnes JM, Hay GA, Kowaleski JL [1991]. The 1989 revision of the U.S. standard certificates and reports. Vital Health Stat 4(28). Washington, DC: US Department of Health and Human Services, National Center for Health Statistics.

Toscano G, Windau J [1998]. Profile of fatal work injuries in 1996. In: Fatal Workplace Injuries in 1996: A Collection of Data and Analysis. Report 922. Washington, DC: US Department of Labor, Bureau of Labor Statistics.

Toscano G, Windau J, Drudi D [1996]. Using the BLS occupational injury and illness classification system as a safety and health management tool. In: Fatal Workplace Injuries in 1994: A Collection of Data and Analysis. Report 908. Washington, DC: US Department of Labor, Bureau of Labor Statistics.

Tucker GC, Kojetin B [1996]. Testing racial and ethnic origin questions in the CPS supplement. In: Monthly Labor Review, September 1996. Washington, DC: US Department of Labor, Bureau of Labor Statistics.

Turner DW, Schumacher MC, West DW [1987]. Comparison of occupational interview data to death certificate data in Utah. Am J Ind Med 12: 145-151.

World Health Organization [1977]. International classification of diseases: manual on the international statistical classification of diseases, injuries, and causes of death. 9th Rev. Geneva, Switzerland.

List of Data Tables

List of Appendices

Back to Top 

NIOSH Home