Starting with the 1992 data year, the National Safety Council (NSC) adopted the Bureau of Labor Statistics’ (BLS) Census of Fatal Occupational Injuries (CFOI) estimate as the authoritative count of work-related deaths. CFOI counts both intentional workplace deaths, like homicides and suicides, as well as preventable deaths. Total deaths and preventable deaths are provided in the fatality trends.
Between 2017 and 2018, preventable work-related deaths increased 2%, from 4,414 to 4,493, while the death rate per 100,000 remained steady at 3.1%. At the same time, total work-related deaths also increased 2%, from 5,147 to 5,250.
Economic trends influence fatality trends because they impact the number of workers and hours worked. Since 2009, the peak of the recession, preventable work-related deaths increased 20%, while the number of workers increased 10%.
Prior to the current rise in fatality trends, preventable work-related fatalities had decreased 24.6% from 1992 to 2009. This decrease occurred even as the number of workers increased 18.4%. In spite of the current increase, preventable work-related deaths are still 9.5% lower than they were in 1992, while the number of workers have increased 30.7%.
- Data Table
- Data Table
Source: Deaths are from CFOI. Employment is from BLS and is based on the Current Population Survey. All other data are NSC estimates.
Deaths include persons of all ages. Workers and death rates include persons 16 and older. Workers are persons 16 years and older who are gainfully employed, including owners, managers, other paid employees, the self-employed, unpaid family workers, and active-duty resident military personnel. Because of adoption of CFOI, deaths and rates from 1992 to present are not comparable to prior years. See the Technical Appendix for additional information.
Starting in 2008, BLS moved from employment-based rates to hours-based rates to measure fatal injury risk per standardized length of exposure, which are generally considered more accurate than employment-based rates. Caution should be used when comparing with rates prior to 2008.