![[ Visit Client Website ]](images/banner.jpg)
The Bureau of Labor Statistics (BLS) has conducted the Survey of Occupational Injuries and Illnesses (SOII) annually since the early 1970s. The SOII reports on the number and rate of OSHA-recordable workplace injuries and illnesses by detailed and State, with additional detail on worker characteristics and case circumstances for cases involving days away from work. Several studies in the mid-2000s matched SOII and workers’ compensation data and expressed concern that the SOII was incomplete. In 2009, Congress appropriated funds to conduct research into the completeness of the SOII.
METHODS:
BLS worked internally and with State partners and a contractor on three research projects: additional matching of SOII and workers’ compensation data; interviews with employers about recording OSHA and workers’ compensation cases; and a multisource enumeration of work-related amputations and carpal tunnel syndrome. These projects began in 2010 and were completed in 2012.
RESULTS:
The research suggests that the SOII undercounts some injuries and illnesses and that the magnitude of the estimated undercount varies based on research methods. The magnitude may differ by State, and it is unknown whether the undercount changes over time or in response to regulatory changes affecting the SOII and workers' compensation.
The multisource enumeration of workplace injuries and illnesses proved impractical on a national scale. Technical challenges in matching across data sets, resource requirements, and variations between data available to each state limit the feasibility of this approach.
The employer interviews were an invaluable tool for understanding some of the results of the data matching, even if their observations were qualitative in nature.
The research did not identify a prominent cause of the undercount, but identified factors that may contribute. Some factors, such as a reduced likelihood to capture injuries or illnesses incurred late in the reporting cycle, appear within BLS’ ability to affect change. Others remain beyond the jurisdiction of BLS, such as those related to OSHA recordkeeping.
CONCLUSIONS:
The results and recommendations of this first phase of SOII research were used to guide the selection of further research in 2013 to improve the completeness and accuracy of estimates from the SOII. Research topics include interviews with a large sample of employers to obtain statistically significant results on various recordkeeping practices; match of twelve years’ worth of SOII data to workers’ compensation records in one state to analyze any undercount trends over time; and pilot test of automated coding of SOII narrative text fields to improve classification consistency.