Research: Underreporting of workplace injuries

A recurring issue with injury statistics is that they tend to be derived from workers’ compensation board claims data and thus under-report the true level of workplace injury. A new article in Occupational and Environmental Medicine entitled “Characteristics of work-related fatal and hospitalised injuries not captured in workers’ compensation data” examines this phenomenon and the patterns within it as well as suggesting a way to cope with it. The study is open access so you can read it by clicking on the link above.

The study found 83% of fatalities (in coroners’ data) and 95% of work-related hospitalizations were captured by workers’ compensation data. This high rate of capture of very serious injuries in workers’ comp-derived stats is not surprising—it is hard to hide a body or convince someone who needs hospitalization to not report an injury.

The study does not engage with the under-reporting of injuries not requiring hospitalization. Injuries that workers can work through (despite some degree of pain) comprise the majority of workplace injuries. These less severe injuries often do not have to be reported and, when they do require reporting, workers can often be persuaded not to report them. Good data on the degree of under-reporting would be helpful—that last really good study on this that I can recall was in about 2003 and didn’t necessaryly tease this issue out.

The study also notes that injuries missing from workers’ compensation claims data tended to be to women, from the fishing/farming industry and injuries to the airway, dislocations and undetermined/unknown injuries. The exclusion of injuries to women is interesting and cries out for more study. It parallels some research my colleague and I are doing around media reports of injuries (where women are basically invisible).

-- Bob Barnetson