To describe incident and recurrent work-related back injuries among union carpenters, describe the hazard function for each and associated risk factors, and explore predictors of subsequent musculoskeletal back injury based on different definitions of the initial injury.Methods:
This study identified a dynamic cohort of 18 768 carpenters who worked in the State of Washington 1989–2003, their hours worked each month, and their work-related back injuries and medical claims for treatment including ICD-9 codes. Using Poisson regression we calculated rates and rate ratios (RRs) of incident and recurrent injury adjusting for age, gender, union tenure and type of carpentry work. Predictors of subsequent musculoskeletal back injury were explored based on different definitions of the incident injury, as were time periods of greatest risk following return to work.Results:
Recurrent back injuries occurred at a rate 80% higher than initial injuries. Survival curves were significantly different for incident and recurrent injuries, but patterns of relative risk were similar. Individuals with greatest union tenure were at lowest risk, likely reflecting a healthy worker effect or lower physical exposures with seniority. Individuals with long periods of work disability with their first injury were at particularly high risk of subsequent musculoskeletal injury compared with those with no prior history (RR 2.3; 95% CI 2.0 to 2.7), as were individuals with degenerative diagnoses (RR 2.0; 95% CI 1.5 to 2.6). Risk for second injury peaked between 1000 and 1500 h after return to work and then gradually declined.Conclusions:
Carpenters with long periods of work disability following back injury warrant accommodation and perhaps better rehabilitation efforts to avoid re-injury. Challenges to workplace accommodation and limited ability to clearly define readiness to return to work following injury demonstrate the need for primary prevention of back injuries through attention to engineering solutions among carpenters involved in strenuous work.