Biomechanical and psychosocial occupational exposures: joint predictors of post-retirement functional health in the French GAZEL cohort.

Abstract

Biomechanical and psychosocial occupational exposures are seldom considered simultaneously and over extended follow-up in occupational epidemiologic studies, although there is some evidence that combined exposures have interactive effects on workers' health during working life. Given high prevalence of functional disability among retirees, research on earlier-life determinants of subsequent functional outcomes can help shape workplace policies and practices. This study investigates whether health effects of combined occupational exposures during working life are observed after individuals retire and are no longer exposed. Analyses were conducted among retirees in the French GAZEL occupational cohort (n=9168). Cumulative exposure during working life to eight biomechanical strains and to one or more reports of psychosocial job strain (high-demand, low-control work) were assessed as predictors of three outcomes: difficulty with physical functioning, role limitations due to physical difficulties, and bodily pain. Individuals were classified by joint exposure to both biomechanical and psychosocial constraints. We modeled risk ratios (RR) between exposure to biomechanical and psychosocial factors at work (separately and in combination) and disability after retirement, and we calculated the relative excess risk due to interaction (RERI) to test whether combined effects departed from additivity. Both psychosocial and biomechanical exposures during working life were independent predictors of the three functional health outcomes. Compared with individuals who had neither biomechanical nor psychosocial exposures, in fully adjusted log-binomial models of the combined effects of biomechanical and psychosocial exposure, among those with low biomechanical exposures, the RR for physical functioning difficulties associated with psychosocial exposures was 1.18 (95% CI 1.01, 1.37). Among those with the highest levels of biomechanical exposures, RR was 1.42 (95% CI 1.21, 1.65) among those with low psychosocial exposures and 1.91 (95% CI 1.61, 2.26) among those with high psychosocial exposures. The two exposure types were modestly super-additive, with an RERI of 0.32 (95% CI 0.00, 0.62) between those with the highest and lowest levels of biomechanical exposures; if the effects were strictly additive, we would have expected an RERI of 0. For the other two outcomes results were similar, although there was no significant departure from additivity. Some effects varied in magnitude by gender. Across the sample, combined biomechanical and psychosocial occupational exposures during working life appear to have additive or perhaps interactive effects on functional health in retirement. However, the relationship is less straightforward among women. Improving overall working conditions may lessen future disability among retirees.