Contribution of working conditions to occupational inequalities in depressive symptoms: results from the national French SUMER survey.

Objectives: Social inequalities in mental health have been observed, but explanations are still lacking. The objectives were to evaluate the contribution of a large set of psychosocial work factors and other occupational exposures to social inequalities in mental health in a national representative sample of employees.

Methods: The sample from the cross-sectional national French survey SUMER 2010 included 46,962 employees: 26,883 men and 20,079 women. Anxiety and depression symptoms were measured using the Hospital Anxiety and Depression scale. Occupation was used as a marker of social position. Psychosocial work factors included various variables related to the classical job strain model, psychological demands, decision latitude, social support, and other understudied variables related to reward, job insecurity, job promotion, esteem, working time/hours, and workplace violence. Other occupational exposures of chemical, biological, physical, and biomechanical nature were also studied. Weighted age-adjusted linear regression analyses were performed.

Results: Occupational gradients were found in the exposure to most psychosocial work factors and other occupational exposures. Occupational inequalities were observed for depressive symptoms, but not for anxiety symptoms. The factors related to decision latitude (and its sub-dimensions, skill discretion, and decision authority), social support, and reward (and its sub-dimensions, job promotion, job insecurity, and esteem) contributed to explain occupational inequalities in depressive symptoms. Decision latitude played a major role in the explanation. Workplace violence variables contributed among men only. Other exposures of physical and biomechanical nature also displayed significant contributions.

Conclusions: Comprehensive prevention policies at the workplace may help to reduce social inequalities in mental health in the working population.