Objectives To examine depression as a potential negative health effect of long work hours. An exposure-response relationship was anticipated.

Methods A prospective cohort study of 4520 Danish senior medical consultants was conducted. Average weekly work hours >40 defined long work hours. Redemption of antidepressive drug prescriptions defined depression. Control variables were gender, age, marital status, medical specialty, decision authority at work, work social support, quantitative work demands and previous redemption of antidepressive drug prescriptions. By means of Proportional Hazards Cox Regression Analyses adjusted hazard ratios (HR) were computed.

Results Among Danish senior medical consultants long weekly work hours were associated with a reduced adjusted risk of redeeming antidepressive drug prescriptions at all times during follow-up compared to reference work hours of 37–40 (41–44 h: HR 0.95, CI 0.5 to 1.8; 45–49 h: HR 0.88, CI 0.4 to 1.8; 50–54 h: HR 0.83, CI 0.3 to 2.1; 55–59 h: HR 0.67, CI 0.2 to 2.9; >60 h: HR 0.48, CI 0.1 to 3.7).

Conclusions Long work hours decrease the risk of developing depression when measured as redemption of antidepressive drug prescriptions among Danish senior medical consultants. Whether this is explained by the healthy worker effect, an unexpected protective effect of long work hours or chance remains to be explored. Study strengths and limitations should be taken into account.