Abstract

Objective: To examine whether high job strain (a combination of high job demands and low job control) is a risk factor for disability pension.

Setting: Ten municipalities and 21 hospitals in Finland.

Design and participants: A prospective cohort study of 20 386 female and 4 764 male Finnish public sector employees aged 19–50 using data from two surveys (baseline in 2000–2 and follow-up in 2005) and employers’ registers. In addition to self-reported job strain, we computed work unit-aggregated job strain for each participant (the average of scores of all workers of participant’s work unit except the participant him/herself).

Main results: 93 employees (0.4%) retired because of disability during the follow-up. In multilevel logistic regression analysis adjusted for demographic characteristics and health risk behaviour, odds for disability pension was 2.60 (95% CI 1.26 to 5.34) times higher for employees with high self-assessed job strain than for those with low self-assessed job strain at baseline. The corresponding OR for passive job versus low job strain was 2.82 (95% CI 1.34 to 5.96). Analysis of work unit-aggregated scores replicated the association for high job strain, OR 2.25 (95% CI 1.17 to 4.35), but not that for passive job. The association between work unit job strain and disability pension remained significant after further adjustment for prevalent diseases, psychological distress and perceived health status.

Conclusions: Job strain is associated with risk of subsequent disability pension. If causal, this association suggests that organisational interventions to reduce job strain may also reduce early exit from work.