Abstract

Objectives: To study the influence of change in self perceived psychosocial work characteristics on subsequent rates of sickness absence.

Methods: Prospective cohort study of British civil service employees. Job control, job demands, and work social supports were measured in 1985/88 and in 1991/93. Analyses included 3817 British civil servants with sickness absence records at baseline (1985–89) and for two follow up periods, early (1994–95) and later follow up (1996–98).

Results: Change in work characteristics predicted subsequent incidence of long spells of sickness absence (>7 days) in the early follow up period after adjustment for covariates including baseline work characteristics, health status, and sickness absence. Adjusted rate ratios were 1.23 (95% CI 1.03 to 1.46) for decreased compared with stable decision latitude; 1.17 (95% CI 1.01 to 1.36) for increased compared with stable job demands and 0.79 (95% CI 0.67 to 0.93) for increased compared with stable work social support. These associations were also seen in a sub-sample who did not change employment grade. In the later follow up period, associations between work change and long spells of sickness absence were similar for decision latitude, less pronounced for job demands, and no longer apparent for social supports. Changes in work characteristics were not associated with subsequent short spells of sickness absence (⩽7 days).

Conclusions: Adverse changes in the psychosocial work environment may lead to increased rates of sickness absence. These findings suggest that workplace interventions to improve psychosocial working conditions may reduce levels of sickness absence.