Study objective

To investigate the role of social supports, social networks, and chronic stressors : (i) as predictors of sickness absence ; and (ii) as potential explanations for the socioeconomic gradient in sickness absence.

Design

A prospective cohort study (Whitehall II study) with sociodemographic factors, health and social support measured at baseline, and spells of sickness absence measured prospectively.

Setting

Twenty London based non-industrial departments of the British civil service.

Participants

Participants were civil servants (n=10 308), aged 35-55 years at baseline, of whom 67% (6895) were men and 33% (3413) were women.

The overall response rate for Whitehall II was 73% (74% for men and 71% for women).

The analysis is based on 41% of the sample who had data on reasons for sickness absence and were administered all social support questions.

Only 4.3% of participants did not complete all necessary questions and were excluded.

Measurements and main results

High levels of confiding/emotional support from the « closest person » predicted higher levels of both short and long spells of sickness absence.

After adjusting for baseline physical and psychological health the effects were increased, suggesting that high levels of confiding/emotional support may encourage illness behaviour rather than generate illness.