OBJECTIVES:In many studies, obesity has been associated with morbidity or mortality, but only a few have studied the relation between obesity and sick leave. Our aim is to analyse the independent effects of both adiposity and body fat distribution in relation to the 1-y incidence of sick leave in a large cohort of male and female workers covering a variety of occupations, taking into account a wide range of socio-demographic, behavioural and bioclinical variables. DESIGN AND SETTING: The baseline survey of the Belstress study was conducted in 25 companies across Belgium between 1994 and 1998. A cohort of 20 463 workers (15 557 males and 4906 females) aged 35-59 y was followed for absenteeism during 1 y. The 75th percentile of the distribution of the total annual sickness days was used as a cutoff to classify the workers with a high 1-y incidence rate of sick leave. The relation between sick leave and both obesity and body fat distribution assessed by the body mass index (BMI) and the waist circumference, respectively, was analysed by multivariate logistic regression models. RESULTS: Using a backward procedure based on the likelihood ratio, we found central abdominal fatness to be an independent predictor of sick leave in both genders (high sick-leave incidence and long spells), but not BMI. In men, the odds ratios was 1.31 (99% CI 1.12-1.52, Po0.0001) and in women it ranged from 1.32 (99% CI 1.03-1.70, P ¼ 0.005) to 1.47 (99% CI 1.14-1.90, Po0.0001). Two baseline covariables, respiratory problems and perceived health, are confounders or mediators. CONCLUSIONS: In this study, body fat distribution was associated with a high annual sick-leave incidence and long spells of sickness absence. If this link is reversible, employers may benefit from programs aiming at the prevention and treatment of central obesity.