RESUMO
AIMS: The aim of this study was to explore the relationship between leisure-time physical inactivity and long-term sickness absence in a representative sample of individuals aged 16-54 years, within the labour market and in good health. It was hypothesised that physically inactive individuals have a higher risk of long-term sickness absence and longer duration of sickness absence. METHODS: The study population was identified from the National Health and Morbidity Survey, 2010. Weekly data on long-term sickness absence were obtained from the National Register on Social Transfer Payments (the DREAM registry). The association of incidence and duration of long-term sickness absence with physical inactivity was explored using logistic and Poisson regression. Data were fitted to models with levels of physical activity, demographic, social and lifestyle characteristics as independent variables. A combined hurdle model was used to estimate the difference in mean number of absence weeks. RESULTS: Logistic regression showed that physically inactive individuals had a 27% higher incidence of long-term sickness absence compared with physically active individuals. The Poisson regression showed that long-term sickness absence was only slightly shorter (1 week less) for moderately active individuals compared with inactive individuals. The hurdle model estimated longer absence periods for inactive individuals (additional 2.5 weeks) in comparison with moderately and highly active individuals. CONCLUSIONS: The study showed that physically inactive individuals have a higher incidence of long-term absence and that physically inactive individuals have longer periods with sickness absence than moderately and highly active individuals. When adjustments for social and health behaviour were included, the estimated associations became statistically insignificant.