The epidemiology of sick leave was studied in the city of Malmö, Sweden, (230,000 inhabitants). Every current and completed sick-leave episode during the year of 1985 was collected for 124,411 persons aged 16 to 65, who were registered with the National Health insurance scheme in 1985. Absence rate, absence incidence, absence prevalence and absence duration were analyzed in relation to age, sex, marital status, nationality, income and place of residence. Absence rate (mean value) in the total population was 25.5 days with a median of three days. The absence rate increased by age. High absence rates were seen for females, single people and some immigrant groups. This was even true for residential areas characterized by a higher proportion of single-person households and households on social welfare, of unemployed and people with a low income and a foreign background. The absence rate gives limited information as to the epidemiology of sick leave. Through adding absence incidence, absence prevalence and absence duration it was possible to get a more comprehensive picture of the phenomenon. Sex-differences in absence rate for instance were mainly explained by differences in absence incidence and prevalence, while differences in absence rate regarding nationality were explained by differences in absence duration. This is an important step towards a better understanding of the factors behind sick leave.