AIM: The purpose of the study was to assess the relationship between aggregate alcohol consumption and sickness absence in Sweden. DATA AND METHODS: Two indicators of sickness absence were used, one based on sickness insurance data, the other on data from the labour force surveys. Alcohol consumption was gauged by sales of pure alcohol (100%) per inhabitant 15 years of age and older. Because changes in the economy may affect alcohol consumption as well as sickness absence, two macroeconomic indicators were included as control variables: unemployment and real wages. The study period was 1935-2002. The data were analysed through the Box-Jenkins method for time-series analyses. FINDINGS: A 1-litre increase in total consumption was associated with a 13% increase in sickness absence among men (P