Alcohol consumption and sickness absence: evidence from microdata.

BACKGROUND Aggregate time-series evidence has shown that overall per capita alcohol consumption is associated with sickness absence. This study re-examines the relationship between alcohol consumption and sickness absence by using individual-level microdata and methods that yield results which are less likely to be due to spurious correlations. METHODS Data on sickness absence and alcohol consumption for 18 Finnish regions over the period 1993-2005 was used. Sickness absence was measured as the number of sickness absence days during 1 year. Alcohol consumption was measured as the number of alcohol drinks consumed per week. The individual-level relationship between alcohol consumption and sickness absence was estimated by using Poisson regression models. Unobserved determinants of lifestyle behaviours associated with the region and survey year were controlled for. Personal characteristics as well as the clustering of observations by regions were also taken into account. RESULTS The estimates show that alcohol consumption is associated with sickness absence. The positive relationship between alcohol consumption and sickness absence is particularly pronounced for low-educated males. CONCLUSIONS Aggregate time-series evidence for the relationship between alcohol consumption and sickness absence is confirmed by using individual-level microdata. The policy lesson is that it is important to take into account the effects of alcohol consumption on the prevalence of sickness absence (i.e. labour supply on an intensive margin) when one is considering the level of taxation of alcoholic beverages.