Is the association between alcohol use and sickness absence modified by socioeconomic position? findings from the Stockholm public health cohort

BackgroundThe distribution of sickness absence tends to be socially patterned less is however known about the underlying mechanisms and pathways of the social gradient found in sickness absence. The present study aims to investigate (i) if the risk function between average volume of alcohol consumption and sickness absence is modified by socio-economic position (SEP), and (ii) whether such an effect modification can be attributed to differences in drinking patterns and other risk factors including other lifestyle behaviours, health status, and working conditions.MethodsThe study was based on data from the Stockholm public health cohort 2006, with an analytical sample of 13 855 respondents aged 18-64 years. Self-reported information on occupational class (a measure of SEP), alcohol consumption, other lifestyle behaviour, health and working conditions was collected from the survey. The outcome of long-term (> 14 days) sickness absence between 2006 and 2008 was obtained from national registers. Negative binomial regression was used to estimate the Incidence Rate Ratios (IRR) with 95% confidence intervals (CI).ResultsIn the initial analyses, heavy drinking manual workers had a 5-fold increased risk of long-term sickness absence compared to non-manual employees who were moderate drinkers, and approximately 60% of the excess risk among heavy drinking manual workers was attributable to an interaction between alcohol use and SEP. Adjusting for working conditions was associated with the largest attenuation of the risk estimate, compared to other lifestyle behaviors and health. In the fully adjusted model, the IRR was further attenuated for the manual workers and the joint effect of SEP and heavy drinking remained in the final model with an attributable proportion of 49%.ConclusionsIndividuals in Sweden with lower levels of SEP appear to be more vulnerable to alcohol consumption in relation to sickness absence, where differences in working conditions explained a large part but not all of the differential vulnerability.