Objective: This study examined the associations between unemployment and alcohol-related hospitalization or mortality and to what extent these associations may be confounded by alcohol consumption and alcohol problems before unemployment.
Method: The study was based on the Stockholm Public Health Cohort (SPHC), a population-based stratified random sample with a baseline questionnaire in 2002/2003 and record linkages up to year 2011. The final sample in the study consists of 15,841 people aged 18-60 years. Unemployment was defined as any registration at the public employment services during 2003-2005. The outcome was alcohol-related hospitalization and alcohol-related mortality during 2006-2011. Confounders were age, sex, and education, and we further adjusted for baseline alcohol consumption and alcohol-related hospitalization before the study period. Cox proportional hazard models were fit, and associations were expressed as hazard ratios (HRs).
Results: In the fully adjusted model, unemployment was associated with an increased risk of alcohol-related hospitalization or mortality, with a more than threefold hazard (HR = 3.38, 95% CI [1.81, 6.31]) compared with no unemployment during the exposure period. There was a moderate attenuating effect of prior alcohol consumption and alcohol-related hospitalization.
Conclusions: Any unemployment in 2003-2005 was highly related to having experienced an alcohol-related diagnosis during the 6-year follow-up, even after controlling for risky use of alcohol and prior hospitalization.