Abstract
This study aims to investigate the relationship between education and alcohol-related morbidity and the role that low job control and heavy physical workload play in explaining these associations among men and women in Sweden. This register-based cohort study (SWIP cohort) includes over three million individuals registered in Sweden in 2005. Job control and physical workload were measured using a job exposure matrix linked to the index person based on their registered occupation at baseline. Alcohol-related morbidity was measured through diagnoses in the national patient registers between 2006 and 2020. Cox proportional hazards regression models were built to estimate associations between education and alcohol-related morbidity. Reductions in hazard ratios (HRs) were calculated after adjusting for job control, physical workload, and other covariates. Models were also stratified by sex. Lower levels of education predicted a higher risk of alcohol-related morbidity (HR: 2.55 95% confidence interval: 2.49-2.62 for the lowest educated compared to the highest). Low job control and heavy physical workload both played roles in explaining educational differences in alcohol-related morbidity even after accounting for sociodemographic and health factors (15.1% attenuation for job control and 18.3% for physical workload among the lowest educated). Physical workload explained a larger proportion of the associations among men compared to women. Lower levels of education were associated with an increased risk of alcohol-related morbidity and working conditions partly explained these associations beyond what was explained by sociodemographic and health factors. Improving working conditions could therefore prevent some cases of alcohol-related morbidity.