Socioeconomic status as an effect modifier of alcohol consumption and harm: an analysis of linked cohort data

Background: Alcohol-related mortality and morbidity are higher in socioeconomically disadvantaged populations. It is unclear if elevated harm reflects differences in consumption, reverse causation or greater risk of harm following similar consumption. We investigated whether the harmful effects differed by socioeconomic status accounting for alcohol consumption and other health-related factors. Methods: Alcohol consumption (weekly units and binge drinking) data (n=50,236; 429,986 person-years of follow-up) were linked to deaths, hospitalisations and prescriptions. The primary outcome was alcohol-attributable hospitalisation/death. The relationship between alcohol attributable harm and socioeconomic status was investigated for four measures (education level, social class, household income and area-based deprivation) using Cox proportional hazards models. The potential for alcohol consumption and other risk factors mediating the social patterning was explored. Downward social selection for high-risk drinkers (reverse causation) was tested by comparing change in area deprivation over time. Findings: Low socioeconomic status was consistently associated with markedly elevated alcohol-attributable harms, including after adjustment for weekly consumption, binge drinking, body mass index and smoking. There was evidence of effect modification: for example, relative to light drinkers living in advantaged areas, the hazard ratio for excessive drinkers was 6.75 (95% CI 5.09-8.93) in advantaged and 11.06 (95% CI 8.53-14.35) in deprived areas. We found little support for downward social selection. Interpretation: Disadvantaged social groups experience greater alcohol-attributable harms compared to the advantaged for given levels of alcohol consumption, even after accounting for different drinking patterns, obesity and smoking status at the individual level