Associations between interrelated dimensions of socio-economic status, higher risk drinking and mental health in South East London: A cross-sectional study

Abstract

Aim: To examine patterns of hazardous, harmful and dependent drinking across different socio-economic groups, and how this relationship may be explained by common mental disorder.

Methods and findings: Between 2011-2013, 1,052 participants (age range 17-91, 53% female) were interviewed for Phase 2 of the South East London Community Health study. Latent class analysis was used to define six groups based on multiple indicators of socio-economic status in three domains. Alcohol use (low risk, hazardous, harmful/dependent) was measured using the Alcohol Use Disorders Identification Test and the presence of common mental disorder was measured using the revised Clinical Interview Schedule. Multinomial regression was used to explore associations with hazardous, harmful and dependent alcohol use, including after adjustment for common mental disorder. Harmful and dependent drinking was more common among people in Class 2 'economically inactive renters' (relative risk ratio (RRR) 3.05, 95% confidence interval (CI) 1.07-8.71), Class 3 'economically inactive homeowners' (RRR 4.11, 95% CI 1.19-14.20) and Class 6 'professional renters' (RRR 3.51, 95% CI 1.14-10.78) than in Class 1 'professional homeowners'. Prevalent common mental disorder explained some of the increased risk of harmful or dependent drinking in Class 2, but not Class 3 or 6.

Conclusions: Across distinct socio-economic groups in a large inner-city sample, we found important differences in harmful and dependent drinking, only some of which were explained by common mental disorder. The increased risk of harmful or dependent drinking across classes which are very distinct from each other suggests differing underlying drivers of drinking across these groups. A nuanced understanding of alcohol use and problems is necessary to understand the inequalities in alcohol harms.

Fig 1. Probability of hazardous and harmful drinking by class, predicted from multinomial regression analysis including age, sex, ethnicity, marital status and number of children, with 95% CIs.