The majority of epidemiological studies have shown that the health effects of alcohol consumption are modified by people’s socio-economic status (SES). There are limited scientific data, however, on the specific mechanisms that cause the differences. Researchers used data on alcohol intake collected in the Scottish Health Surveys from 50,236 people to investigate whether the harmful effects of alcohol consumption differ by SES (factors include: education level, social class, household income, and area-based deprivation).
- Without considering alcohol intake, people with lower SES were overall at much greater risk for poor health outcomes than those in higher SES categories.
- Compared with people with “light” consumption (1–10 units/week for men, 1–7 units/week for women), the risk of alcohol-attributable hospital admission or death for people who reported “moderate” consumption (11–20 units/week for men; 8–13 units/week for women) was higher for all but more so among those with lower SES (hazard ratio, 1.35 in higher-SES groups and 2.95 in lower-SES groups).
- Compared with people with “light” consumption living in advantaged areas, the hazard ratios for alcohol-attributable hospital admission or death for people with “excessive” consumption (≥51 units/week for men, ≥36 units/week for women) were 6.12 in advantaged areas and 10.22 in deprived areas.
Comments: Unfortunately, the authors of this study had limited information on participants’ access to health care and patterns of alcohol consumption, and no data on the types of beverages. While the study showed marked differences in adverse health outcomes according to SES, the estimated amount of alcohol consumed and the factors considered inadequately explain the differences in health effects. The specific reasons why people with lower SES have worse health overall after alcohol consumption remain unclear.
R. Curtis Ellison, MD
Reference: Katikireddi SV, Whitley E, Lewsey J, et al. Socioeconomic status as an effect modifier of alcohol consumption and harm: analysis of linked cohort data. Lancet Public Health. 2017;2(6):e267-e276.