Evaluation of the risk relationship between average alcohol volume consumed and suicide: An analysis of mortality linked cohort data

ObjectiveTo evaluate the relationship between average alcohol volume consumed per day and suicide. MethodsData from the annual, cross-sectional US National Health Interview Survey, 1997-2018, was obtained, and linked to the 2019 National Death Index. The association between average alcohol volume consumed in grams per day (g/day) and suicide was quantified using Cox proportional hazards model (multiplicative) and Aalens additive hazard model. All analyses were stratified by sex, and adjusted for education, marital status, race/ethnicity, and survey year. ResultsOn the multiplicative scale, for males, former drinkers and those who consumed on average (40, 60] g/day had about 53% (HR=1.53, 95% CI: 1.10, 2.13) and 77% (HR=1.77, 95% CI: 1.17, 2.66) greater risk of dying by suicide, compared to lifetime abstainers, respectively. There was no significant association found for former or current drinkers among females, on the multiplicative scale. On the additive scale, for males and females, being a former drinker was associated with 11.4 (95% CI: 2.3, 20.4) and 5.6 (95% CI: 0.8, 10.4) additional deaths per 100,000 person years, compared to lifetime abstainers. For males only, drinking (40, 60] g/day on average was associated with 23.2 (95% CI: 6.7, 39.7) additional deaths per 100,000 person years. Level of education was not found to modify the focal relationship for males or females. ConclusionsThe findings suggest that the relationship between average alcohol volume consumed per day and suicide is nuanced. Additional research on the respective relationship is needed, including repeated measures of average alcohol consumption over time. What is already known on this topicThere is a dearth of studies on the sex-specific relationship between average alcohol volume consumed per day and suicide. The one existing study, from South Korea, found that for males as average alcohol volume consumed increased, the likelihood of death by suicide also increased. For females it was not possible to estimate the risk associated with the upper level of consumption due to a zero-cell count. What this study addsThis is the first systematic investigation of the sex-specific relationship between average alcohol volume consumed per day and death by suicide using a large linked dataset from the United States. It is also the first to evaluate the modifying effect of education, an important indicator of socioeconomic status, on the respective relationship. How this study might affect research, practice or policyThe findings were not in line with the sparse existing literature, indicating that this line of research is not yet resolved and more research is needed.