Polygenic risk moderation of stressful life events in alcohol use disorder severity

Advances in gene–environment (GE) research underscore the need for novel statistical methods to clarify how social stress and genetic liability jointly influence Alcohol Use Disorder (AUD). This study examined the effect of Stressful Life Events (SLEs) on past-year DSM-5 AUD severity using propensity score matching and tested whether Polygenic Risk Scores (PRS) moderated the association between stress exposure and AUD severity. A secondary analysis was performed using the data from National Epidemiologic Survey on Alcohol and Related Conditions III (NESARC III; N = 36,309) and the newly available NESARC III genomic dataset (N = 22,848). Of 10,775 self-reported White subjects, individuals in the high-stress group (i.e., more than two stressors) (N = 4436) were matched to those in the low-stress group (N = 5048) using propensity score matching. Further, a moderation analysis was conducted to evaluate the interaction between PRS and SLE. Exposure to two or more SLEs in the past year significantly increased the odds of AUD severity after covariate adjustment (estimate: 0.37; 95%CI: 0.27–0.48; t-value: 6.96; p < 0.001). PRS was independently associated with higher AUD criteria counts after adjustment (estimate: 0.23; 95%CI: 0.07–0.39; t-value: 2.87; p = 0.01). PRS significantly moderated the relationship between the stress exposure and the AUD severity (interaction effect = 0.30; 95% CI: 0.06–0.54; t = 2.47; p = 0.01), indicating stronger stress-related effects among individuals with higher genetic risk. Additional sensitivity analyses suggested that PRS moderation by stress exposure is small and sensitive to model specification. This novel GE interaction model provides evidence that polygenic risk can potentiate the effects of high-stress exposure on the severity of AUD, thereby advancing post-GWAS modeling efforts and support precise identification of high-risk subpopulations for targeted prevention and intervention. Not applicable.