Genetic Correlates of Treatment Utilization for Alcohol Use Disorder: A Family‐Based and Polygenic Investigation

ABSTRACT Background The widespread prevalence of alcohol use disorder (AUD) and availability of treatments contrasts with limited treatment utilization. Here, in the Collaborative Studies on the Genetics of Alcoholism (COGA), we examined the extent to which genetic factors (i.e., family‐based heritability ( h 2 GCTA ), polygenic scores [PGS]) are associated with individual differences in AUD treatment utilization that could be leveraged to target intervention efforts. Methods Non‐proband COGA participants with a lifetime Diagnostic and Statistical Manual of Mental Disorders, 5th Edition (DSM‐5) AUD diagnosis, whose genomic ancestry resembled that of European (EUR‐like: N = 3148; 370 nuclear families; 222 extended family pedigrees) and African (AFR‐like: N = 962; 92 nuclear families; 79 extended family pedigrees) reference panels were selected for analyses. AUD treatment h 2 GCTA and associations with PGS (i.e., alcohol involvement, cigarette use, psychopathology, externalizing behavior, and cognition‐related) were estimated, along with between‐ and within‐family effects of the PGSs. Results AUD treatment was heritable ( h 2 GCTA = 0.36; SE = 0.06) and positively associated with PGS for problematic alcohol use, drinks/week, externalizing behavior, depression, anxiety and schizophrenia/bipolar (ORs > 1.12) while being negatively associated with education attainment (OR = 0.90) in the EUR‐like sample after accounting for AUD severity ( p FDR ≤ 0.04). When all significant PGSs were simultaneously included, only alcohol PGSs remained significant. There were no significant PGS associations in within‐family analyses. Despite a nominally significant positive association with cigarettes per day, no PGSs were associated with AUD treatment following FDR adjustment in the AFR‐like sample. Conclusions Genetic factors underlying alcohol use influence AUD treatment utilization, but this phenotype is largely not significantly associated with current estimates of polygenic scores for other complex traits; this suggests a potential unique genetic etiology requiring further investigation with a genome‐wide association study. The lack of significant within‐family PGS associations may have likely arisen from power reductions due to restricting the sample to full siblings. Together, these findings highlight potential facets influencing the underutilization of AUD treatment.