Similar to the findings for EAs, social support from friends (but not family) was associated with lower levels of alcohol use among AAs, suggesting that friend support is particularly important in adulthood. However, there was no significant main effect of alc-PRS on alcohol use among AAs. There was also no significant interaction effect between alc-PRS and social support from friends or family in relation to alcohol use among the whole AA subsample, which could be, at least in part, due to the low predictive ability of alc-PRS among AAs. This finding of limited predictive utility of polygenic scores among AAs is likely due to the underrepresentation and small sample size of AAs in the GWAS studies that the alc-PRS was derived from (Martin et al., 2019; Su et al., 2018). These findings highlight the continued need for increased representation of AAs in genetic research to better characterize genetic risk among this population. Additionally, we note that alc-PRS for AAs were generated for the AUDIT-C while alc-PRS for EAS were generated for drinks per week, which matched more closely to the
