Consistent with many previous studies, we found that adolescent and young adult participants who carried the ALDH2*2 allele exhibited reduced alcohol use compared to ALDH2*1 homozygotic participants. For every phenotypic outcome (drinking index, alcohol dependence and abuse symptom count, and alcohol dependence or abuse diagnosis), the best-fitting model included terms for the effect of ALDH2 genotype (table 2). Furthermore, best-fitting models for each of these primary phenotypes included an age by ALDH2 genotype interaction, such that the protective effect of ALDH2*2 increased over the course of adolescence and young adulthood,. This pattern is in accordance with earlier research showing that ALDH2 genotype is not related to the initiation of alcohol use, or to ever having been intoxicated (Wall, et al., 2001), and that evidence for association between the ALDH2*2 allele and diminished alcohol use may not be apparent in young samples (Hendershot, et al., 2005) but may increase during young adulthood (Doran et al., 2007; Kim et al., 2010).
