Meta-analyses of the association between alcohol dependence diagnosis and ALDH2 polymorphism suggest a rather large genetic effect (Luczak, et al., 2006; Zintzaras, et al., 2006), although these meta-analyses largely comprise case-control studies in fully adult samples, unlike our own. In contrast, for the alcohol use outcomes we examined, the magnitude of the protection associated with possession of the ALDH2*2 allele ranged from small to moderate (table 3). There are multiple possible reasons why possession of the ALDH2*2 allele might afford only a small-to-moderate level of protection in our sample. First, as described above, the estimated effect of possession of the ALDH2*2 allele in this sample of adopted participants should be free from inflation due to confounding introduced by the environmental influence of a biological parent or parents who also possesses the allele. Further, the maximum ALDH2*2 effect that we observed for each alcohol use phenotype occurred at the later ages included in the sample, so it is possible that the effect may continue to increase with age, beyond the ages at which our participants were assessed. Alternatively, though, rather than
