note that the increase in drunkenness observed at the transition to adulthood in our sample is not sustained. There are no significant differences in rate of change in drunkenness after age 19, and mean levels of drunkenness do not differ significantly by genotype by the mid-20s, as illustrated by the overlapping error bars in the figures. Other analyses exploring the association between GABRA2 and a variety of impulsivity phenotypes in this sample show that this gene is associated with subclinical levels of externalizing behavior, as measured by the Achenbach Externalizing Scale, but not with clinical-level DSM symptom counts of alcohol dependence, other drug dependence, or antisocial behavior, as in the older adult COGA samples (Dick Aliev et al. 2013). These findings illustrate that genes can impact different outcomes across different developmental stages and underscore the importance of understanding the longitudinal pathways of risk associated with particular genes.
