Abstract

Early alcohol use initiation predicts onset of alcohol use disorders (AUD) in adulthood. However, little is known about developmental trajectories of alcohol use initiation or their putative biological and environmental correlates. Here we report the results of latent class analyses identifying two trajectories of alcohol use initiation in a prospective study of adolescents selected for the presence or absence of familial risk for depression at baseline. The latent class analyses identified two distinct patterns of initiation: early initiators (EI; n=32) who reported greater baseline alcohol use (M = 1.12, SE = .35) and exhibited a more rapid rate of change in use between baseline and each follow up wave (M = 4.43, SE = .94); and in contrast, late initiators (LI; n=298) who reported lower baseline use (M = .23, SE = .03) and exhibited a slower rate of change between baseline and each of the subsequent follow up waves (M = .12, SE = .03). Early initiators had more positive expectancies regarding alcohol (p = .002 - p = .005) and reported higher levels of stressful life events during the year prior to baseline assessment (p = .001). Additionally, fMRI analyses revealed that EI displayed heightened threat-related amygdala activity at baseline compared to LI (p = .001), but no differences in reward-related ventral striatum activity. Lastly survival analyses revealed that EI initiators were 6.7 times more likely to develop an AUD by age 19 when compared to the LI (p = .005). These patterns, which were independent of broad familial risk for depression, suggest that early initiation of alcohol use during adolescence associated with later risk for AUD is reflected in both higher levels of stressful life events and higher neural reactivity to threat, the combination of which may inform ongoing efforts to prevent persistent dysfunction.