We failed to find evidence of developmental specificity in terms of the timing of the first drink – that is, there were no apparent differences in the association between divorce/separation or any of the other predictors and initiation as a function of the age at first drink. The inconsistency with prior studies by Waldron et al. (2014a, 2014b) and Grant et al. (2015), in which risk conferred by parental separation/divorce was most pronounced in the pre- to early adolescent years, may in part be due to the younger age of the current sample, for whom onset of drinking occurred (by definition) at a young age. Subsequent studies either with older adolescent samples or with additional follow-up of this sample may reveal different relationships between divorce and alcohol use for earlier versus later ages of initiation.
