The TRF externalizing results were similar to the CBCL externalizing results. In terms of main effects, adolescents who were male, who were missing a parental alcohol problem report, and whose parents had higher antisocial behavior and alcohol problems had higher TRF externalizing (see Model 1 in Tables 5 and 6). Missing a parental antisocial behavior report, life events, and GABRA2 genotype were not significantly associated with TRF externalizing. As shown in Model 2 of Tables 5 and 6, the interaction between cross-lagged life events and GABRA2 was significant in the parental antisocial behavior and alcohol problems models. The interaction effects accounted for 1.54% and 1.48% of the variance in the respective models. Figure 3 depicts the mean levels of TRF externalizing as a function of life events and GABRA2 genotype for the model that included parental antisocial behavior. A consistent pattern of effects was found for the model that included parental alcohol problems (available upon request from the first author).
