As the roster of efficacious preventive interventions for drug abuse risk grows, prevention researchers have become increasingly interested in refining and elaborating their understanding of intervention effects. In addition to documenting average outcome differences between intervention and control groups researchers are increasingly interested in better understanding how long term (distal) intervention effects are related to immediate (proximal) intervention effects. The present proposal is to apply this perspective to evaluating the long term impact (to age 16) of a two-stage, efficacious, preventive intervention (the SAFE Children program) offered to families residing in high risk (inner-city) communities, with a child entering elementary school. Our intent, in extending data collection and conducting these additional data analyses, is to model intervention effects as developmental alteration to help provide a more elaborate understanding of prevention impact, By building upon the original sampling and 9 previous points of data, we have four specific aims: (1) to estimate the overall long term effects (at age 16) of the SAFE Children I and II interventions for reducing the likelihood of substance use, the extent of substance use, and the age of first substance use; (2) to examine the relation between early intervention effects and long-term outcomes; (3) to test the relations of conditions and characteristics, pre-existing intervention, to initial (proximal) and long-term (distal) effects of the intervention, to test the relations of circumstances and events during the intervention to initial and long-term effects, and to test for the influence of post-intervention behavioral status and conditions and events on the relation between immediate and long-term effects; and (4) to document the developmental trajectories for substance use among this inner-city sample, including identification of trajectories related to antisocial/delinquent behavior and academic failure, and to then test for effects of the SAFE Children I and II interventions on distribution in membership across these trajectories.
