Given the high rate of psychiatric comorbidity with depression in adolescents, treatment interventions that have been shown to be effective with """"""""pure"""""""" samples of depressed adolescents need to be evaluated with youth who have comorbid conditions. The goal of this proposal is to evaluate the efficacy of a cognitive-behavioral group treatment intervention for adolescents whose depression is comorbid with conduct disorder. Adolescents, ages 13-17, who have been charged with at least two criminal violations but are not in juvenile detention in the Lane Country Department of Youth Services (DSY) will be referred to participation in the study. Adolescents meeting initial inclusion criteria will be recruited to participate in a diagnostic interview (K-SADS for DSM-IV) assessing all major psychiatric disorders, including depression (major depression and dysthymia) and conduct disorder. Over a four-year period, 200 adolescents with comorbid depression/conduct disorder will be randomly assigned to one of two interventions conditions: (1) the Adolescent Coping With Depression (CWD-A) course, or (2) academic tutoring. Adolescents will be assessed pre- and post-treatment (estimated n=150) and at 6 and 12 months post-treatment (estimated n=120). In addition, academic and criminal arrest records will be monitored for 12 months pre- and post-treatment (a total of 24 months). Analyses will focus on 3 primary issues: (1) Group differences in treatment outcome (depression and conduct disorder). The main hypotheses are that the CWD-A intervention will be superior to academic tutoring in reducing depression and conduct disorder, both of which will be evaluated as numerical and categorical (diagnosis vs. no diagnosis) measures. The intervention is predicted to impact depression more significantly than conduct disorder. (2) Group differences in additional outcome measures (e.g., future suicidal behavior, criminal recidivism, and academic performance). (3) Predictors of participation vs. attrition, improvement vs. nonrecovery, and maintenance of gains vs. relapse. Predictors will include pre-treatment patient variables, in-session variables, and interventionist variables.
