PREDICTORS OF TREATMENT EFFICACY IN A CLINICAL-TRIAL OF 3 PSYCHOSOCIAL TREATMENTS FOR ADOLESCENT DEPRESSION

Objective: To assess the predictors of treatment outcome across treatm ents, as well as those associated with differential treatment response . Method: One hundred seven adolescent outpatients, aged 13 to 18 year s, with DSIM-III-R major depression were randomly assigned to one of t hree manual-based, brief (12 to 16 sessions) psychosocial treatments: cognitive-behavioral therapy (CBT), systemic-behavioral family therapy or nondirective supportive therapy. Those with good and poor outcomes were compared. Results: Continued depression was predicted by clinica l referral (versus via advertisement) and was in part mediated by hope lessness. Other predictors of depression were comorbid anxiety disorde r and higher levels of cognitive distortion and hopelessness at intake . Achievement of clinical remission was predicted by a higher level of self-reported depression. Poorer functional status was predicted by a higher level of initial interviewer-rated depression. Comorbid anxiet y and maternal depressive symptoms predicted differential treatment ef ficacy. CBT's performance continued to be robust with respect to nondi rective supportive therapy, even in the presence of the above-noted ad verse predictors. Conclusion: Predictors of poor outcome may give clue s as to how to boost treatment response. Subjects who come to treatmen t for clinical trials via advertisement (versus clinical referral) may show more favorable treatment responses. CBT is likely to be a robust intervention even in more complex and difficult-to-treat patients.