Predicting dropout in adolescents receiving therapy for depression.

OBJECTIVE: Therapy dropout is a common occurrence, especially in adolescence. This study investigated whether dropout could be predicted from a range of child, family, and treatment factors in a sample of adolescents receiving therapy for depression. METHOD: This study draws on data from 406 participants of the IMPACT study, a randomized controlled trial, investigating three types of therapy in the treatment of adolescent depression. Logistic regression was used to estimate the effects of predictors on the odds of dropout. RESULTS: Few pre-treatment predictors of dropout were found, with the only significant predictors being older age, antisocial behaviour, and lower scores of verbal intelligence. Missed sessions and poorer therapeutic alliance early in treatment also predicted dropout. Most child and family factors investigated were not significantly associated with dropout. CONCLUSIONS: There may be little about depressed adolescents' presentation prior to therapy starting that indicates their risk of dropout. However, within-treatment factors indicated that warning signs of dropout may be identifiable during the initial phase of therapy. Identifying and targeting early treatment indicators of dropout may provide possibilities for improving engagement. Clinical and methodological significance of this article: In the literature, a great deal of attention has been paid to child and family factors that predict therapy dropout, yet in this study, few pre-treatment characteristics were predictive of dropout. However, findings revealed possible warning signs of dropout in the early part of treatment, as poor therapeutic alliance and missed sessions were both found to be predictive of dropout. These findings call for therapists to be aware of such warning signs and clinical guidelines for managing cases at risk of dropout are warranted