Objective: The purpose of this study was to examine predictors of treatment response to cognitive-behavioral therapy (CBT) for depression in Parkinson's disease (PD). Method: The sample comprised 80 depressed ("DSM-IV" criteria) adults with PD (60% male) and their caregivers who participated in an National Institutes of Health-sponsored randomized-controlled trial of CBT vs. clinical monitoring from April 2007 until July 2010. Individually administered CBT was provided to people with PD for 10 weeks, modified to address the unique needs of the medical population, and supplemented with up to 4 separate caregiver educational sessions. Treatment response was defined a priori as a rating of depression much improved or very much improved on the Clinical Global Impression-Improvement Scale or [greater than or equal to] 50% reduction in the baseline Hamilton Depression Rating Scale score. It was hypothesized (a priori) that caregiver participation in treatment, motor disability, psychiatric comorbidity, and executive functioning would be significant predictors of response to CBT at end-of-treatment (Week 10) and short-term follow-up (Week 14). Results: At Week 10, caregiver participation was the only significant predictor of treatment response in the CBT group. At Week 14, both caregiver participation and executive functioning predicted response to CBT. Treatment group, baseline depression severity, executive functioning, motor disability, psychiatric comorbidity, marital status, and caregiver burden were also related to change in depression scores, for all participants, in secondary and exploratory models. Conclusions: Caregiver participation may enhance acute treatment response to psychosocial interventions for depression in PD. Further research is needed to extend and replicate these findings. (Contains 4 tables and 2 footnotes.)