While clinical science has yielded a number of psychotherapeutic interventions that benefit individuals with a broad range of psychopathology, not everyone benefits from these treatments and it is still unclear how these treatments work. Consequently, there is a significant need for research into the predictors and mediators of treatment response, particularly beyond traditional randomized clinical trial settings. We have examined factors involved in treatment response for a complex psychiatric sample presenting for treatment in a partial hospital setting, finding support for the relevance of therapeutic skill change, rumination, and treatment expectancy in prediction of therapeutic response.