OBJECTIVE: A randomized placebo-controlled trial of treatment for common mental disorders in Goa, India found that psychological treatment (problem-solving) was no more effective than placebo. The study aimed to identify factors predicting outcome amongst participants receiving the psychological treatment, as measured by score on the Revised Clinical Interview Schedule (CISR). METHOD: The CISR scores at 2 months, 6 months and 12 months were modelled using multiple linear regression with a random effect for patients. Two types of variables were examined. Patient variables were age; sex; religion; education; marital status; severity of psychiatric morbidity at recruitment; and severity of social and life problems faced by the subject. Treatment variables were number of sessions attended and the hospital site. RESULTS: About half the participants had a depressive disorder; most of the remainder had mixed anxiety-depression. Overall adherence with the intervention was low; more than a third of patients attended only one session. Attending a greater number of sessions and facing a severe life problem were significantly associated with a worse outcome (p<0.005). CONCLUSION: The severe nature of life problems faced by some patients with common mental disorders may limit the efficacy of clinic-based psychological treatments in low income countries.