The efficacy of acceptance and commitment therapy (ACT) in psychosis has been reported but not for medication-resistant psychosis.To test the efficacy of ACT in a sample of community-residing patients with persisting psychotic symptoms. (Australian New Zealand Clinical Trials Registry: ACTRN12608000210370.) METHOD: The primary outcome was overall mental state at post-therapy (Positive and Negative Syndrome Scale - total); secondary outcomes were psychotic symptom dimensions and functioning. In total, 96 patients were randomised to ACT (n = 49) or befriending (n = 47). Symptom, functioning and process measures were administered at baseline, post-therapy and 6 months later.There was no group difference on overall mental state. In secondary analyses the ACT group showed greater improvement in positive symptoms and hallucination distress at follow-up: Cohen's d = 0.52 (95% CI 0.07-0.98) and 0.65 (95% CI 0.24-1.06), respectively.Improvements reflected the treatment focus on positive symptoms; however, absence of process-measure changes suggests that the ACT intervention used did not manipulate targeted processes beyond befriending. Symptom-specific therapy refinements, improved investigation of process and attention to cognitive functioning and dose are warranted in future research.