Background: When treatment-refractory schizophrenia shows an insufficient response to a trial of clozapine, clinicians commonly add a second antipsychotic, despite the lack of robust evidence to justify this practice. Objectives: The main objectives of the study were to establish the clinical effectiveness and cost-effectiveness of augmentation of clozapine medication with a second antipsychotic, amisulpride, for the management of treatment-resistant schizophrenia. Design: The study was a multicentre, double-blind, individually randomised, placebo-controlled trial with follow-up at 12 weeks. Settings: The study was set in NHS multidisciplinary teams in adult psychiatry. Participants: Eligible participants were people aged 18-65 years with treatment-resistant schizophrenia unresponsive, at a criterion level of persistent symptom severity and impaired social function, to an adequate trial of clozapine monotherapy. Interventions: Interventions comprised clozapine augmentation over 12 weeks with amisulpride or placebo. Participants received 400 mg of amisulpride or two matching placebo capsules for the first 4 weeks, after which there was a clinical option to titrate the dosage of amisulpride up to 800 mg or four matching placebo capsules for the remaining 8 weeks. Main outcome measures: The primary outcome measure was the proportion of 'responders', using a criterion response threshold of a 20% reduction in total score on the Positive and Negative Syndrome Scale. Results: A total of 68 participants were randomised. Compared with the participants assigned to placebo, those receiving amisulpride had a greater chance of being a responder by the 12-week follow-up (odds ratio 1.17, 95% confidence interval 0.40 to 3.42) and a greater improvement in negative symptoms, although neither finding had been present at 6-week follow-up and neither was statistically significant. Amisulpride was associated with a greater side effect burden, including cardiac side effects. Economic analyses indicated that amisulpride augmentation has the potential to be cost-effective in the short term [net saving of between 329 and 2011; no difference in quality-adjusted life-years (QALYs)] and possibly in the longer term. Limitations: The trial under-recruited and, therefore, the power of statistical analysis to detect significant differences between the active and placebo groups was limited. The economic analyses indicated high uncertainty because of the short duration and relatively small number of participants. Conclusions: The risk-benefit of amisulpride augmentation of clozapine for schizophrenia that has shown an insufficient response to a trial of clozapine monotherapy is worthy of further investigation in larger studies. The size and extent of the side effect burden identified for the amisulpride-clozapine combination may partly reflect the comprehensive assessment of side effects in this study. The design of future trials of such a treatment strategy should take into account that a clinical response may be not be evident within the 4-to 6-week follow-up period usually considered adequate in studies of antipsychotic treatment of acute psychotic episodes. Economic evaluation indicated the need for larger, longer-term studies to address uncertainty about the extent of savings because of amisulpride and impact on QALYs. The extent and nature of the side effect burden identified for the amisulpride-clozapine combination has implications for the nature and frequency of safety and tolerability monitoring of clozapine augmentation with a second antipsychotic in both clinical and research settings.