SummaryBackground: Group psychoeducation (PEd) is a cheap NICE recommended treatment for bipolar disorder. Its clinical effectiveness and acceptability are unclear compared to unstructured group support (PS) matched for delivery and aim of treatment, and previous bipolar history. Methods: Randomised, parallel group, observer blinded superiority controlled trial recruiting at eight sites from secondary care or self-referral with two year follow up. Participants aged 18 years or over with bipolar disorder, no episode in the preceding four weeks. Randomisation using minimisation for number of previous episodes (1-7, 8-19, 20+ bands) and site by clinical trials unit to either 21 two-hour weekly sessions of PEd or PS, allocation conveyed to trial co-ordinator and masking of assessors to outcome. Intention to treat primary outcome was time to the next bipolar episode with planned moderator analysis of number of previous bipolar episodes and qualitative interview of participant experience. Trial registration number ISRCTN62761948, ended. (IQR) 14 (15.5) versus 9 (15) sessions, p=0.026). Over 96 weeks, time to first bipolar episode did not differ between PEd and PS (number of bipolar episodes 89 (58%) versus 95 (65%), hazard ratio (HR) = 0.83 (95% CI 0.62-1.11) p=0.217. The treatment was moderated in favour of PEd for participants with fewer previous bipolar episodes (X 2 = 6.80, p = 0.034, HR 0.28 (95% CI 0.12-0.68) for 1-7 previous episodes). Four participants (1 PEd, 3 PS) died in the follow-up period of the study. Interpretation: Structured group psychoeducation was no more clinically effective than similarly intensive unstructured group support but was more acceptable and improved outcome in participants with few previous episodes.