Background Efficacy trials suggest that structured psychological therapies may significantly reduce recurrence rates of major mood episodes in individuals with bipolar disorders. Aims To compare the effectiveness of treatment as usual with an additional 22 sessions of cognitive–behavioural therapy (CBT). Method We undertook a multicentre, pragmatic, randomised controlled treatment trial (n=253). Patients were assessed every 8 weeks for18 months. Results More than half of the patients had a recurrence by 18 months, with no significant differences between groups (hazard ratio=1.05; 95% CI 0.74–1.50). Post hoc analysis demonstrated a significant interaction (P=0.04) such that adjunctive CBT was significantly more effective than treatment as usual in those with fewer than 12 previous episodes, but less effective in those with more episodes. Conclusions People with bipolar disorder and comparatively fewer previous mood episodes may benefit from CBT. However, such cases form the minority of those receiving mental healthcare.