Reducing relapse and suicide in bipolar disorder: practical clinical approaches to identifying risk, reducing harm and engaging service users in planning and delivery of care – the PARADES (Psychoeducation, Anxiety, Relapse, Advance Directive Evaluation and Suicidality) programme

Bipolar disorder (BD) costs £5.2B annually, largely as a result of incomplete recovery after inadequate treatment.A programme of linked studies to reduce relapse and suicide in BD.There were five workstreams (WSs): a pragmatic randomised controlled trial (RCT) of group psychoeducation (PEd) versus group peer support (PS) in the maintenance of BD (WS1); development and feasibility RCTs of integrated psychological therapy for anxiety in bipolar disorder (AIBD) and integrated for problematic alcohol use in BD (WS2 and WS3); survey and qualitative investigations of suicide and self-harm in BD (WS4); and survey and qualitative investigation of service users’ (SUs) and psychiatrists’ experience of the Mental Capacity Act 2005 (MCA), with reference to advance planning (WS5).Participants were from England; recruitment into RCTs was limited to certain sites [East Midlands and North West (WS1); North West (WS2 and WS3)].Aged ≥ 18 years. In WS1–3, participants had their diagnosis of BD confirmed by the Structural Clinical Interview for theDiagnostic and Statistical Manual of Mental Disorders.In WS1, group PEd/PS; in WS3 and WS4, individual psychological therapy for comorbid anxiety and alcohol use, respectively.In WS1, time to relapse of bipolar episode; in WS2 and WS3, feasibility and acceptability of interventions; in WS4, prevalence and determinants of suicide and self-harm; and in WS5, professional training and support of advance planning in MCA, and SU awareness and implementation.Group PEd and PS could be routinely delivered in the NHS. The estimated median time to first bipolar relapse was 67.1 [95% confidence interval (CI) 37.3 to 90.9] weeks in PEd, compared with 48.0 (95% CI 30.6 to 65.9) weeks in PS. The adjusted hazard ratio was 0.83 (95% CI 0.62 to 1.11; likelihood ratio testp = 0.217). The interaction between the number of previous bipolar episodes (1–7 and 8–19, relative to 20+) and treatment arm was significant (χ2 = 6.80, degrees of freedom = 2;p = 0.034): PEd with one to seven episodes showed the greatest delay in time to episode. A primary economic analysis indicates that PEd is not cost-effective compared with PS. A sensitivity analysis suggests potential cost-effectiveness if decision-makers accept a cost of £37,500 per quality-adjusted life-year. AIBD and motivational interviewing (MI) cognitive–behavioural therapy (CBT) trials were feasible and acceptable in achieving recruitment and retention targets (AIBD:n = 72, 72% retention to follow-up; MI-CBT:n = 44, 75% retention) and in-depth qualitative interviews. There were no significant differences in clinical outcomes for either trial overall. The factors associated with risk of suicide and self-harm (longer duration of illness, large number of periods of inpatient care, and problems establishing diagnosis) could inform improved clinical care and specific interventions. Qualitative interviews suggested that suicide risk had been underestimated, that care needs to be more collaborative and that people need fast access to good-quality care. Despite SUs supporting advance planning and psychiatrists being trained in MCA, the use of MCA planning provisions was low, with confusion over informal and legally binding plans.Inferences for routine clinical practice from WS1 were limited by the absence of a ‘treatment as usual’ group.The programme has contributed significantly to understanding how to improve outcomes in BD. Group PEd is being implemented in the NHS influenced by SU support.Future work is needed to evaluate optimal approaches to psychological treatment of comorbidity in BD. In addition, work in improved risk detection in relation to suicide and self-harm in clinical services and improved training in MCA are indicated.Current Controlled Trials ISRCTN62761948, ISRCTN84288072 and ISRCTN14774583.This project was funded by the National Institute for Health Research (NIHR) Programme Grants for Applied Research programme and will be published in full inProgramme Grants for Applied Research; Vol. 6, No. 6. See the NIHR Journals Library website for further project information.