Why this is important

There has been very little research regarding the clinical effectiveness of structured individual and group psychological interventions for children and young people with bipolar disorder. Research on unipolar depression in children and young people supports the effectiveness of cognitive behavioural therapy (CBT), interpersonal therapy (IPT) and short-term family therapy. However, there have been no published trials investigating clinical and functional outcomes for young people with bipolar depression. Given the increasing emphasis on early interventions in related conditions such as psychosis and unipolar depression, it is important to know the form of psychological therapy that can benefit young people with bipolar disorder.

A high-quality, non-inferiority randomised controlled trial (RCT) should recruit young people during an acute episode of bipolar depression who are treated with structured psychological interventions (CBT compared with IPT). Interventions should be offered over 6–9 months, with a 9-month follow-up period. Key outcomes should include clinical recovery, symptom change, personal recovery or functional outcomes at the end of treatment and at 9-month follow-up, and cost effectiveness.

Why this is important

Lithium and antipsychotic medication are known to reduce the risk of relapse when used long term in people with bipolar disorder. Relapses do still occur and the response is usually to add another mood-stabilising drug. However, lithium and antipsychotics are associated with a number of side effects, some of which can adversely affect physical health. The relative effects of lithium, an antipsychotic or a combination of these drugs, regarding efficacy, tolerability, cost effectiveness and quality of life are unknown. Such information is important to people with bipolar disorder to help them make an informed choice about the treatment options available to them, and to the NHS to inform the best use of resources.

The suggested programme of research should involve a pragmatic -arm RCT comparing lithium monotherapy with antipsychotic monotherapy (haloperidol, olanzapine, quetiapine or risperidone) and a combination of lithium and an antipsychotic. The study should last at least 1 year with the primary outcome being quality of life. Symptom control, relapse, function and economic outcomes should also be measured.

Why this is important

Bipolar depression occurs three times more frequently than mania and is associated with suicide and impaired function and quality of life. The Guideline Development Group found that the combination of fluoxetine and olanzapine was the most clinically- and cost-effective treatment for bipolar depression. Antidepressants (imipramine, paroxetine and moclobemide) alone were ineffective compared with placebo. Olanzapine alone was an effective treatment for bipolar depression, but not as effective as olanzapine and fluoxetine in combination. However, for many people some antidepressants are ineffective or cannot be tolerated. For these people the NICE clinical guideline on depression in adults³ recommends changing to another antidepressant from the same or a different class.

A two-arm non-inferiority RCT of the combination of fluoxetine and olanzapine within BNF therapeutic levels compared with an alternative SSRI and olanzapine for moderate or severe bipolar depression with a 12-week follow-up period should be carried out. The primary clinical outcome should be depression response. Secondary outcomes should be depression remission, function, anxiety symptoms, emergent mania or hypomania symptoms, other adverse outcomes, quality of life and cost effectiveness.

Why this is important

There is moderate-quality evidence of the effectiveness of a specialised collaborative care service compared with usual treatment in reducing hospitalisation in two studies (from Denmark and the US). There is no overall evidence of an effect on relapse or other outcomes. An economic analysis from one study showed that better clinical outcomes were achieved at two-thirds of the overall cost of usual treatment. If similar results were obtained in England, then better care for a substantially reduced cost might be achieved.

A two-arm multicentre RCT of a specialist collaborative care service for people admitted to hospital with bipolar disorder compared with usual treatment, with follow-up of at least 2 years, is needed. Community alternatives to hospitalisation should be included. The specialist intervention should be based on collaborative care principles, including group or other psychoeducation to promote self-management, care coordination and algorithm-derived psychotropic medication (in line with this guideline) under the direction of a psychiatrist. Feasibility and acceptability development work should involve collaborating with people with bipolar disorder and professionals to demonstrate sustainable service delivery and recruitment, and identify and address possible barriers to such a service. Clinical outcomes should include time to next bipolar episode, mania and depression symptoms, function, recovery and quality of life. Economic outcomes should consider health, social care and personal costs.

Why this is important

The Guideline Development Group found that individual structured psychological interventions are clinically effective. Studies support the efficacy of individual CBT, but evidence for long-term benefits of internet-based interventions is less conclusive. Internet-facilitated CBT has the potential to deliver the key components of face-to-face CBT in a more cost-effective and accessible format. If this proves to be the case then increased access to cost-effective psychological care could be rapidly achieved. The proposed research programme would have 2 phases: (1) software development of internet-facilitated CBT including alpha, beta and feasibility and acceptability testing to confirm that the intervention is safe, acceptable and used by potential patients; (2) a two-arm non-inferiority RCT comparing internet-facilitated CBT with individual face-to-face CBT designed for bipolar disorder. Participants should be aged 16 years and over and be in a state of euthymia when recruited. The primary outcomes should be personal recovery and quality of life at 12-month follow-up. Secondary outcomes should be time to relapse, social and occupational functioning, and cost effectiveness.