Do psychological interventions reduce symptoms of depression for patients with bipolar I or II disorder? A meta-analysis.

Background: Psychological therapies may play an important role in the treatment of bipolar disorders. Several meta-analyses that examine the effectiveness of psychotherapies for patients with bipolar disorder include conclusions about the impact upon bipolar depression. However, these tend not to consider differences in depression outcome depending upon whether the therapy primarily targets acute depression, nor severity of baseline depression. This may affect the conclusions drawn about the effectiveness of these therapies for acute bipolar depression treatment.

Objectives: This meta-analysis explored the effectiveness of psychological therapies in reducing bipolar depression, in particular examining whether: (1) the effect of therapy is greater when baseline depressive symptoms are more severe, and (2) the effect of therapy is greater when the primary focus of the therapy is the treatment of acute bipolar depression?

Data Sources: A systematic search was conducted using the following electronic databases; Cochrane Controlled Register of Trials (1996), MEDLINE (1966 onwards), EMBASE (1980 onwards), PsycINFO (1974 onwards), Scopus, Web of Science and Clinical Trials Registries (listed at:https://www.hhs.gov/ohrp/international/clinical-trial-registries/index.html).

Eligibility Criteria: Eligible studies were randomized controlled trials evaluating a psychological intervention for adults diagnosed with Bipolar I or II disorder. The comparators were usual care, wait-list, placebo, active treatment control. Post-treatment depression status was required to be measured continuously using a validated self- or observer- report measure, or categorically by a validated diagnostic instrument or clinical diagnosis by a suitably qualified person.

Data Extraction And Synthesis: Titles and abstracts were screened, followed by full texts. Two reviewers conducted each stage until agreement was reached, and both independently extracted study information. Means, standard deviations (SDs) and number of participants were retrieved from articles and used to perform a meta-analysis. The primary outcome was depressive symptom score.

Results: The database search identified 6388 studies. After removing the duplicates, 3298 studies remained, of which, 28 studies were included in the qualitative review and 22 in the meta-analysis. Effect sizes range from -1.99 [-2.50, -1.49] to 0.89 [-0.12, 1.90]. There was low quality evidence of a significant effect on symptoms of depression for cognitive behavioral therapy and dialectical behavior therapy. Trials of psychoeducation, mindfulness-based therapy, family therapy and interpersonal and social rhythm therapy showed no evidence of any effect on depression. We found no significant relationship between baseline depression score and depression outcome post-treatment when we controlled for therapy type and comparator. The result also showed that the effect sizes for studies targeting acute depression to be tightly clustered around a small overall effect size.

Conclusions: Some psychological therapies may reduce acute bipolar depression although this conclusion should be viewed with caution given the low quality of evidence. More research using similar therapy types and comparators is needed to better understand the relationship between depression status at baseline and outcome.