Review articleA systematic review of yoga for major depressive disorder

- Seven RCTs have investigated the effects of yoga in major depressive disorder.
- There is some evidence of positive effects beyond placebo.
- Yoga's efficacy seems to be comparable to that of antidepressant medication.
- Larger and adequately powered RCTs using non-inferiority designs are needed.

BackgroundThe purpose of this review was to investigate the efficacy and safety of yoga interventions in treating patients with major depressive disorder.MethodsMEDLINE, Scopus, and the Cochrane Library were screened through December 2016. Randomized controlled trials (RCTs) comparing yoga to inactive or active comparators in patients with major depressive disorder were eligible. Primary outcomes included remission rates and severity of depression. Anxiety and adverse events were secondary outcomes. Risk of bias was assessed using the Cochrane tool.ResultsSeven RCTs with 240 participants were included. Risk of bias was unclear for most RCTs. Compared to aerobic exercise, no short- or medium-term group differences in depression severity was found. Higher short-term depression severity was found for yoga compared to electro-convulsive therapy; remission rates did not differ between groups. No short-term group differences occurred when yoga was compared to antidepressant medication. Conflicting evidence was found when yoga was compared to attention-control interventions, or when yoga as an add-on to antidepressant medication was compared to medication alone. Only two RCTs assessed adverse events and reported that no treatment-related adverse events were reported.LimitationsFew RCTs with low sample size.ConclusionsThis review found some evidence for positive effects beyond placebo and comparable effects compared to evidence-based interventions. However, methodological problems and the unclear risk-benefit ratio preclude definitive recommendations for or against yoga as an adjunct treatment for major depressive disorder. Larger and adequately powered RCTs using non-inferiority designs are needed.