Yoga as a modality in occupational therapy practice for adults experiencing mood disorders: A systematic review and meta-analysis.

The review pooled 34 different outcomes from 22 trials into one number. Those outcomes measured different things, including mood symptoms, memory and a blood marker of oxidative stress, so the pooled figure does not describe any single thing a clinician could act on. The authors report considerable disagreement between the trials and admit caution is needed when generalising; on their own scoring, 10 of 22 studies had an unclear risk of bias and 1 a high risk. Common problems across the included studies were a lack of blinded raters and inconsistent reporting of the yoga itself; details such as instructor qualifications and how the yoga was delivered were often missing, making it impossible to say which kind of yoga, at what dose, might work. Several yoga programmes were intensive, up to 120 minutes five times a week for 12 weeks, which many patients would be unable to attend. Only English-language, peer-reviewed studies with complete standard deviation data could be used; 8 studies were dropped from the analysis for incomplete data, and the language restriction may have left out relevant work. No included study measured occupational participation or daily functioning, which the authors themselves describe as a gap in the literature.

The authors declared no competing interests and no financial support for the research, authorship or publication of the article.

Do not read the large pooled result as proof that yoga works better than other options for mood disorders. The review merged 34 different outcomes from very different trials into one number, the trials disagreed with each other enormously (I-squared 92.9%), the publication-bias check suggested 14 missing negative studies, and three of the studies that measured symptoms found yoga did no better than what it was compared with, which included support groups, mindfulness and usual care.