A re-analysis of Chiang et al. (2017): Cognitive-behaviour therapy does not have a statistically significant effect on levels of depression or mania in bipolar disorder

There is an ongoing debate concerning the efficacy of cognitive behaviour therapy (CBT) for bipolar disorder. Chiang et al.’s (2017) meta-analysis concluded that CBT was effective in reducing symptoms of depression and mania severity in patients with bipolar disorder, and that article has been cited more than 200 times. However, Chiang et al.’s conclusions are based, in part, on a pair of implausibly large effect sizes that were extracted from a randomized controlled trial published in 2010. In this paper, we repeat Chiang et al.’s analyses, but replace the pair of implausibly large effect sizes with much more plausible effect sizes taken from a second journal article that reported results of that 2010 randomized controlled trial. Our findings suggest that Chiang et al. should have reported that CBT does not have a statistically significant effect on symptoms of depression and mania severity in patients with bipolar disorder. Data supporting our findings are available at this – doi.org/10.17605/osf.io/x3qwp – page on the Open Science Framework.