Pharmacological and non-pharmacological treatments for major depressive disorder in adults: A systematic review and network meta-analysis

Abstract Depression has brought huge disease burden to the world. This systematic review aimed to compare the efficacy and safety of pharmacological and non-pharmacological treatments for major depressive disorder (MDD). We searched electronic databases with time range from 1990.1.1 to 2018.9.5. Randomized controlled trials (RCTs) including adult patients with MDD were eligible for inclusion. We conducted network meta-analyses using multivariate meta-analyses models under the frequency framework. Primary outcomes were efficacy (response rate) and safety (overall risk of adverse events). We estimated summary odds ratios (ORs) based on group-level data. 20,937 citations were identified, 91 trials comprising 10,991 participants were included in efficacy study, and 32 trials comprising 5,245 participants were included in safety study. In terms of efficacy, all treatments studied (acupuncture, mirtazapine, herbal medicine, venlafaxine, physical exercise, cognitive-behavioral therapy (CBT), bupropion, fluoxetine, and vortioxetine) except for probiotics were significantly more effective than placebo. In terms of safety, bupropion, fluoxetine, venlafaxine, and vortioxetine were significantly less safe than placebo. Herbal medicine and mirtazapine had no significant difference in overall risk of adverse events compared with placebo. Acupuncture, CBT, physical exercise and probiotics were lack of eligible safety data.