Efficacy and Safety of Psilocybin-Assisted Therapy for Depression: A Meta-Analysis of Randomised Controlled Trials
Siti Nashria Rusdhy, Andrian Fajar Kusumadewi, Carla Raymondalexas Marchira, Mustika Suci Mahardikaningrum, Teresa Lalita Wiryarini, Devira Ayu Wulandari
Open Access Indonesian Journal of Medical Reviews April 21, 2026 DOI: 10.37275/oaijmr.v6i2.883 via OpenAlex
Summary
AI-generated from the abstractPsilocybin-assisted therapy produces a large reduction in depressive symptoms, but the evidence is preliminary and limited by methodological problems. A meta-analysis of nine randomized controlled trials involving 514 participants found a large effect size (SMD = 1.270). However, the certainty of the evidence is rated low due to risk of bias, high heterogeneity, short-term follow-up, and publication bias. Effects were much larger when psilocybin was compared to a waitlist rather than an active placebo, and blinding is compromised by the drug's subjective effects. The authors conclude that robust Phase 3 trials are needed before routine clinical use.
Study at a glance
| Characteristics | Systematic review and meta-analysis Randomized Peer reviewed |
|---|---|
| Sample size | 514 |
| Population | Adults with major or treatment-resistant depression |
| Intervention | Psilocybin-assisted therapy |
| Key finding | Psilocybin-assisted therapy yields a large pooled effect on depressive symptom reduction, but the evidence is rated low certainty due to methodological limitations. |
Abstract
Psilocybin-assisted therapy shows promise for depression, though current evidence relies on Phase 2 trials with notable methodological limitations. We conducted a systematic review and meta-analysis of randomized controlled trials (RCTs) evaluating psilocybin-assisted therapy for major or treatment-resistant depression up to February 2024. We evaluated depressive symptom severity using random-effects meta-analysis, moderator analyses, Cochrane Risk of Bias 2, and GRADE methodology. Nine RCTs (N=514) were included. Psilocybin therapy demonstrated a large pooled effect size for symptom reduction (SMD = 1.270, 95% CI: 0.865–1.676, p<0.001). However, substantial heterogeneity was observed (I² = 79.1%). Comparator type significantly moderated outcomes, with waitlist controls showing substantially larger effects than active/placebo controls. Overall GRADE certainty of evidence was rated LOW due to risk of bias, heterogeneity, short-term outcomes, and publication bias concerns. In conclusion, while psilocybin-assisted therapy yields a large pooled effect estimate for depression, current findings are preliminary. Results are heavily qualified by methodological constraints, including waitlist-inflated efficacy, compromised blinding from subjective psychedelic effects, and the confounding influence of integrated psychological support. Confirmation through robust Phase 3 trials is required before supporting routine clinical implementation.