Effect of psilocybin-assisted psychotherapy on anxiety symptoms: A systematic review and meta-analysis

A systematic review of 25 studies found that psilocybin-assisted psychotherapy (PAP) produces a large reduction in anxiety symptoms within groups (Hedge's g = 0.96) and a small reduction compared to control groups (Hedge's g = 0.48). However, the studies varied widely in psychotherapy format, dosing, session structure, and outcome timing, and high heterogeneity persisted even after accounting for these differences. The authors conclude that PAP shows promise for treating anxiety across different diagnoses but caution that variability in study quality, design, sample representativeness, and high heterogeneity limit confidence in the findings. More rigorous trials with diverse populations are needed.

Study at a glance

Abstract

Abstract Background and Aims Psilocybin-assisted psychotherapy (PAP) is a novel, transdiagnostic treatment in which the 5-HT2A receptor agonist psilocybin is combined with psychotherapy. Studies to date have evaluated PAP's effects on depression, substance use, and end-of-life adjustment. Relatively less attention has been given to its effects on anxiety symptoms, which are highly comorbid with other psychiatric conditions and are a leading cause of global disability. This review systematically evaluated evidence for PAP's effects on anxiety symptoms across diagnoses, with attention to variations in interventional components across studies. Methods A systematic review was conducted following PRISMA guidelines. Searches were completed in six databases and independent reviewers screened records. Study quality was assessed and data extracted on participant demographics and intervention features. Random-effects models estimated within- and between-group effects from baseline to primary endpoint. Results Twenty-five studies were determined eligible for inclusion. Considerable heterogeneity was observed in psychotherapy format, dosing, session structure, and outcome timing. Pooled results showed a large within-groups effect on anxiety after controlling for measurement artifacts (Hedge's g = 0.96) and a small between-groups effect (Hedge's g = 0.48). High heterogeneity persisted even after controlling for the influence of different anxiety measures and moderators related to intervention formulation and delivery. Conclusions PAP shows promise for reducing anxiety across primary diagnoses. However, variability in study quality, interventional design, sample representativeness, and high heterogeneity warrant caution in interpretation. More rigorous, high-quality trials with diverse populations are needed. Implications and directions for future research are summarized.