A Bayesian meta-analysis of 10 clinical trials involving 515 adults with diagnosed depression found that psilocybin-assisted therapy produced a pooled mean reduction of 10.08 points on the 17-Item Hamilton Depression Rating Scale. The psychological protocols used alongside psilocybin varied: manualized directive psychotherapy, manualized nondirective psychological support, non-manualized nondirective psychological support, and non-manualized supportive psychotherapy. Compared with manualized nondirective psychological support, the other three approaches did not differ significantly in their effect on depressive symptoms. The improvement in depressive symptoms was not associated with the type of psychological protocol employed.
The antidepressant effects of psychedelics may be overestimated in trials using pre-post single-arm, non-active-drug-as-placebo, and waitlist-control designs. A systematic review of 19 trials found that psilocybin and MDMA showed large to medium effect sizes in non-active-placebo designs (psilocybin: Hedges' g = 0.87; MDMA: g = 0.65), but effects were not statistically significant in active-placebo designs. Psilocybin effect sizes were very large in pre-post (g = 2.51) and waitlist-control (g = 2.88) designs. Ayahuasca also showed large effects in pre-post (g = 1.88) and non-active-placebo (g = 1.60) designs. LSD was significant only in non-active-placebo design (g = 1.49). Limited sample sizes, difficulty maintaining participant blinding, and high expectancy likely inflate apparent efficacy.