Psilocybin for Depression: The First Meta-Analysis That Asks How, Not Just If
- Meta-analysis of 7 RCTs (522 participants) found psilocybin-assisted therapy (PAT) has a large and significant antidepressant effect — confirming the clinical signal across pooled trials
- Larger effects associated with: body-weight-adjusted dosing, longer preparation sessions, longer dosing sessions, longer integration sessions, and non-manualized psychotherapy
- No significant between-subgroup heterogeneity — meaning these methodological moderators show trends, not definitive differences, and need confirmatory research
- First systematic examination of how PAT methodology influences antidepressant outcomes — moving the field from "does it work?" to "how should we do it?"
The psilocybin-for-depression evidence has accumulated to the point where "does it work?" is no longer the interesting question. Seven RCTs, 522 participants, a large pooled effect. The Toronto group behind this meta-analysis asks the question that matters next: does it matter how you deliver it?
The methodology matters
The subgroup analyses reveal a pattern. Larger effects are associated with:
Body-weight-adjusted dosing — rather than fixed doses. This makes pharmacological sense: a 60 kg patient and a 100 kg patient receiving the same 25 mg dose are not having the same pharmacological experience.
Longer preparation sessions — more time preparing the patient for the psychedelic experience. The therapeutic frame before the drug matters.
Longer dosing sessions — more time during the active experience. Rushed sessions may cut short therapeutically productive states.
Longer integration sessions — more time afterward to process what happened. The psychedelic experience is not the therapy; the integration is.
Non-manualized psychotherapy — flexible, responsive therapeutic support rather than rigid protocol adherence. This is counterintuitive for evidence-based medicine, which generally favors manuals. But psilocybin experiences are highly individual, and a therapist locked into a script may miss the patient's most therapeutically productive material.
The cautionary note
None of these subgroup differences reached statistical significance for between-group heterogeneity. The trends are clinically plausible but not confirmed. This is a hypothesis-generating meta-analysis, not a prescriptive one. The field should not yet build protocols around these findings — but it should test them in head-to-head protocol comparison trials.
For your practice
For clinicians tracking psilocybin therapy developments: this meta-analysis is your current evidence summary. The pooled effect is large and robust. If you are involved in clinical trial design or protocol development: the methodological moderators identified here — particularly the integration session length and non-manualized flexibility — deserve prospective testing. For patients asking about psilocybin: the evidence base now includes 7 RCTs. Legal access currently exists in Oregon, Colorado, and Australia (authorized prescriber pathway). Czechia joined in January 2026.
Seven RCTs confirm psilocybin works for depression. The next question is not whether to do it, but how to do it well — and the answer may be: take more time.
Only 7 RCTs available — still a small literature. Subgroup analyses are exploratory and underpowered. Heterogeneous depression definitions (MDD, TRD). Blinding is imperfect in psychedelic trials. Comparison to active psychotherapy controls is lacking.