Largest Meta-Analysis Finds No Evidence Cannabis Treats Depression, Anxiety, or PTSD
- Systematic review of 5,774 studies, 54 RCTs included (2,477 participants) — the largest RCT-only meta-analysis of cannabinoids for mental health conditions
- No significant evidence that medicinal cannabis effectively treats anxiety disorders, PTSD, or depression — the three most common conditions for which it is prescribed
- Zero RCTs found assessing cannabinoids specifically for depression
- Limited evidence of benefit only for cannabis use disorder itself, insomnia, and Tourette's syndrome
Your patients are asking about cannabis for depression and anxiety. Some already use it. Many jurisdictions are prescribing it. This Lancet Psychiatry meta-analysis — the largest to date — delivers a clear verdict: there is no RCT evidence that cannabinoids treat the mental health conditions they are most commonly prescribed for.
The scale of the problem
The research team screened 5,774 studies and found only 54 RCTs meeting inclusion criteria — 2,477 participants total, spanning 1980 to 2025. For a substance prescribed to millions for mental health conditions, the evidence base is remarkably thin.
For anxiety disorders: no significant treatment effect. For PTSD: no significant treatment effect. For depression: not a single RCT exists. The conditions driving the majority of medical cannabis prescriptions have no randomised evidence supporting the practice.
Where limited evidence does exist — cannabis use disorder, insomnia, and Tourette's syndrome — the effect sizes are modest and the trial quality variable. This is not a blanket condemnation of cannabinoid research, but a precise assessment of what the current evidence does and does not support.
The clinical conversation
This matters because the gap between public perception and evidence is wider for cannabis than for almost any other psychoactive substance. Patients arrive with strong priors — anecdotal reports, media coverage, legal access interpreted as medical endorsement.
The practitioner's role is not to dismiss but to inform. When a patient with treatment-resistant depression asks about cannabis, the honest answer is: "There is no randomised trial evidence that it helps your condition. Zero trials for depression specifically. I cannot recommend it as treatment, but I can discuss what the research actually shows and what alternatives have evidence behind them."
This is not a prohibition conversation. It is an informed consent conversation.
The largest meta-analysis of cannabinoids for mental health found no RCT evidence for the three conditions they are most prescribed for — depression, anxiety, and PTSD.
Many included RCTs were small and short-term. Heterogeneity in cannabinoid formulations, doses, and delivery methods limits pooling. Absence of evidence is not evidence of absence — but it is evidence that prescribing has outpaced science.