Largest-ever review finds no evidence medicinal cannabis treats depression, anxiety, or PTSD

More than 700,000 Australians have used medicinal cannabis. Twenty-seven percent of adults aged 16 to 65 in the United States and Canada have tried it for medical purposes, with roughly half using it to manage mental health. Prescriptions have tripled in Australia over four years. The market is booming.

The evidence is not.

What 54 trials over 45 years actually show

A systematic review and meta-analysis published in The Lancet Psychiatry - the largest ever conducted on cannabinoids for mental health and substance use disorders - found no evidence that medicinal cannabis is effective for treating depression, anxiety, or post-traumatic stress disorder. The review, led by Dr. Jack Wilson at the University of Sydney's Matilda Centre, analyzed 54 randomized controlled trials published between 1980 and 2025.

The findings are stark. For the three conditions most commonly cited by patients seeking medicinal cannabis prescriptions, the pooled trial data show no benefit over placebo. This is not a matter of insufficient data. Forty-five years of trials, aggregated and analyzed, point in the same direction: these products do not work for these conditions.

"The routine use of medicinal cannabis could be doing more harm than good by worsening mental health outcomes, for example a greater risk of psychotic symptoms and developing cannabis use disorder, and delaying the use of more effective treatments," Wilson said.

Where cannabis might help - and where it backfires

The picture is not entirely negative. The review found evidence suggesting potential benefits for a handful of conditions: cannabis use disorder (dependence on cannabis itself), autism symptom reduction, insomnia, and tics or Tourette's syndrome. But the evidence quality for each of these was low.

The cannabis dependence finding carries an interesting parallel. Similar to how methadone treats opioid addiction, oral cannabis formulations - administered alongside psychological therapy - appeared to reduce cannabis smoking. The approach treats the addiction with a controlled version of the substance causing it.

But the same logic fails spectacularly elsewhere. When medicinal cannabis was tested in people with cocaine use disorder, it increased cocaine cravings. "It should not be considered for this purpose and may, in fact, worsen cocaine dependence," Wilson said.

The review also noted that medicinal cannabis has established evidence for certain non-psychiatric conditions - reducing seizures in some forms of epilepsy, managing spasticity in multiple sclerosis, and treating certain types of pain. The mental health evidence, though, falls short.

A prescription system running ahead of the science

The gap between prescribing practice and evidence has drawn attention from major medical bodies. In Australia, the Australian Medical Association and the Pharmacy Guild have raised concerns about largely unregulated growth in cannabis prescribing. The Therapeutic Goods Administration initiated a regulatory review, receiving more than 500 public responses published in February 2026.

In North America, the regulatory landscape is even more fragmented. State-level medical cannabis programs in the United States frequently list anxiety, PTSD, and depression among qualifying conditions - precisely the indications for which this review found no supporting evidence.

The study's authors frame their work as a tool for regulators and clinicians making evidence-based decisions. "Our study provides a comprehensive and independent assessment of the benefits and risks of cannabis medicines," Wilson said. Whether that assessment changes prescribing patterns remains to be seen. The commercial momentum behind medicinal cannabis is substantial, and patient demand - fueled in part by anecdotal reports and marketing - shows no sign of slowing.

What this review cannot answer

The review has inherent limitations. Randomized controlled trials of cannabis face significant methodological challenges: blinding is difficult because patients often recognize the psychoactive effects of THC, dosing varies widely across products and jurisdictions, and trial durations tend to be short relative to the chronic conditions being treated. The 54 included trials span different formulations (CBD, THC, and combinations), different dosing regimens, and different patient populations.

The review also cannot speak to why so many patients report subjective benefit from medicinal cannabis for anxiety and depression. Possible explanations include placebo response, the anxiolytic effects of THC intoxication being mistaken for therapeutic benefit, or self-selection bias - people who feel better continue using cannabis, while those who do not simply stop and are not captured in observational data.

What the review can say, with the weight of the largest evidence synthesis to date, is that controlled trials do not support the use of medicinal cannabis for depression, anxiety, or PTSD. The gap between what patients believe and what the data show is wide, and it is getting wider as prescriptions multiply.