A sweeping new study published in The Lancet Psychiatry has reignited a long-running debate about medical cannabis for mental health conditions. The findings make for uncomfortable reading for a fast-growing private prescribing sector that has largely outpaced the science behind it.
Researchers described this as the largest and most comprehensive review of its kind. They analysed 54 randomised controlled trials involving 2,477 participants across a range of mental health and substance use disorders. The central conclusion is stark: for the vast majority of psychiatric conditions where doctors currently prescribe medical cannabis, no reliable evidence exists that it works.
A Booming Market With a Thin Evidence Base
Medical cannabis has been legal on prescription in the United Kingdom since November 2018. High-profile campaigns on behalf of children with treatment-resistant epilepsy initially drove that change. Since then, the market has expanded dramatically. More than 30 specialist cannabis clinics are now registered with the Care Quality Commission in England, serving an estimated 80,000 patients.
The numbers are striking. Between 2023 and 2024, private prescriptions more than doubled, rising from around 283,000 to 659,000. Yet that figure needs context. Patients typically receive repeat prescriptions every four to eight weeks, so those 659,000 prescriptions likely represent roughly 82,000 individuals rather than a surge of entirely new cases.
What is harder to contextualise, however, is the mismatch between who receives medical cannabis prescriptions and what clinical trials actually support. According to available data, approximately 42 per cent of UK medical cannabis patients receive it for psychiatric conditions such as anxiety, depression, post-traumatic stress disorder (PTSD) and obsessive-compulsive disorder (OCD). Notably, the new Lancet Psychiatry review found no significant benefit for any of these conditions.
What the Research Found on Cannabinoids for Mental Disorders
Researchers at the University of Sydney and several partner institutions conducted this systematic review. They examined RCT evidence published between 1980 and May 2025, and what they found was sobering.
For anxiety disorders, including social anxiety, the data showed no significant reduction in symptoms. For PTSD, results were equally discouraging. Furthermore, for psychotic disorders, OCD, bipolar disorder, ADHD and anorexia nervosa, researchers identified no meaningful benefit. Most strikingly, not a single randomised controlled trial has examined cannabinoids for depression.
Cannabinoids did show promise in a few areas. A combination of cannabidiol (CBD) and delta-9-tetrahydrocannabinol (THC) reduced withdrawal symptoms and cannabis use in people with cannabis use disorder. Additionally, insomnia patients showed improved sleep time, and those with Tourette’s syndrome experienced a reduction in tic severity. Researchers also recorded a reduction in autistic traits in autism spectrum disorder studies, though they rated this evidence as very low certainty.
Overall, people taking cannabinoids were significantly more likely to experience an adverse event than those on placebo. Specifically, roughly one in every seven patients was affected. The most commonly reported side effects included dry mouth, nausea, diarrhoea and dizziness.
The Prescribing Paradox
The disconnect here is considerable. Sleep problems, anxiety, depression and PTSD rank among the most common reasons people seek medical cannabis in the UK, the US and Australia. Yet the evidence base for these very indications is, at best, weak.
Professor Owen Bowden-Jones, former chair of the Advisory Council for the Misuse of Drugs (ACMD), said the findings give the “clearest indication yet” that benefits of medical cannabis may have been overstated for many conditions.
Defenders of the practice argue the picture is more complicated than clinical trials alone can capture. Dr Niraj Singh, a consultant psychiatrist with over six years of prescribing experience, points to what he describes as outstanding outcomes in patients who had exhausted every other option.
“Tens of thousands of patients would testify that it benefits them across an array of symptoms,” he said. “In my experience, it has had some incredible outcomes.”
The Lancet Psychiatry review deliberately excluded this real-world evidence, drawn from patient registries and observational studies. Authors justified this on methodological grounds: observational data cannot establish causation and is more prone to bias. Nevertheless, critics including Professor David Nutt, founder of the charity Drug Science, have pushed back on the idea that RCTs should hold a monopoly on clinical decision-making.
Medical Cannabis for Mental Health: Safety Is Not Efficacy
Proponents of medical cannabis prescribing often argue that a relatively benign safety profile justifies its use even where efficacy evidence is limited. The review offers some support for this position. While cannabinoid users experienced more adverse events overall, serious adverse events and study withdrawals did not increase significantly.
However, researchers flag a critical caveat. Most clinical trials involved registered, pharmaceutical-grade cannabinoid products. These differ markedly from the high-THC preparations increasingly common in Australian, American and Canadian markets, and becoming more available privately in the UK. Products with high THC content carry well-documented risks of increased anxiety, paranoia and, in vulnerable individuals, psychosis.
Dr Marta Di Forti, professor of drug use, genetics and psychosis at King’s College London, has raised concerns about patients with pre-existing mental health conditions receiving high-THC prescriptions. In her view, those with a history of anxiety, depression or paranoia face a greater risk of worsening symptoms and dependency if exposed to products with THC content of 10 per cent or above.
In response to these concerns, the Medical Cannabis Clinicians Society recommends that prescriptions exceeding 60 grammes per month, or those containing more than 25 per cent THC, go through peer panel review.
The Research Gap Nobody Is Filling
Perhaps the most troubling part of this debate is not what the evidence shows, but what it does not yet examine. Back in 2018, the Department of Health committed to conducting efficacy trials in children with epilepsy. According to Professor Nutt, nothing has been done. Meanwhile, pharmaceutical companies show little interest in a botanical product that cannot be patented.
The Lancet Psychiatry review authors identified 14 ongoing trials through clinical registries that could not yet contribute to the analysis. They call urgently for trials with larger and more representative samples. Importantly, many of the included studies were small, with a median sample size of just 31 participants, which significantly limits what conclusions researchers can draw.
Professor Mike Barnes, chair of the Medical Cannabis Clinicians Society, has been candid that the industry itself bears responsibility here. “The industry has not been rigorous enough in collecting and analysing patient outcomes,” he said. “Clinics have a moral obligation to collect their own data.”
What This Means for Patients Using Medical Cannabis for Mental Health
For the tens of thousands of people in the UK (alone) currently using prescribed medical cannabis for mental health conditions, these findings present a genuine dilemma. Many have tried multiple treatments without success. Others describe the relief they found as life-changing. The United Patients Alliance has warned that dismissing medical cannabis on the basis of incomplete evidence risks pushing patients back to the unregulated illegal market, where product quality, THC content and medical oversight are entirely absent.
That concern deserves to be taken seriously. Even so, it cannot substitute for a proper evidence base. The Lancet Psychiatry review does not claim that medical cannabis never helps any individual. Rather, it argues, carefully and on the available data, that routine prescribing across these indications lacks the clinical trial support we would expect for any other controlled medicine.
Consider the scale of the problem. An estimated 8.7 million people in England received antidepressants in 2023/24. Over 1.5 million adults were in contact with NHS mental health services as of January 2026. The pressure to find better options is entirely understandable. Yet the answer to a struggling mental health system is to invest properly in research that could one day tell us, with confidence, which patients might genuinely benefit and under what conditions.
(Source: WRD News)
Also Read
- Cannabis as ‘Medicine?’: Pot Propaganda, Emotive Anecdote, Marketing Manipulation, and the Side Stepping of Science
- Regulation of Cannabis as Medicinal: Playing the Game of ‘Licensed Credibility’ in a State and Federal Legislation Disconnect.
- Failing Mental Health and ‘Vote for Medicine’ Models: Cannabis Conundrum Grows