For many people living with anxiety, depression or trauma, medicinal cannabis has become a symbol of hope. But hope and evidence are not always the same thing.

Over the past decade, prescriptions for medicinal cannabis have increased dramatically in countries including Australia, Canada, the United States and parts of Europe. Many people report that cannabis products help them feel calmer, sleep better or cope with difficult emotions.

Yet one of the largest scientific reviews ever conducted has reached a far more cautious conclusion.

Researchers from the University of Sydney and several international institutions examined every high-quality randomized clinical trial they could find, asking a straightforward question: Does medicinal cannabis actually work for mental health disorders? The answer was more nuanced than either advocates or critics might expect.

Looking at the strongest evidence, the researchers conducted a systematic review and meta-analysis of 54 randomized controlled trials involving 2,477 participants published between 1980 and 2025. Randomized controlled trials remain the highest standard for determining whether a treatment truly works.

For many of the conditions most commonly treated with medicinal cannabis, the evidence simply was not convincing. The review found no meaningful benefit for anxiety disorders, depression (where no randomized trials existed), post-traumatic stress disorder, psychotic disorders, opioid use disorder or anorexia nervosa. There was also insufficient evidence for ADHD, bipolar disorder and obsessive-compulsive disorder.

Despite depression being one of the most common reasons patients seek medicinal cannabis, the researchers found no randomized clinical trials evaluating cannabinoids as a primary treatment for depression. In other words, clinical practice has expanded much faster than the scientific evidence.

There were, however, some areas where cannabinoids showed promise. The review found evidence that cannabinoid medicines may reduce withdrawal symptoms in cannabis use disorder, modestly increase total sleep time in insomnia, reduce tic severity in Tourette syndrome and improve some measures of autistic traits. Importantly, the certainty of this evidence was generally low or very low, meaning future research could substantially change these conclusions.

Safety was equally important. Across all conditions, participants receiving cannabinoids experienced more adverse effects than those receiving placebo. Common side effects included dizziness, dry mouth, nausea, diarrhoea and fatigue. Overall, approximately one additional person experienced an adverse event for every seven people treated, although serious adverse events were not increased.

Many patients genuinely report that medicinal cannabis helps them. Their experiences should not be dismissed. However, personal experience and scientific evidence answer different questions. Symptoms naturally fluctuate, expectations influence perception, and other treatments may contribute to improvement. Carefully designed randomized trials are intended to separate these influences from the effects of the medication itself.

The authors expressed concern that expanding use of medicinal cannabis could unintentionally delay or replace treatments with stronger evidence, such as psychological therapies or established medications. Their conclusion was not that medicinal cannabis has no role, but that its use should be guided by high-quality evidence rather than assumptions.

Patients currently using medicinal cannabis should not stop treatment based solely on this study. Instead, it provides an opportunity to discuss with their healthcare professional what evidence exists for their particular condition, what benefits can realistically be expected, what side effects may occur, and whether other evidence-based treatments should also be considered.

Lydia™ Reflection: Science rarely gives simple yes-or-no answers. This study does not conclude that medicinal cannabis is either good or bad. Rather, it reminds us that for many mental health conditions we simply do not yet have strong evidence that it works. Hope is important, but hope is strongest when it is supported by careful research. Sometimes the wisest path is neither unquestioning enthusiasm nor outright rejection, but the patience to follow where the best evidence leads.

Further Reading & Sources:
• Wilson J, Dobson O, Langcake A, et al. The efficacy and safety of cannabinoids for the treatment of mental disorders and substance use disorders: a systematic review and meta-analysis. The Lancet Psychiatry. 2026.
• University of Sydney. No evidence to suggest medicinal cannabis is effective for depression, anxiety and other mental health conditions.