Cannabinoids and mental health: between increasing use and limited scientific data

Cannabinoids, particularly CBD (cannabidiol) and the THC (delta-9-tetrahydrocannabinol), are increasingly being prescribed to treat mental disorders and substance use disorders (SUD) in countries where the medical use of cannabis is authorized.

A large-scale systematic review and meta-analysis published in March 2026 in the Lancet reveal a more nuanced reality: despite growing adoption, the scientific evidence supporting their effectiveness remains limited and uneven.

A comprehensive review of clinical trials

The study analyzed 54 randomized controlled trials (RCTs) conducted between 1980 and 2025, involving a total of 2,477 participants. These trials examined the effects of cannabinoids on a wide range of disorders, including anxiety disorders, the post-traumatic stress disorder, theinsomnia, the autism spectrum disorders and various forms of dependency.

RCTs are considered the gold standard in clinical research, but almost half of the studies included in the analysis were judged to have a high risk of bias. As a result, the overall certainty of evidence was judged to be low for most outcomes.

The gap between methodological rigor and clinical use is striking. While prescriptions for medical marijuana continue to increase worldwide, the evidence base remains fragile and often inconclusive.

Areas where cannabinoids show promise

Despite the general lack of solid evidence, this review identified several areas in which cannabinoids have demonstrated potential therapeutic effects.

Cannabis use disorder

One of the most consistent findings concerns the cannabis use disorder. Treatments combining CBD and THC were associated with a reduction in withdrawal symptoms and overall consumption.

This substitution effect, i.e. the use of regulated cannabinoid formulations to reduce problematic consumption, is in line with harm reduction strategies already explored in other substance use contexts.

Sleep and insomnia

In patients suffering from’insomnia, Cannabinoids were associated with an increase in total sleep time, measured both by electronic devices and self-reported sleep diaries.

However, improvements in sleep quality or latency were not statistically significant, suggesting that, although users sleep longer, the restorative quality of that sleep remains uncertain.

Tourette's syndrome and tics

The analysis also revealed a reduction in the severity of tics in patients with Tourette's syndrome, especially when treated with combined formulations of CBD and THC.

This effect did not appear when either compound was used alone, suggesting a potential «entourage effect» that merits further study.

Autism spectrum disorders

Some data suggest a reduction in autistic features in people with autism spectrum disorders. The authors point out, however, that this conclusion is based on very limited data and studies with a high risk of bias.

Limited or no evidence for most disorders

For many of the conditions most frequently cited by patients and prescribers, cannabinoids showed no significant benefit over placebo. These include:

• Anxiety disorders
• Post-traumatic stress disorder (PTSD)
• Psychotic disorders
• Opioid use disorder
• Cocaine use disorder
• Anorexia nervosa

The magazine found no randomized controlled trials evaluating cannabinoids as a treatment for depression, This is one of the most common reasons for the medical use of cannabis.

In some cases, the results were not only inconclusive, but potentially worrying. For example, in people suffering from cocaine use disorder, cannabinoids have been associated with a increased cravings (the craving), which raises questions about their relevance in certain dependency contexts.

Safety profile: common but non-serious adverse events

When it comes to safety, the conclusions are just as nuanced.

Participants receiving cannabinoids were significantly more likely to have side effects (such as nausea, dizziness or dry mouth) compared with those receiving a placebo. The study estimates that’one patient in seven treated with cannabinoids will experience such effects.

However, no no increase in serious adverse events or treatment discontinuation rates. This distinction is important: while cannabinoids do not appear to present major acute safety risks under controlled conditions, their tolerability remains a matter of concern.

It is also worth noting that most of the studies have focused on pharmaceutical-grade cannabinoid products, and not on the wide variety of high-THC formulations available on the market. This limits the generalizability of the results to real-world use.

A growing gap between practice and evidence

One of the central conclusions of the study is the widening gap between clinical practice and scientific evidence. In many countries mental health disorders are among the main indications for prescribing medical cannabis. Yet for many of these disorders, data supporting efficacy are either weak or non-existent.

The authors warn that the use of cannabinoids could potentially delay or replace more effective treatments, These include cognitive-behavioral therapies, whose benefits are well established for disorders such as anxiety and depression.

The need for better research and regulation

This review calls for a significant improvement in research quality. Future studies should include:

• Larger, more representative patient samples
• Transparent communication of results
• Reduced influence of industry financing

At the same time, the authors consider it necessary to strengthen regulatory oversight and the training of healthcare professionals. Many practitioners report feeling unprepared to assess the risk-benefit profile of cannabinoid-based therapies.

Public health messages also need to evolve. As cannabinoids move from the illicit to the medical setting, communication should avoid both exaggeration and dismissal, focusing instead on evidence-based information.

Overall, the results suggest that if the cannabinoids can offer advantages in specific contexts, in particular for cannabis-related disorders, l’insomnia and certain neurological conditions, their systematic use in mental health treatment is rarely justified based on current data.