News Release

The Lancet Psychiatry: Insufficient evidence that medicinal cannabinoids improve mental health

Meta-analysis finds inadequate evidence that cannabinoids relieve depression, anxiety disorders, attention-deficit hyperactivity disorder, Tourette syndrome, post-traumatic stress disorder, or psychosis

Peer-Reviewed Publication

The Lancet

Meta-analysis finds inadequate evidence that cannabinoids relieve depression, anxiety disorders, attention-deficit hyperactivity disorder, Tourette syndrome, post-traumatic stress disorder, or psychosis.

The most comprehensive analysis of medicinal cannabinoids and their impact on six mental health disorders - combining 83 studies including 3,000 people - suggests that the use of cannabinoids for mental health conditions cannot be justified based on the current evidence. This is due to a lack of evidence for their effectiveness, and because of the known risks of cannabinoids.

The new findings, published in The Lancet Psychiatry journal, find insufficient evidence medicinal cannabinoids improve disorders overall or their symptoms, although there is a very low quality evidence that pharmaceutical tetrahydrocannabinol (THC) may lead to a small improvement in symptoms of anxiety in individuals with other medical conditions, such as chronic pain or multiple sclerosis.

Medicinal cannabinoids include medicinal cannabis and pharmaceutical cannabinoids, and their synthetic derivatives, THC and cannabidiol (CBD). Around the world, these are increasingly being made available for medicinal purposes (e.g. in the United States, Australia, and Canada), including for the treatment of mental health disorders. However, there are concerns around the adverse effects of this availability, as there is a large body of evidence indicating that non-medicinal cannabis use can increase the occurrence of depression, anxiety, and psychotic symptoms.

Professor Louisa Degenhardt of the National Drug and Alcohol Research Centre (NDARC) at UNSW Sydney, Australia, and lead author of the study says: "Our findings have important implications in countries where cannabis and cannabinoids are being made available for medical use. There is a notable absence of high-quality evidence to properly assess the effectiveness and safety of medicinal cannabinoids compared with placebo, and until evidence from randomised controlled trials is available, clinical guidelines cannot be drawn up around their use in mental health disorders." [1]

She continues: "In countries where medicinal cannabinoids are already legal, doctors and patients must be aware of the limitations of existing evidence and the risks of cannabinoids. These must be weighed when considering use to treat symptoms of common mental health disorders. Those who decide to proceed should be carefully monitored for positive and negative mental health effects of using medicinal cannabinoids." [1]

This study follows The Lancet Series on Drug Use, which includes a paper on cannabis where the authors assess the current and possible future public health impacts of the legalisation of cannabis production, sale, and use in the Americas. They summarise the overall evidence on medicinal use of cannabinoids, regulation, and how medicinal use may have affected recreational use. [2]

The authors set out to examine the available evidence for all types of medicinal cannabinoids. They included all study designs and investigated the impact on remission from and symptoms of six mental health disorders in adults: depression, anxiety, attention-deficit hyperactivity disorder (ADHD), Tourette syndrome, post-traumatic stress disorder (PTSD), and psychosis.

They included published and unpublished studies between 1980 and 2018 and included 83 eligible studies, 40 of which were randomised controlled trials (RCTs) (the others were open-label trials, where participants knew which treatment they were taking). Of the 83 studies, 42 looked at depression (including 23 RCTs), 31 looked at anxiety (17 RCTs), eight looked at Tourette syndrome (two RCTs), three were on ADHD (one RCT), 12 were on PTSD (one RCT), and 11 were on psychosis (six RCTs).

In most RCTs examining depression and anxiety, the primary reason for cannabinoid use was for another medical condition such as chronic non-cancer pain or multiple sclerosis. In the studies looking at the other four disorders, the cannabinoid was used to treat the mental health disorder. Few randomised controlled trials examined the role of pharmaceutical CBD or medicinal cannabis; most looked at THC, with or without CBD.

The authors found that pharmaceutical THC (with or without CBD) improved anxiety symptoms among individuals with other medical conditions (seven studies of 252 people), though this may have been due to improvements in the primary medical condition. The authors suggest further research should explicitly study the effects of cannabinoids on anxiety and depression.

Pharmaceutical THC (with or without CBD) worsened negative symptoms of psychosis (one study, 24 people) and did not significantly affect any other primary outcomes for the mental health disorders examined. It also increased the number of people who had adverse events (ten studies; 1,495 people) and withdrawals due to adverse events (11 studies; 1,621 people) compared with placebo across all mental health disorders examined.

The study highlights the limited evidence and the low quality of the evidence that exists around using cannabinoids for treatment of mental health conditions. There is a need for high-quality research to understand the effects of different cannabinoids on a range of outcomes for people with mental health disorders.

Professor Degenhardt says: "Cannabinoids are often advocated as a treatment for various mental health conditions. Countries that allow medicinal cannabinoid use will probably see increased demand for such use. Clinicians and consumers need to be aware of the low quality and quantity of evidence for the effectiveness of medicinal cannabinoids in treating mental health disorders and the potential risk of adverse events. Given the likely interest but scant evidence to guide patient and clinician decisions around cannabinoids for mental health, there is an urgent need for randomised controlled trials to inform whether there are benefits of cannabinoids for these indications." [1]

The authors highlight that their analysis and conclusions are limited by the small amount of available data, small study sizes, and the differences in findings between small studies. There is no recommended approach for addressing these issues in systematic reviews, but they tried to minimise them by keeping the focus of the review narrow. They also note that most studies are based on pharmaceutical cannabinoids, rather than medicinal cannabis, but plant products are most often used by those taking cannabinoids for medicinal purposes in the USA.

In a related Comment article, Professor Deepak Cyril D'Souza of Yale University School of Medicine, USA, says: "The process of drug development in modern medicine is to first demonstrate efficacy and safety in clinical trials before using the drug clinically. With cannabinoids, it seems that the cart (use) is before the horse (evidence). For cannabinoids to be used in the treatment of psychiatric disorders they should be tested in RCTs and subjected to the same regulatory approval process as other prescription medications."



This study was funded by Australian Therapeutic Goods Administration, the Commonwealth Department of Health, Australia, Australian National Health and Medical Research Council and the US National Institutes of Health. It was conducted by researchers from the National Drug and Alcohol Research Centre, UNSW, the University of Brisbane, Australia and King's College London, UK.

The labels have been added to this press release as part of a project run by the Academy of Medical Sciences seeking to improve the communication of evidence. For more information, please see: if you have any questions or feedback, please contact The Lancet press office

[1] Quote direct from author and cannot be found in the text of the Article.

[2] Paper available here (begins page 29):

A press release for this report is also available.

Peer-reviewed / Meta-analysis / People

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.