This article covers what research shows about cannabis for Tourette syndrome and autism. Both conditions start in childhood. A third condition, ADHD, is covered in Part 2. The evidence ranges from one published patient to a randomised trial with 150 participants. None of it is a treatment recommendation.
Many families see only the encouraging results, because those are the ones quoted in material that promotes cannabis. This article puts the positive and negative results side by side, so you can see how little is settled.
Read this first.
This article is not medical advice. It does not recommend giving cannabis to any child. Treatment decisions for a child with one of these conditions belong to a specialist who knows the child. That means a paediatrician, a child psychiatrist or a paediatric neurologist (a doctor for children’s brain and nerve conditions).
Nobody knows the long-term effects of cannabinoids on the developing brain. Cannabinoids are the active compounds in cannabis. Separately, large population studies link cannabis use in the teenage years with a higher risk of psychosis. Psychosis is a mental illness in which a person loses touch with reality.
This article gives no doses. Be sceptical of anyone who does.
Translator’s note: In Thailand, cannabis flower is a controlled herb. It may be sold only against a PT 33 (ภ.ท.33) prescription from an authorised practitioner. Using it for a minor is a decision for a specialist.
Research Words Used in This Article
These terms come up in the trial descriptions below. Each one affects how much weight a result deserves.
- THC and CBD. THC (tetrahydrocannabinol) is the cannabinoid that causes the “high”. CBD (cannabidiol) does not cause a high. Medical products often combine the two in a fixed ratio.
- Randomised controlled trial (RCT). Participants are split into two groups by chance. One group gets the real product. The other gets a placebo, which is a dummy that looks the same. The difference between the groups shows what the product itself does.
- Double-blind. Neither the participants nor the researchers know who got the real product until the trial ends. This keeps hopes and expectations from shaping the results.
- Crossover trial and washout. Each participant takes the real product for one period and the placebo for another. Between the two periods there is a washout, a break with no treatment, so the first product can leave the body.
- Pilot trial. A small first trial that tests whether a larger trial is workable. Its results are a hint, not proof.
- Primary outcome. The main measure that decides whether the product worked. Researchers choose it before the trial starts. If it fails, the trial has not shown a benefit, even when other measures look good.
- Statistically significant (p-value). The p-value estimates how often a difference this large would appear by luck alone if the product did nothing. By convention, a value of about 0.05 or lower counts as significant.
When you read a claim about cannabis for these conditions, look for these words. A post that mentions no placebo group and no primary outcome is describing someone’s experience. It is not describing a tested result.
1. Three Conditions With Very Different Levels of Evidence
Some books and websites that campaign for cannabis discuss ADHD, autism and Tourette syndrome together. That makes the evidence look equally strong for all three. When you look at each condition on its own, the three are very different.
The table uses a grade from NASEM, the US National Academies of Sciences, Engineering, and Medicine. In 2017, NASEM reviewed the evidence on cannabis for many conditions (see section 2).
| Condition | Best evidence so far | How strong |
|---|---|---|
| Tourette syndrome | Two early small RCTs: tics improved. Three adult trials in 2023: one positive, two missed their primary outcome. NASEM (2017): “limited evidence” for THC capsules | Small trials, mixed recent results |
| Autism | One placebo-controlled crossover RCT in 150 children and young people, with mixed results. Smaller trials of CBD-rich extracts in children | Largest trial, but it measured behaviour, not autism itself |
| ADHD (Part 2) | One pilot RCT in adults. Primary outcome not significant | Weakest. The explanation of how it would work is only a hypothesis |
Whenever someone claims cannabis helps “ADHD and autism” together, ask which condition the study actually covered. Then ask whether the participants were children or adults. Only the autism trials studied children.
2. Tourette Syndrome
Tourette syndrome is a neurological condition that causes tics. Tics are repeated movements and sounds the person cannot control. Severe cases include jerking limbs, shouting, inappropriate words and sometimes spitting. This is distressing for the person and for their family.
The early reports
In 1998, a research team led by Müller-Vahl interviewed 64 patients with tics. The same researcher later led the 2023 German trial below. Of the 64 patients, 17 said they used cannabis, and 14 of these reported fewer tics. They also reported weaker premonitory urges, which are the uncomfortable feelings just before a tic. They also reported fewer obsessive-compulsive symptoms, meaning unwanted repeated thoughts and the urge to repeat actions.
A separate report in 1999 described a 25-year-old man with tics and ADHD. He also had OCD (obsessive-compulsive disorder) and hurt himself. One dose of THC lowered his tic score on the Tourette’s Syndrome Global Scale from 41 to 7, two hours after the dose. This scale is a doctor-rated measure of how severe tics are. A higher score means worse tics.
His premonitory urges also fell. Psychophysiological tests (measures of attention and reaction) showed better signal detection, meaning he was better at picking out a target among distractions. They also showed he could keep his attention up for longer.
Why these reports prove little
The 1998 study asked people who had chosen to use cannabis to look back and report on it. So every weakness of personal testimony applies. The patients selected themselves, memory is unreliable, and there was no comparison group.
The 1999 report covers a single patient. A dramatic result in one patient is a reason to run a trial. It is not a conclusion.
What the controlled trials show
A systematic review collects and assesses all trials on one question. One review looked at RCTs of cannabis-based medicines for movement disorders. It found two small RCTs in Tourette syndrome, and both showed fewer tics.
In 2017, NASEM concluded there is limited evidence that THC capsules improve Tourette syndrome symptoms. “Limited” is a defined grade in that review. It is well below “substantial”, the grade given to chronic pain. It is also below “conclusive”, the grade given to nausea from chemotherapy. But it is above “insufficient”, the grade several other conditions received.
Three trials published in 2023
- CANNA-TICS, by Müller-Vahl and colleagues (Psychiatry Research, 2023). This double-blind trial ran for 13 weeks at several centres in Germany. It compared nabiximols with placebo in 97 adults with Tourette syndrome or chronic tic disorders. Nabiximols is a prescription mouth spray made from cannabis extract that contains both THC and CBD. More patients on nabiximols reached the trial’s threshold for a real improvement: about 22% versus 9%. But the primary outcome was not statistically significant, so the trial did not show that nabiximols works better than placebo.
- Mosley and colleagues (NEJM Evidence, 2023). This crossover trial compared an oral THC:CBD oil with placebo in 22 adults with severe Tourette syndrome. Tic scores fell significantly more on the real oil. Some participants had slowed thinking, memory lapses and poor concentration.
- Abi-Jaoude and colleagues (Cannabis and Cannabinoid Research, 2023). This pilot crossover trial tested single vaporised doses of three cannabis products against placebo. Vaporised means heated until it gives off vapour, without burning. There were 12 adults, and 9 finished the trial. The primary outcome, tics rated from video recordings, showed no significant effect. Most participants could tell whether they had received cannabis or placebo.
Every one of these trials was in adults.
Overall, the evidence is small and mixed. The two early trials and the Mosley crossover trial were positive. The largest trial so far missed its primary outcome. That is far from an established treatment.
Someone may show you the 1998 or 1999 reports as proof. Ask them two questions: was there a placebo group, and were the patients children? For both reports the answer to both questions is no. Use the 2023 trials instead to judge where the evidence actually stands.
3. Autism
Autism spectrum disorder is a lifelong developmental condition that affects social communication and behaviour.
One mother’s account
Some campaign writing that promotes cannabis reproduces a mother’s account. She describes legally treating her autistic son with medical cannabis over several years. In her account, it eased his stomach pain and reduced his violent behaviour. The family was able to stop the psychiatric medicine he had been taking for violent outbursts.
She counted it a success, but only in a narrow sense. Cannabis did not cure her son, who was 11 at the time. It eased some severe symptoms enough that the family’s daily life was no longer run by his outbursts.
Her claim is careful and limited, and you should read it the way she put it. It is also one family’s story, and nobody can check it independently.
The largest trial: Aran and colleagues, 2021
Since then, a proper trial has been run. Aran and colleagues published it in Molecular Autism in 2021. It was a double-blind, placebo-controlled crossover trial at a single centre in Israel.
- Participants: 150 children and young people aged 5–21 (average age 11.8), 80% boys. All had autism spectrum disorder and disruptive behaviour.
- Products: two cannabinoid preparations, a whole-plant extract and purified cannabinoids. Both contained CBD and THC in a 20:1 ratio.
- Schedule: the dose was raised gradually over a 12-week period. After a 4-week washout, participants switched to a second 12-week period.
- Two co-primary outcomes (two main measures, both chosen in advance):
- the Home Situations Questionnaire-ASD (HSQ-ASD), where parents rate how disruptive the child’s behaviour is in everyday home situations;
- the Clinical Global Impression-Improvement scale (CGI-I), where a clinician rates the child’s overall change, from “very much improved” to “very much worse”.
The results, positive and negative
- HSQ-ASD: no significant difference between the groups. One of the two main measures was negative.
- CGI-I: 49% were much or very much improved on whole-plant extract, versus 21% on placebo (p = 0.005).
- The Social Responsiveness Scale, a questionnaire on social behaviour, improved by 14.9 points on whole-plant extract versus 3.6 on placebo (p = 0.009).
- Safety: no serious adverse events (serious harmful effects) were judged to be caused by treatment. Six participants had serious events that were judged unrelated. Somnolence (sleepiness) and reduced appetite affected about 21–28% of participants on cannabinoids, versus 8–15% on placebo.
Other trials in children
Aran’s trial is not the only randomised trial in children. There are also smaller trials of CBD-rich extracts. For example, da Silva Junior and colleagues in Brazil gave a CBD-rich extract or placebo to 60 children aged 5–11 for 12 weeks. The trial was published online in Trends in Psychiatry and Psychotherapy in 2022. The researchers reported improvements in social interaction, anxiety and agitation.
Across the field, results are mixed. None of these trials is large or long.
What this adds up to
Aran 2021 is by far the strongest study in this article. Its result is genuinely mixed. One main measure was negative and the other positive, in a trial large enough to take seriously.
This is a real but partial signal on behaviour. It is not a signal on autism itself. Cannabinoids do not treat autism.
If you are a parent thinking about this, take the Aran 2021 paper to your child’s specialist. Ask which one behaviour you would be trying to improve, and how you would both measure it. That turns a general hope into a specific question the specialist can answer.
This article is part of the Asiannabis Community educational series on medical cannabis. It is general information, not medical advice. It is not a recommendation to use cannabis in children. The long-term effects of cannabinoids on the developing brain are unknown.
References for Part 1
- Müller-Vahl KR, Kolbe H, Schneider U, Emrich HM. Cannabinoids: possible role in patho-physiology and therapy of Gilles de la Tourette syndrome. Acta Psychiatrica Scandinavica, 1998: interviews with 64 patients with tics.
- Müller-Vahl KR, Schneider U, Kolbe H, Emrich HM. Treatment of Tourette’s syndrome with delta-9-tetrahydrocannabinol. American Journal of Psychiatry, 1999 (letter): THC in a 25-year-old man with tics, ADHD, OCD and self-injury.
- Oikonomou P, Jost WH. Randomized controlled trials on the use of cannabis-based medicines in movement disorders: a systematic review. Journal of Neural Transmission, 2022.
- National Academies of Sciences, Engineering, and Medicine (NASEM), 2017: review of the health effects of cannabis and cannabinoids.
- Müller-Vahl and colleagues. CANNA-TICS. Psychiatry Research, 2023.
- Mosley and colleagues. NEJM Evidence, 2023.
- Abi-Jaoude and colleagues. Cannabis and Cannabinoid Research, 2023.
- Aran and colleagues. Molecular Autism, 2021.
- da Silva Junior and colleagues. Trends in Psychiatry and Psychotherapy, published online 2022.
Continued in Part 2: Cannabis and Neurodevelopmental Conditions in Children: ADHD, the Gaps in All the Trials and the Thai Rules (Part 2 of 2)