In 2022, Australian doctors lodged 575 Special Access Scheme applications to prescribe medicinal cannabis to children and teenagers with autism. That made autism the single most common reason cannabis was sought for any under-18 in the country — ahead of anxiety, ahead of epilepsy, and well ahead of ADHD, which accounted for another 131 applications of its own. Four years on, those numbers have only grown. And for every prescription that moves through a doctor’s office, the research tells us there is at least one bottle being bought by a parent without one.
Something is happening in the living rooms of Australian AuDHD families that the evidence base has not yet caught up with. This week, as the Therapeutic Goods Administration tightens its rules for paediatric cannabis prescribing and a long-awaited Australian trial prepares to report its findings, it is worth stopping to ask the harder question underneath the numbers. Not simply does medicinal cannabis work for AuDHD kids — but why are so many parents trying it anyway, and what does that tell us about the system they are living inside?
A treatment in search of evidence
The honest scientific answer, as of April 2026, is that we still do not know whether medicinal cannabis helps children with autism, ADHD or the common ground between them. The best-designed trial to date, published in Molecular Autism by Aran and colleagues in 2021, randomised 150 autistic participants aged five to 21 to either a whole-plant cannabis extract, a purified cannabidiol-and-THC combination at the same ratio, or a placebo. Half the children got the real medicine first, then crossed over. The results were more complicated than either the sceptics or the true believers wanted. The purified formulation did nothing. The whole-plant extract did appear to reduce disruptive behaviour on one clinician-rated measure, with a statistically robust effect. On the parent-rated scale, the signal was softer and less certain. It was enough to justify more research. It was not enough to justify the 575 prescriptions.
Closer to home, an Australian team led by Associate Professor Daryl Efron at Murdoch Children’s Research Institute and the Royal Children’s Hospital in Melbourne has just completed a Medical Research Future Fund-backed trial of cannabidiol in children and adolescents with severe behavioural problems and an intellectual disability — most of whom were also autistic. It is the first study of its kind in this country. The earlier open-label phase, published in 2020, found the medication was well tolerated and parents were willing to recommend it to other families. The full placebo-controlled results are expected later this year. Whatever they show, they will arrive into a marketplace that has long since stopped waiting.
What parents already know that researchers don’t
Here is the part that does not fit neatly on a TGA fact sheet. According to research reviewed by paediatricians including Efron’s team, more than half of the medicinal cannabis used by Australian children for emotional and behavioural difficulties is purchased without a medical prescription at all. In roughly a quarter of cases, the child’s treating doctor has not been told. Parents are not being reckless. They are being cornered.
Think about what the average AuDHD family in 2026 is trying to hold together. Their child is eight. They have co-occurring autism and ADHD — a combination Australia has no dedicated clinical guideline for, no National Disability Insurance Scheme primary category for, and no coordinated care pathway through. They have been on a waitlist for a paediatric psychologist for fourteen months. They have cycled through stimulants, at least one of which has been on national shortage for parts of the last eighteen months. They may have tried guanfacine, melatonin, an SSRI, maybe a low-dose atypical antipsychotic that a well-meaning specialist suggested to take the edge off the meltdowns. Sleep is a nightly negotiation. School is a fortnightly emergency. The family is exhausted, and the child is exhausted, and someone in a Facebook group says cannabidiol oil changed their son’s life.
This is the soil in which an unregulated market grows. It is not a story about gullible parents or reckless doctors. It is a story about a medical system that has, for this specific cohort, run out of tools before it runs out of need.
What the TGA is trying to do
The regulator is not ignoring this. In guidance refreshed over the past year, the Therapeutic Goods Administration has clarified that Special Access Scheme Category B applications for Schedule 8 medicinal cannabis in children under 18 can now only be approved for a maximum of 24 months at a time, and — critically — any higher-strength product must be backed by a specialist, not a general practitioner, with expertise relevant to the condition being treated. The TGA has also reminded practitioners, in language that is unusually pointed for a regulator, that exposure to THC-containing products carries real risks for developing brains, and that caution is warranted.
This is a reasonable response to a difficult problem. It is also, from the perspective of a family already paying hundreds of dollars out of pocket for an unapproved product, another door closing. The single medicinal cannabis medicine with full TGA approval in Australia remains nabiximols, licensed for spasticity in multiple sclerosis. Every other product prescribed in the country — including those going to AuDHD children — is unapproved, meaning the TGA has never formally assessed it for safety, quality or efficacy. Parents are being asked to weigh up a treatment whose evidence is incomplete, whose regulatory status is provisional, whose cost is not subsidised, and whose alternative, often enough, is nothing at all.
The AuDHD-specific complication
It would be easier to have this conversation if autism and ADHD sat in tidy separate boxes. They do not. The estimated 650,000 Australians living with both — AuDHD, as the community now calls it — experience an interaction of traits that is often more than the sum of its parts. Sensory overload feeds executive dysfunction, which feeds emotional dysregulation, which feeds burnout, which feeds more sensory overload. Stimulants that calm a purely ADHD brain can amplify anxiety in an AuDHD one. SSRIs that help one child’s rumination can flatten another’s hyperfocus until they cannot function. There is no pharmacological off-switch for the loop, and the behavioural interventions that work well for a single diagnosis can fall over when the profile is mixed.
That is the context in which families start Googling cannabidiol. It is not naive hope. It is the rational behaviour of people who have tried everything else on the official list and still cannot get their child through a school day.
What the evidence can fairly say right now
A fair summary of the research, as of this month, looks something like this. Cannabidiol — CBD, without THC — appears to be reasonably well tolerated in short-term paediatric studies. The signal for reduced disruptive behaviour exists but is inconsistent, stronger on clinician-rated measures than parent-rated ones, and is almost entirely derived from whole-plant extracts rather than isolated compounds. No high-quality trial has ever shown that medicinal cannabis improves the core features of autism or ADHD. Long-term safety data in developing brains is thin. THC-containing products carry documented risks and should not be first-line for children. And almost everything we know comes from studies in autism or intellectual disability populations — not from studies designed specifically around the AuDHD profile.
None of that is a no. It is a not yet, which is a very different thing, and the distinction matters enormously if you are the parent in the kitchen at 11pm deciding what to do tomorrow.
What would help
There is a version of this story in which the regulator, the research community and the AuDHD community all end up on the same side. It begins with the Efron trial reporting its results, transparently, with the parent-rated and clinician-rated data broken out so families can see for themselves what the effect sizes really look like. It continues with targeted funding for an Australian trial that enrols children with confirmed co-occurring autism and ADHD, not just one or the other. It includes a clinical guideline for AuDHD, published by a body like the Royal Australasian College of Physicians, that treats medicinal cannabis the way it treats every other off-label medicine — as an option worth discussing honestly, with shared decision-making, when the evidence-based alternatives have been exhausted. And it includes a Medicare pathway, or a subsidy, so that the parents who do choose this road are not pushed underground by cost alone.
None of that is radical. Most of it is already how Australia handles every other difficult paediatric medicine.
The uncomfortable takeaway
The cannabis question is not really a question about cannabis. It is a question about what we owe AuDHD children when the standard playbook runs out. It is a question about whether we can be honest enough, as a system, to admit that 575 applications and a thriving grey market represent a failure of care, not a failure of parenting. And it is a question about whether the next decade of Australian neurodevelopmental research will actually ask the questions AuDHD families are living — or whether it will keep studying autism and ADHD as if they were separate countries, while the people who live at the border keep improvising their own medicine.
If you are a parent weighing this up, the most useful thing anyone can tell you is this: do not do it in secret. Whatever you decide, tell your child’s paediatrician. Not because they will necessarily agree, but because a decision like this is safer — medically, legally, and emotionally — when it is held by more than one person. The evidence is not yet with us. But the conversation can be.
Sources
Therapeutic Goods Administration — Medicinal cannabis Special Access Scheme data
Leave a Reply