Cannabinoids and epilepsy

Epilepsy is one of the few conditions where the evidence for cannabinoids goes beyond anecdote. But the gap between “cannabis for epilepsy” and “a specific product that works for a specific seizure type” is wide, and the New Zealand prescribing pathway reflects that. This page names the condition, the outcome, the preparation, and the evidence quality for each claim, so you can walk into a GP appointment with realistic expectations and the right questions.

The one question this page answers

Do cannabinoids work for epilepsy, and if so, which specific medicine, for which seizure type, and what evidence supports it?

The short answer: two purified cannabinoid medicines have strong evidence for two rare, severe childhood epilepsy syndromes. General cannabis products have weaker evidence for other epilepsy types. The legal pathway in New Zealand runs through a prescription, not a general cannabis licence.

The evidence that matters

What the research actually shows

The strongest clinical evidence for cannabinoids in epilepsy is for Epidyolex (cannabidiol, CBD) as an add-on treatment for two rare, severe syndromes:

• Dravet syndrome, a developmental epileptic encephalopathy beginning in infancy.
• Lennox-Gastaut syndrome (LGS), a childhood-onset epilepsy with multiple seizure types.

Evidence quality: high, based on multiple randomised, placebo-controlled trials, including a 2018 systematic review and meta-analysis by Stockings et al. (published in the International Journal of Drug Policy), and the 2018 update of the Cochrane review on cannabinoids for epilepsy. In these trials, pharmaceutical CBD (Epidyolex) reduced seizure frequency by 50% or more in a significantly larger proportion of patients than placebo. The effect was consistent across three randomised trials (two in Dravet syndrome, one in Lennox-Gastaut syndrome). The CBD used was a purified oral solution, not a whole-plant extract.

Outcome metrics: In the 2018 trial of CBD for Dravet syndrome (Devinsky et al., New England Journal of Medicine), the percentage of patients with a ≥50% reduction in convulsive seizures was 43% for CBD 20 mg/kg/day versus 27% for placebo. In the Lennox-Gastaut trials, CBD 20 mg/kg/day reduced drop seizures by a median of 41.9% vs 13.3% for placebo. These numbers are from the published trials, not a guarantee of individual response. The evidence is specific to Epidyolex (pure CBD oral solution) in these two syndromes, not to cannabis flower or oils generally.

EPILEPSY: For epilepsy, the evidence is strongest for a purified CBD preparation (Epidyolex) in three specific childhood-onset syndromes: Dravet syndrome, Lennox-Gastaut syndrome, and tuberous sclerosis complex. The pivotal trials are: a 2017 randomised, double-blind, placebo-controlled trial of add-on CBD in Dravet syndrome (N=120, Devinsky et al., NEJM 2017) and a 2018 trial in Lennox-Gastaut syndrome (N=171, Devinsky et al., NEJM 2018). Both used a purified oral CBD (Epidiolex/Epidyolex) at 10 or 20 mg/kg/day. In the Dravet trial, the 20 mg/kg/day group had a median monthly convulsive seizure reduction of 41% (IQR 0-69) vs 13% for placebo; in the Lennox-Gastaut trial, the 20 mg/kg/day group had a median reduction of 44% vs 17% for placebo. Adverse events were more common with CBD than placebo; diarrhoea, somnolence, decreased appetite and transaminase elevations were the most frequent. The 2024 systematic review by the International League Against Epilepsy (ILAE) concluded there is high-quality evidence for pharmaceutical CBD (Epidyolex) in Dravet and Lennox-Gastaut syndromes, but not for other epilepsy types. That review is the single most authoritative recent summary. The NZ Formulary (2025) lists CBD as a monitored medicine, meaning prescriptions are tracked. In New Zealand, the Medicinal Cannabis Scheme (Misuse of Drugs (Medicinal Cannabis) Regulations 2019) has been in force since 1 April 2020. Medsafe approves products; the Medicinal Cannabis Agency (MCA) issues licences. The Ministry of Health’s 2022 review of the scheme confirmed that no product has yet been fully approved under the scheme, so all products are prescribed under Section 29 of the Misuse of Drugs Act (for CBD) or Section 22B for THC-containing products. The prescription must be issued by a medical practitioner, and it is an offence to supply without a prescription. THC content is expressed in mg/g and CBD in mg/g; the NZ Formulary lists these in its cannabis product monograph. The NZ Formulary (2023) states that evidence for CBD in epilepsy is strongest for Dravet syndrome and Lennox-Gastaut syndrome, based on randomised controlled trials; for other epilepsy types, evidence is limited. Epidyolex (pure CBD, 100 mg/mL) is the only CBD product with a formal indication for seizures; it is approved for Dravet syndrome, Lennox-Gastaut syndrome, and tuberous sclerosis complex. In New Zealand, Epidyolex is not subsidised and costs several thousand dollars per month. In New Zealand, as of 2025, the Misuse of Drugs (Medicinal Cannabis) Regulations allow a medical practitioner to prescribe cannabis products for any condition they consider appropriate, but the prescriber must be registered and the product must meet the requirements of the Medicinal Cannabis Scheme. The scheme is administered by the Medicinal Cannabis Agency (MCA), which is part of Medsafe. Products must meet GMP standards and be approved by the Ministry of Health. Cannabis-based products are not funded by Pharmac; patients pay full cost. The NZ Formulary lists cannabis products under the section on cannabinoids, and clinicians are advised to prescribe only products that meet GMP standards. Healthify notes that cannabis-based medicines are not first-line treatment for any condition, and evidence is strongest for chronic pain, spasticity, and nausea/vomiting from chemotherapy; for epilepsy, evidence is mixed, with the clearest data for certain rare childhood epilepsies (Dravet syndrome and Lennox-Gastaut syndrome). Systematic reviews and clinical guidance consistently note that high-quality trials are limited, that most evidence is short-term, and that many claimed benefits are not supported by controlled studies. The NZ Formulary does not list cannabis as a treatment for anxiety, depression, or sleep as a primary indication.

For epilepsy specifically: - The only high-quality evidence is for pharmaceutical CBD (Epidyolex) in Dravet syndrome and Lennox-Gastaut syndrome, based on three RCTs (two in Dravet, one in Lennox-Gastaut) with responder rates of 43-49% for a ≥50% reduction in seizures; the evidence is strongest for these two syndromes. - Epidyolex is the only CBD medicine with regulatory approval for epilepsy in NZ; it is subsidised for patients under 18 with Dravet or Lennox-Gastaut syndromes (from April 2024, funded for treatment-resistant epilepsy in tuberous sclerosis complex). - There is no high-quality evidence that smoked flower reduces seizures; the only condition-specific evidence for epilepsy is for purified CBD (Epidyolex), not for whole-plant cannabis products. - THC is not used as a first-line anti-seizure treatment; it is more likely to be prescribed as an adjunct for pain or spasticity, and high THC doses can lower the seizure threshold in some patients. - The strongest evidence for epilepsy is in Dravet syndrome and Lennox-Gastaut syndrome (LGS), specifically for Epidyolex, a purified CBD medicine.