France will publish medical cannabis regulations in July 2026, Health Minister told AFP, ending a months‑long delay and starting the final administrative steps toward generalised prescribing.
Officials say the unpublished decree will define how cannabis-based medicines are evaluated and priced under France’s social security system. Once the decree appears, Haute Autorité de Santé (HAS) can issue reimbursement opinions, companies can register products with Agence nationale de sécurité du médicament (ANSM), and national prescribing can move forward. Government sources estimate that, if the decree clears Conseil d’État review and HAS meets its expected timetable, the first prescriptions under a general framework are unlikely before 2027.
The announcement follows publication of the U.CANNABIS study, a national evaluation that tracked 1,971 patients enrolled in France’s medical cannabis experimentation programme. ANSM commissioned the study, which Inserm UMR 1219 and the AHeaD epidemiology team at Université de Bordeaux completed in September 2025. The results were withheld from the public until June 2026, when cannabis industry association UIVEC secured release through a request to the Commission d’accès aux documents administratifs (CADA).
U.CANNABIS used reimbursement claims from the SNDS (national health data system) linked to the ReCann clinical registry to compare healthcare consumption in the 12 months before and six months after patients started treatment. Because it relies on administrative reimbursement data rather than self-reporting, the study measures actual changes in medication dispensing and healthcare use.
Key findings: – Sample size: 1,971 patients across five authorised indications; largest cohort was refractory neuropathic pain with 1,105 patients. – Refractory neuropathic pain: share receiving any analgesic fell from about 73% to 69% after starting cannabis treatment (≈4 fewer patients per 100). Gabapentinoid use declined ≈3 per 100; strong opioid use declined ≈1.4 per 100; benzodiazepine use declined ≈1.6 per 100. Monthly specialist chronic pain day‑hospital visits fell ≈0.9 per 100 per month. – Palliative care: 21 fewer patients per 100 were on strong opioids after initiation; average daily opioid dose fell by 13.6 defined daily doses per patient. The study notes opioid use in this group had been declining before cannabis initiation, which complicates direct attribution. – Oncology: analgesic use declined by about 12 per 100 patients after initiation. – Multiple sclerosis spasticity: non‑opioid analgesic use frequency fell by about six per 100, but dose analysis suggested an apparent increase attributed by authors to modelling artefact. – Refractory epilepsy: reductions in benzodiazepine dispensing observed but did not meet the study’s statistical thresholds, likely due to limited sample size.
Authors conclude that initiation of medical cannabis corresponds with measurable reductions in several drug classes and some healthcare contacts, though effect sizes are modest and causation cannot be established by this observational design. The study used interrupted time series methods to identify changes associated with treatment start but did not randomise patients.
Regulatory implications For HAS, the SNDS‑based measures matter because reimbursement evaluation focuses on whether cannabis‑based medicines replace or supplement existing therapies. The neuropathic pain cohort showed reductions across five treatment lines—gabapentinoids, non‑opioid analgesics, benzodiazepines, strong opioids, and specialist pain visits—which provides concrete, claims‑based evidence for HAS to weigh.
Safety data from ANSM’s pharmacovigilance monitoring across the experimentation recorded zero cases of addiction or misuse. The most recent monitoring period listed 30 adverse event reports, eight of which were serious and mainly neurological or psychiatric; none involved dependency.
Background and timeline France launched a two‑year medical cannabis pilot in 2021 as a proof of concept. Political delay and repeated extensions kept the pilot running well beyond its original term; the number of participating patients peaked near 3,000 and later dropped to roughly 700. The 2024 Social Security Finance Act made generalisation a legal requirement, and draft regulatory work accelerated in 2025 and early 2026. At Cannabis Europa Paris in February 2026, Direction Générale de la Santé and Direction de la Sécurité Sociale presented a draft decree describing evaluation and pricing processes. That draft remains unpublished pending the ministry’s July release.
Industry reaction UIVEC president Ludovic Rachou urged faster publication, noting the study is “French, public and solid” and calling for texts to appear before the end of June 2026. The ministry’s July commitment falls short of that demand but sets a concrete month for the next step.
Next steps If the July decree is published and passes Conseil d’État review, HAS will evaluate reimbursement case‑by‑case. ANSM can then process product registrations. Officials expect HAS opinions in the autumn; even with that schedule, national prescribing under a general framework will likely begin in 2027 rather than 2026. Policymakers, clinicians and manufacturers will now use the U.CANNABIS claims data alongside pharmacovigilance records to shape pricing and reimbursement decisions.
