Cannabis in Switzerland: what the pilot trials really show, and what it changes for CBD
Over-the-counter CBD, cannabis prescribed by a doctor and recreational cannabis tested in seven pilot trials are three different things. Here is what the first official and scientific results say, with their limits.
Between over-the-counter CBD, cannabis prescribed by a doctor and pilot trials reserved for a few thousand adults, the Swiss cannabis framework is often misunderstood. In September 2026, two important sets of results came out: an interim report from the SCRIPT study (Bern, Biel, Lucerne) and a one-year analysis of the Weed Care project in Basel. Added to these are the federal synthesis published in January 2026 and the figures from Lausanne.
We take no side in the debate on legalisation, and these trials are not about CBD: they concern THC-rich cannabis, sold to adults who already use it. But they shed light on the context in which our sector operates, and they show how to read preliminary results without making them say more than they do.
01Three legal statuses, three regimes
The starting criterion is the THC content, the psychoactive substance of cannabis. Products whose total THC content is below 1% are not subject to the Narcotics Act (NarcA), unlike cannabis with a higher THC content (FOPH).
| Status | Condition | Framework |
|---|---|---|
| CBD products | Total THC content below 1% | Outside the Narcotics Act. Depending on the product, other laws apply (food, chemicals, product safety). |
| Medical cannabis | Prescribed by a doctor | Authorised without an exceptional permit from the FOPH since 1 August 2022. Production and trade controlled by Swissmedic. |
| Recreational cannabis | THC of 1% or more, non-medical use | Prohibited, except for adults admitted to a scientific pilot trial (art. 8a NarcA). |
For hemp flowers, it is the total THC content that counts, THC and THCA added together, according to the federal guidance on products containing CBD.
02Medical cannabis: what changed on 1 August 2022
Before 2022, a doctor who wanted to prescribe cannabis had to obtain an exceptional permit from the Federal Office of Public Health (FOPH). On 22 June 2022, the Federal Council decided to lift this ban in the Narcotics Act, with effect from 1 August 2022 (Federal Council press release). Since then, the doctor prescribes on their own responsibility, by means of a narcotics prescription.
The cultivation, manufacture and trade of cannabis for medical use are subject to Swissmedic's authorisation and control system, like other medical narcotics, and commercial export is possible with a Swissmedic permit. Doctors must also send the FOPH, through an online reporting system, data on the treatment during the first years of therapy, in order to follow the evolution of prescriptions and document their effects. No reporting is required for medicines based on pure CBD or on extracts whose total THC content is below 1% (FOPH).
Recreational cannabis, for its part, has remained prohibited outside the pilot trials.
Prescribed does not mean proven
Lifting the ban makes medical cannabis easier to access; it does not settle the question of its efficacy. The 2026 Cochrane and Annals of Internal Medicine reviews on pain are detailed in our dossier on benefits and research.
03The pilot trials in brief: seven projects, about 10,400 adults
Article 8a of the Narcotics Act, in force since 15 May 2021, allows the controlled sale of non-medical cannabis to be tested scientifically within a limited framework. Each trial is limited to five years (extendable by two) and to 5,000 participants. Products are grown in Switzerland to organic standards, with THC content capped at 20%, no advertising and limited purchase quantities. Participants must be of age and already users, and may not consume in public spaces.
In June 2025, about 10,400 people were taking part in the seven trials, a little under 5% of the roughly 220,000 people who reported in 2022 having used cannabis in the previous month, according to the Federal Statistical Office (federal synthesis).
| Trial | Place | Sales model | Participants |
|---|---|---|---|
| La Cannabinothèque | Vernier (GE) | Non-profit membership shop | approx. 1,300 |
| Cann-L | Lausanne | Non-profit specialised shop | approx. 1,500 |
| SCRIPT | Bern, Biel, Lucerne | Pharmacies | 762 (end of 2024) |
| Cannabis Research Zürich | Zurich | For-profit specialised shops | approx. 3,200 |
| Weed Care | Basel-Stadt | Pharmacies | 314 |
| ZüriCan | Zurich | Pharmacies, information centre, social clubs | approx. 2,217 |
| Grashaus | Basel-Landschaft | For-profit specialised shops | approx. 1,100 |
Participants in June 2025 according to the FOPH, except SCRIPT (end of 2024). Weed Care had 378 participants at the start; SCRIPT totals 1,177 since the end of its recruitment.
04What the federal synthesis says
The FOPH commissioned the University of Lausanne and the University of Applied Sciences and Arts Northwestern Switzerland to compare the seven trials. The second report, dated 17 November 2025, covers the period up to mid-2025 and was made public in January 2026. Its authors are political science specialists, independent of the trials.
No notable incident, a shift towards the legal market, signals that are still fragile
No notable negative effect on health or public order is reported, and the police have passed on no information about problems linked to participants or products.
For most participants, legal points of sale have become the main source of supply. In three non-profit trials (La Cannabinothèque, Cann-L, Weed Care), between half and two thirds of participants obtain their cannabis exclusively or almost exclusively on the legal market. At Grashaus (for-profit shops), illegal purchases fell from 20 to 10 days per month on average.
High-THC products remain the most in demand, but interest in vaporisers, e-liquids and lower-dose products is growing. The average amount of THC bought per person per month remained relatively stable over the first twelve months.
The first analyses show an improvement in anxiety and depression symptoms at Weed Care and Grashaus. The trial managers explain this by the end of the stress linked to illegal supply and by destigmatisation. In Basel, problematic use falls after one and then two years, especially among those whose use was already problematic.
Source: FOPH executive summary, UNIL and FHNW, November 2025. Details at the end of the article.The authors' own reservations
- The period is short, so results on social and health aspects remain limited.
- Participants are not representative: 78% men, about a third university graduates, young people and light users under-represented.
- Satisfaction is measured in only three trials.
- The offer tested is more restricted than that of a commercial market.
- International experience suggests expecting a slight rise in use at the start of regulation, and effects that only become visible after several years.
05Basel: Weed Care, the randomised trial
The six-month randomised result (Addiction, 2025)
Weed Care is designed as a randomised trial (protocol). 378 adult regular users were recruited between August 2022 and March 2023, 374 of them with usable baseline measurements. Half had access to cannabis in pharmacies from the start; the other half kept buying on the illegal market for six months. The trial was open-label: participants knew which group they were in. The primary outcome was a cannabis misuse score (CUDIT-R, from 0 to 32) at six months.
Result in the article published in Addiction: 10.1 points in the pharmacy group versus 10.9 in the control group, a difference of 0.69 points, with a 95% confidence interval from -1.4 to 0.0 and a p-value of 0.052. The authors speak of indications of a difference, not of an established effect. A subgroup analysis suggests that the decrease is limited to participants who also used other drugs. No secondary outcome (depressive, anxiety or psychotic symptoms, amount of cannabis, alcohol, other drugs) varied in a statistically significant way.
What can be taken from it: at six months, no statistically detectable worsening, and a signal of modest, uncertain benefit on the primary outcome.
The two- and three-year follow-up (canton communications)
After six months, the control group also gained access to the pharmacies: what follows is an observational follow-up, without a comparison group. According to the Basel-Stadt Department of Health, 87 kg of cannabis had been sold after two years, a third of it products with less than 13% THC, which is less than the usual black-market products. The canton estimates at about CHF 900,000 the sales taken away from the illegal market. It states that neither days of use nor quantities had increased, and that depressive symptoms, anxiety and addictive behaviour had decreased. 299 of the 378 participants were still active.
After three years, 265 people still were; the canton reports a clear drop in the use of joints mixed with tobacco, stable quantities bought per person, and the extension of the study until January 2027 (Keystone-SDA). These are press releases from an administration, not scientific publications.
The one-year analysis (Drug and Alcohol Dependence, September 2026)
According to the Hanf Magazin report of 23 September 2026, an analysis published in early September in Drug and Alcohol Dependence followed 374 participants for one year. The depressive symptom score (PHQ-9) goes from 5.37 to 4.53 points on average, and anxiety and problematic use both fall slightly. These decreases are statistically significant but small, and the share of people above clinical thresholds remains about the same. Participants use slightly more often than at the start, without an increase in the total quantity. This analysis has no comparison group: according to the same report, its authors themselves write that no causal link can be drawn from it.
Press releases, syntheses and peer-reviewed articles do not always say the same thing with the same nuance. That is one more reason to look at who measured what, and with which comparison group.
06Bern, Biel, Lucerne: SCRIPT
SCRIPT (Safer Cannabis Research In Pharmacies Trial) is a randomised trial supported by the Swiss National Science Foundation and run by the University of Bern. Its interim report, dated 4 September 2026 and presented to the press on 7 September, has not yet been peer-reviewed. Its authors chose to publish it ahead of the scientific journal, after a broadened review (researchers, authorities, specialised services, participants), because the study sits at the interface between science and politics.
1,177 adult regular users were randomly assigned: half could buy cannabis in pharmacies immediately, the other half after a six-month wait. Products are limited to 20% THC, sold in plain packaging, with a high simulated tax and caps of 2 g of THC per purchase and 10 g per month. The report presents the six-month results.
- Frequency of use: it changed little, for cannabis as for tobacco, in both groups. Nothing indicates that pharmacy access modified it.
- Form of use: the share of participants who exclusively smoked their cannabis goes from 78% to 58% in the authorised group, whereas it barely moved in the control group (82% at the start). Vaping, oral products and mixed use increase. The share of people who stopped is comparable in both groups.
- Supply: in the authorised group, 51% of participants had used smoked products from the study in the last seven days, and 40% smoked products from the illicit market, the two sources possibly being combined. The authors point out that SCRIPT is not meant to drive out the illicit market: prices aligned with that market plus a simulated tax, plain packaging, THC cap.
- Satisfaction: 87% of the authorised group find buying in a pharmacy pleasant, 71% are satisfied with the advice, 79% feel they have enough privacy.
- Safety: 45 participants (4%) reported 51 serious adverse events between enrolment and the six-month visit: 26 in the authorised group (24 people) and 25 in the control group (21 people), mostly linked to medical interventions in hospital. For one event, a link with the intervention cannot be excluded at this stage, and validation by an independent committee is pending.
- Illicit market: 127 illicit cannabis samples brought by participants were analysed, with on average 15% THC for flowers and 22% for resin. Of 69 participants who had estimated the content of their sample, only one had estimated it correctly. Nine samples showed anomalies linked to synthetic cannabinoids. These samples were handed in voluntarily, mostly by acquaintances: they do not represent the illicit market as a whole.
The study continues until 2029, with pharmacy sales extended until 31 October 2028 (extension authorised by the FOPH in September 2025).
07Lausanne and Geneva
Cann-L (Lausanne). In its May 2026 report, the Vaud Council of State indicates that about 1,700 people have registered since December 2023 (77% men, 36 years old on average). The volumes of cannabis bought illegally by participants would have fallen from about 14.7 kg to 5.5 kg per month (2024 data), and about 70% of registrants stated in 2024 that they obtained their supply exclusively or almost exclusively from Cann-L. Addiction Suisse estimates that its sales would represent about 18% of Lausanne's consumption. No complaint has been recorded by the cantonal police, and batch analysis certificates are available at the point of sale. These figures are estimates from an ongoing scheme, not results of a randomised trial.
La Cannabinothèque (Vernier, Geneva). About 1,300 participants in June 2025 according to the FOPH. The preliminary data published in June 2026 in the Revue médicale suisse describe a mostly male cohort, about 38 years old on average, made up of frequent users motivated by product quality, with no notable adverse event to date.
08What these trials cannot say
Five limits to keep in mind
- Who is studied: adults who already used regularly. Nothing about young people, non-users, or about starting use.
- What is sold: a restricted offer, capped at 20% THC, with no advertising. The effects of a commercial market cannot be deduced from it.
- How long: from six months to three years depending on the trial. The effects of regulation are measured over years.
- How it is measured: mostly by self-report. The randomised result published in Basel is borderline on its primary outcome, and the longer-term improvements come from a follow-up without a comparison group.
- What status: the SCRIPT report has not yet been peer-reviewed, and its own team points out that its results concern individual effects in users, not the effects of general regulation.
09Where the law stands
Parliament has not legalised cannabis. A preliminary draft law on cannabis products (LPCan), arising from the Siegenthaler parliamentary initiative, is being worked on. The National Council's Social Security and Health Committee (SSHC-N) put it out for consultation from 29 August to 1 December 2025, and the report on the consultation results is dated 8 May 2026 (FOPH). The preliminary draft notably provides for cultivation for personal use and the purchase of cannabis products by adults, non-profit sale in a limited number of points of sale under cantonal licence, a ban on advertising, an incentive tax and the maintenance of zero tolerance at the wheel.
In May 2026, the SSHC-N decided to continue the work (16 votes to 8 with 1 abstention) and to send the project back to a sub-committee. It refused, by 17 votes to 8, to suspend it while awaiting the final results of the pilot trials. Several cantons are asking for more resources for implementation, and online sales and measures for minors must be reworked (GREA).
Positions remain divided. According to RTS, the SP and the Greens support the project, while the SVP and The Centre oppose it, The Centre judging in particular that youth protection is not achieved (watson). Among the professional organisations consulted, only the Blue Cross rejects the project.
As for the timetable, the authors of the SCRIPT study mention in their September 2026 report that the future law would come into force in 2031 at the earliest. Until then, recreational cannabis remains prohibited outside the pilot trials.
10And CBD in all this?
CBD whose total THC content is below 1% remains outside the Narcotics Act: according to the FOPH, only THC is controlled by this law. CBD products fall under other texts depending on their nature (food, chemicals, product safety), as detailed in the federal guidance.
The pilot trials, for their part, do not concern CBD: their products contain up to 20% THC, twenty times the 1% limit applicable to CBD products, and are only sold to adults who already use cannabis. Their results, whether on THC use, mental health or the illicit market, therefore do not transfer to a CBD oil or flower.
What a CBD buyer can still take from it
- A label is not an analysis. In SCRIPT, only one participant out of 69 had correctly estimated the THC content of their sample, and in Lausanne batch analysis certificates are available at the point of sale. The reflex applies to CBD: a stated percentage only has value if verified by a laboratory analysis. We detail this point in our article on concentrations.
- Terpenes and the entourage effect. The federal evaluators note that two for-profit trials put these notions forward in their commercial messaging, specifying that no scientific evidence supports them. Our reading is a little more nuanced: the hypothesis rests on real preclinical data, but it is not established clinically in humans. See our page on the entourage effect.
- Residual THC and the road. In SCRIPT, participants are not allowed to drive under the influence of cannabis, and the preliminary draft law maintains zero tolerance at the wheel. If you use a Full Spectrum, the question of THC traces and tests is covered in our article Full Spectrum or Broad Spectrum.
11Frequently asked questions
Is cannabis legal in Switzerland in 2026?
Not for recreational use: it remains prohibited, except for adults admitted to a scientific pilot trial. Medical cannabis has been authorised on prescription since 1 August 2022. CBD products whose total THC content is below 1% are not subject to the Narcotics Act.
Since when has medical cannabis been authorised in Switzerland?
Since 1 August 2022, a doctor can prescribe medical cannabis without requesting an exceptional permit from the FOPH. They must, however, report data on the treatment during the first years of therapy.
Do the pilot trials concern CBD?
No. They concern THC-rich cannabis, up to 20%, sold to adults who already use it. Their results say nothing about the efficacy or safety of a CBD oil or flowers with less than 1% THC.
Do the pilot trials prove that legalisation would be risk-free?
No. Their own authors stress that results are still limited: short duration, unrepresentative participants, no young people or non-users, a more restricted offer than a commercial market. They note the absence of a notable negative effect at this stage, which is not the same as proof of safety.
Can one still take part in a pilot trial?
Conditions vary by trial: being of age, already using cannabis regularly and living in the city or canton concerned. Recruitment has ended or is limited in several trials, SCRIPT's having ended in August 2025. Ask the trial in your region directly.
When could cannabis be legalised in Switzerland?
No date is set. The bill is being worked on in the National Council committee. The authors of the SCRIPT study mention the future law coming into force in 2031 at the earliest.
Are the SCRIPT results final?
No. They cover the first six months, have not yet been peer-reviewed and will be completed until 2029. The safety analyses are awaiting validation by an independent committee.
A question about the status of your CBD product?
Our team can answer your questions on analyses, THC content and the legal framework.
Contact our customer serviceSources
- Federal Council. Access to cannabis for medical use will be made easier for patients, press release of 22 June 2022. Read the press release
- FOPH. Cannabis-based medicines: amendment of the law. See the page
- FOPH. Use of cannabis for medical purposes. See the page
- FOPH. Cannabis, addictions and health. See the page
- Swissmedic, FSVO, FOPH. Products containing cannabidiol (CBD), guidance document. Read the guidance document
- Hadorn S, Mavrot C, Novet B. Analysis of the results of the cannabis pilot trials in Swiss cities, part two (mid-2024 to mid-2025), executive summary, 17 November 2025, for the FOPH. Read the summary
- Baltes-Flueckiger L, et al. Effects of cannabis regulation in Switzerland: study protocol of a randomized controlled trial. Frontiers in Psychiatry, 2023. Read the study
- Baltes-Flueckiger L, et al. Effects of legal access versus illegal market cannabis on use and mental health: a randomized controlled trial. Addiction, 2025;120(10):1982-1992. Read the study
- Basel-Stadt Department of Health. Cannabisstudie Weed Care: Positive Bilanz nach zwei Jahren, press release of 28 March 2025 (in German). Read the press release
- Keystone-SDA. Basel-Stadt takes positive stock after three years of cannabis dispensing, 19 March 2026 (in English). Read the dispatch
- Hanf Magazin. Weed Care Basel: one year of pharmacy cannabis without psychological deterioration, 23 September 2026, report on an analysis published in Drug and Alcohol Dependence (2026). Read the report
- Tomaschek R, et al. SCRIPT, interim report 2026, University of Bern, 4 September 2026 (not peer-reviewed). Read the report
- Vaud Council of State. Report to the Grand Council on the pilot trials of controlled cannabis dispensing, May 2026. Read the report
- Cina N, et al. Regulated cannabis distribution pilot studies in Switzerland: an example in Geneva. Revue médicale suisse, 2026;22(965):1022-1026. Read the reference
- FOPH. New law on cannabis products (preliminary draft and report on the consultation results). See the page
- GREA. Cannabis: Parliament continues the work, 12 May 2026. Read the article
- RTS. Cannabis in Switzerland: the debate on legalisation divides parties and organisations. Read the article
- watson. Cannabis legalisation: Swiss parties disagree. Read the article
This article is for information purposes. It constitutes neither legal nor medical advice. It reflects the state of the sources consulted as of 20 October 2026; the regulatory situation and the results of the pilot trials are evolving. CBD is not a medicine.
Leave a comment