Since January 2023, several hundred adults in Basel have been buying cannabis at a pharmacy counter — not on prescription, but as participants in a government-authorised research trial. The question hanging over the experiment from day one is the same one that shadows every legalisation debate: when access gets easier and legal, do people end up worse off? A new Swiss cannabis study on THC and quality of life now adds a sharper answer than “yes” or “no”.
The analysis was published online in mid-October 2026 in the journal European Addiction Research (read the study on PubMed). A team led by first author Jacqueline Curiger, which includes study head Marc Walter and Lavinia Baltes-Flückiger, followed 300 participants for 18 months. On average, their quality of life did not move. But participants with higher cumulative THC exposure reported somewhat lower quality of life at the end, while the total amount of cannabis they bought through pharmacies showed no measurable link. In other words, how much drug went in mattered more than how many grams were on the receipt.
Key Takeaways
The 18-month analysis of Switzerland’s regulated pharmacy model found no overall decline in quality of life among 300 participants: the average score on the EQ-5D scale started at 0.92 and changed by 0.002 over the follow-up, a shift with a p-value of 0.418, meaning it is indistinguishable from no change. Three factors did independently predict somewhat lower quality of life at 18 months: higher cumulative THC exposure, higher problematic-use scores at the start, and an increase in self-reported consumption. Total cannabis quantity purchased through regulated channels did not predict outcomes, at a p-value of 0.134. The effects are small, the THC finding sits just under the conventional significance threshold at p = 0.048, and the study is observational, so it cannot show cause and effect. It fits with the earlier one-year analysis, which found small reductions in depression, anxiety and problematic-use scores alongside slightly more frequent use. The defensible reading across both is that regulated access did not make this group worse off, and that rules counting only grams may be measuring the wrong thing.
Swiss Cannabis Study on THC and Quality of Life: The 18-Month Numbers
Quality of life was measured with the EQ-5D, a short, widely used questionnaire. It asks about five areas — mobility, self-care, usual activities, pain or discomfort, and anxiety or depression — and turns the answers into a single index where 1 represents the best possible health state. A starting average of 0.92 is high. This group began in good shape, which leaves little room for improvement and is worth keeping in mind when reading the “stable” result.
What sets this analysis apart from most cannabis research is how consumption was tracked. Participants did not just describe their use: their pharmacy purchases were recorded, along with the THC content of the products they bought. On top of that, they reported how much they consumed and completed the CUDIT-R, a questionnaire that tallies warning signs of problematic use — closer to a checklist than a diagnosis.
The team then used regression analysis, a statistical method that estimates how strongly each factor is linked to an outcome while holding the others constant. All three significant coefficients are small: −0.003 for cumulative THC exposure, −0.001 for the baseline CUDIT-R score and −0.016 for an increase in self-reported consumption. The study name is anonymised in the abstract, but the author list largely overlaps with the Basel Weed Care team, and the sample size matches that cohort.
Why Grams Say Little About Dose
The logic is familiar from alcohol. A glass of beer and a glass of whisky can hold the same volume and very different amounts of alcohol. Cannabis works the same way: two grams of flower at 8 percent THC contain less than half the active compound of two grams at 20 percent. A rule that counts only weight treats them as identical.
Basel’s model was built differently from the start. According to the one-year analysis, products were capped at 20 percent THC and the monthly purchase limit was set at 10 grams of THC — 10 grams of active compound, not 10 grams of plant material. One third of products sold contained less than 13 percent THC, according to the canton.
The quantity finding needs careful handling. That purchased quantity did not predict outcomes does not mean quantity is irrelevant: self-reported consumption did predict lower quality of life. And pharmacy receipts do not capture everything. When the trial was extended in July 2025, the study team reported that around half of participants were still also buying on the black market despite legal access. The pharmacy record is a partial view of total use.
What the 18-Month Data Cannot Show
The study shows associations, not causes. The arrow could easily point the other way: people whose lives are going worse may reach for stronger products, in which case lower quality of life would drive higher THC exposure rather than result from it. By this phase there was no longer a comparison group without legal access.
The familiar constraints apply as well. Participants were volunteers who already used cannabis regularly. Quality of life was self-reported. The setting is one Swiss region with its own health system. And a p-value of 0.048 means the THC result only just clears the line researchers conventionally use to call a finding statistically significant. At the time of writing, only the abstract was available to us; details such as the unit used for THC exposure and the full list of control variables are not given there.
A First-of-Its-Kind Legal Cannabis Trial
Conducted by a consortium including the University of Basel and the University Psychiatric Clinics Basel, the Weed Care project recruited around 370 participants. Half were given legal access to cannabis through licensed pharmacies, while the rest continued sourcing cannabis illegally.
Researchers used regular self-reported questionnaires to monitor both consumption habits and psychological well-being. The findings from that first randomised phase — published in the journal Addiction — show a slight but measurable decline in problematic cannabis use among those with legal access. In drug policy terms, this is significant: consumption is deemed problematic when it causes health, social, or mental difficulties, even if dependency isn’t formally diagnosed. The study protocol was published in advance, which matters: the outcomes were specified before the data came in, not chosen afterwards to fit a story.
The benefit was clearest among people who used cannabis alongside other substances. “In this subgroup, problematic cannabis use experienced a significant drop,” said lead author Lavinia Baltes-Flückiger, deputy head of the study at Psychiatric Services Aargau, in the University of Basel’s announcement.
The core measurement tool is called CUDIT-R. Think of it as a risk tally rather than a diagnosis: it counts how many warning signs a person reports about their own cannabis use. Depression and anxiety were captured through separate standardised questionnaires. Everything rests on what participants said about themselves, not on clinical examination — a limitation worth holding onto while reading the rest.
Mental Health Impact: No Harm, Some Relief
Importantly, the study debunks a common concern: that legalisation might worsen cannabis-related mental health issues. After six months, both groups showed no significant difference in rates of depression, anxiety or other psychiatric symptoms.
By the two-year mark, however, researchers observed a general improvement in mental well-being among participants — most of whom had by then transitioned to legal cannabis. “Legal access eases the burden on consumers,” said Professor Marc Walter, head of the study.
The One-Year Numbers: Small Shifts, Hard Limits
An analysis published online in September 2026 in the journal Drug and Alcohol Dependence (read the study) extends past the randomised six-month phase and looks at a full year inside the pharmacy model, following 374 regular adult cannabis users. The operating conditions were tight: dispensing by trained pharmacy staff, THC content capped at 20 percent, and purchases limited to 10 grams of THC per month.
The result is mixed, which is exactly what makes it useful. Depression and anxiety scores declined slightly. So did problematic-use scores. All three shifts were statistically significant and all three were small. Use frequency climbed from 18.7 to 20.1 days per month. Quantity per use day barely moved. The proportion above the screening threshold for possible cannabis use disorder dropped from 32.7 to 28.6 percent, at a p-value of .126 — which means the change is not distinguishable from noise.
Using more often without using more, and scoring slightly better while doing it, is not a victory lap. It is a finding that inconveniences the simple version of both arguments. The 18-month analysis adds a third dimension to the same picture: beyond depression and anxiety scores, overall quality of life also held steady on average.
Why “No Deterioration” Is Not the Same as “Benefit”
This is the boundary that matters most, and the authors draw it themselves: the one-year phase was uncontrolled. There was no comparison group. People who spend a year being observed, filling in questionnaires and receiving offers of voluntary counselling may change for reasons that have nothing to do with where they bought their cannabis. Without a control arm, the effect of legal access cannot be separated from the effect of time, attention, or simple statistical regression toward the mean.
Stack the usual constraints on top of that. The sample was self-selected: adults who already used cannabis regularly and volunteered to take part. The outcomes are self-reported, not clinically assessed. The setting is a single Swiss city with its own health system and its own black market. And a fall from 32.7 to 28.6 percent looks like movement until the p-value is placed beside it.
The honest summary, then, is narrow. Across twelve months of regulated pharmacy access, mental health did not get worse in this group, and across eighteen months, neither did quality of life. That is a statement about a risk that failed to materialise, not evidence of therapeutic effect. Anyone reading it the other way is reading more than is there. The same caution applies to small-sample psychiatric research generally — a point we made recently about a four-person pilot comparing THC and psilocybin, where the design mattered far more than the headline.
What Basel’s Operational Data Adds
Alongside the peer-reviewed publications, the Basel authorities report programme figures. After two study years, the Canton of Basel-Stadt health department’s interim report recorded 87 kilograms of cannabis sold and roughly 900,000 Swiss francs diverted from the illegal market. One third of products contained less than 13 percent THC, below what the black market typically supplies. Thirty-five participants approached the study physician on their own initiative to discuss their consumption. Of the 378 people originally enrolled, 299 were still active after two years, and the trial was extended to January 2027.
The study team’s updated interim overview now spans three years and reports declines in depressive symptoms, anxiety symptoms and psychosis-like symptoms, alongside reduced alcohol and other psychoactive substance use. Across three study years, 113 kilograms of flower and hash were dispensed, plus around 21 decilitres of vapes, oil and spray. The non-smokable products introduced in autumn 2025 accounted for roughly 18 percent of monthly purchase volume between September 2025 and February 2026.
These are operational records, not controlled measurements. They show the model works and is being used. They do not demonstrate that it improves anyone’s mental health.
SCRIPT: A Second Swiss Dataset
Basel is not the only Swiss trial. The Federal Office of Public Health oversees several authorised pilot projects, operating inside a legal carve-out that lets prohibition and scientific exception coexist.
In September 2026, the SCRIPT project released interim results covering 1,177 participants in Bern, Biel and Lucerne. After six months, the report found no recognisable effect of pharmacy access on how often people used cannabis. What did shift was method: the share of access-group participants who only smoked fell from 78 to 58 percent, while the control group barely changed. The report is not yet peer-reviewed, and it also documents 51 serious adverse events among 45 people across both arms.
On use frequency, Basel and SCRIPT disagree — stable in the randomised six-month comparison, modestly higher in the uncontrolled one-year observation. Smoothing that over would be sloppy. Different designs answer different questions, and the disagreement is itself the information.
What the Potency Finding Means for U.S. Markets
American adult-use rules mostly work the way the Swiss data suggest is incomplete: they cap purchases by weight. Potency caps are the exception. A February 2025 comparison by Connecticut’s Office of Legislative Research found that, among the Northeast states it reviewed, only Connecticut and Vermont limited flower to 30 percent THC, and none set a potency limit for vapes. That 30 percent ceiling is well above Basel’s 20 percent cap, and no state ties its monthly limit to grams of THC the way Basel does.
The 18-month study does not show that potency caps would improve anyone’s quality of life — an observational cohort of 300 Swiss volunteers cannot settle that. What it does show is that a weight-based limit and a dose-based limit measure different things, and that in this cohort the dose-based measure was the one linked to how people were doing.
Towards Full Legalisation
The Weed Care project is one of several ongoing pilot schemes in Switzerland. Collectively, these experiments are shaping the country’s path toward full legalisation.
Earlier government assessments show participants increasingly favouring safer consumption methods and reporting less stigma — especially when obtaining cannabis through pharmacies. Those data may have helped sway lawmakers: the Social Security and Health Committee of Switzerland’s National Council approved a draft law to legalise the commercial sale and possession of cannabis by a 14–9 vote. Public consultation on the preliminary Cannabis Products Act ran from August 29 to December 1, 2025, and the draft is now being revised at federal level. It would keep cannabis classified as a narcotic while giving adults strictly regulated access, with sales overseen by the cantons.
If it passes, Switzerland could become one of the first countries in Europe to offer a fully regulated recreational market — built on trial data rather than ideology. The caution the Swiss researchers apply to their own findings is precisely what makes that foundation worth anything.
More Insights on Cannabis from The Cannex
This Swiss work sits alongside a growing body of international research on what regulated markets actually change. A U.S. study found that legal marijuana may reduce workers’ compensation claims and costs, pointing to knock-on effects far outside health statistics. Other evidence suggests cannabis may curb alcohol cravings in heavy drinkers, which is directly relevant to Basel’s polydrug findings. And on the product side, geneticist Anna Schwabe’s work on how THC content varies between strains explains why a 20 percent cap in a pharmacy means something the illegal market cannot promise — and why the new 18-month finding about THC exposure is about labels as much as limits.
Frequently Asked Questions
Not in this Swiss cohort. Over 18 months, the average EQ-5D quality-of-life score barely changed. Because there was no comparison group in this phase, the data support the statement that quality of life did not decline, not that legal access improved it.
Grams measure the weight of plant material, not the amount of active compound, so two purchases of the same weight can deliver very different THC doses. Many participants also bought outside the pharmacy, so the recorded purchases only partly reflect total use. Self-reported consumption, by contrast, was linked to lower quality of life.
That conclusion is not available from this data. The one-year analysis found small, statistically significant reductions in depression and anxiety scores, but it ran without a control group. What can be said is that mental health did not deteriorate in this group across twelve months.
No. Cannabis remains a federally controlled substance, and no federal pathway permits pharmacy sale of non-medical cannabis; ongoing rescheduling proceedings have not created one. State-licensed retail programmes are commercial markets, not public-health dispensing trials with THC caps, purchase limits and pharmacist counselling. Swiss pilots run under a specific federal exemption that has no American equivalent, so the results describe a system US consumers cannot currently access.
Disclaimer
This article is science journalism and is provided for information only. It does not constitute medical, pharmaceutical or legal advice, makes no health claim, and is not a recommendation to purchase or consume any cannabis product. The findings described come from a Swiss pilot trial operating under a national exemption that has no direct equivalent in other jurisdictions. Cannabis laws differ by country and, in the United States, by state as well as at federal level. Readers with health concerns should consult a qualified physician or pharmacist.
