Since 2019, Curaleaf International has invested in building rigorous, peer-reviewed evidence on patient outcomes in everyday clinical care. Research from the UK Medical Cannabis Registry, run by Curaleaf Clinic, has now been published in more than 50 peer-reviewed papers. The papers draw on data from over 55,000 patients. Here is what that body of real-world evidence covers, what it can and cannot tell us, and why it matters beyond the journals.
A milestone built by patients
The UK Medical Cannabis Registry was set up in 2019, shortly after specialist doctors in the UK were first allowed to prescribe cannabis-based medicines. It is run by Curaleaf Clinic, part of Curaleaf International. Today it holds data from more than 55,000 patients who have chosen to share how their treatment is going, many of them over years.
That makes it the largest real-world dataset on medical cannabis in Europe. The registry has now contributed to more than 50 peer-reviewed publications.
“Reaching 50 peer-reviewed papers is a milestone that belongs to our patients. Every one of these studies exists because people chose to share how their treatment was going, often over months and years. Real-world evidence can’t replace clinical trials, but it can show researchers where to look next, and that is what this data is now doing,” said Dr Simon Erridge, Research Director, Curaleaf Clinic.
Why real-world evidence matters
Randomised controlled trials remain the gold standard for testing any medicine. Most trials of cannabis-based medicines so far have focused on chronic pain. Taken together, systematic reviews point to a trend towards a small benefit, alongside side effects that need to be weighed. However, many of the underlying trials were judged to be at high risk of bias, so firm conclusions are hard to reach.
Trials also tend to exclude the people most often prescribed cannabis-based medicines: those with several conditions, or taking other medicines. As well as this, many trials follow people for weeks rather than years.
This is where a registry adds value. Patients complete validated questionnaires before and during treatment, covering pain, sleep, mood and quality of life. Clinicians record every product prescribed and every side effect reported. Over time, that builds a picture of everyday clinical practice that trials alone cannot provide.
What the research covers
As of November 2026, the registry has 51 publications: 50 published and one accepted for publication. They span more than 20 conditions, many of which have had little or no trial research. Output has held at around ten papers a year since 2023. Five of the 2026 papers report two-year outcomes, in depression, cancer pain, endometriosis, insomnia and PTSD.
| Area | Papers | Conditions studied |
| Pain | 21 | Chronic pain, fibromyalgia, cancer pain, migraine and headache, osteoarthritis, inflammatory arthritis, complex regional pain syndrome, endometriosis, hypermobility |
| Mental health | 14 | Generalised anxiety, depression, PTSD, ADHD, autism, substance use disorder |
| Whole-cohort and safety | 6 | Overall outcomes, safety and patient profiles across all registry patients |
| Neurology | 4 | Multiple sclerosis, epilepsy (including in children) |
| Sleep | 3 | Insomnia |
| Gastrointestinal | 2 | Inflammatory bowel disease |
| Palliative care | 1 | Palliative care |
One of the registry’s largest analyses, published in The Journal of Clinical Pharmacology in 2026, followed 8,945 patients for up to two years. Patients’ average quality-of-life scores rose from the start of treatment and stayed higher across the two years. By 24 months, 54% reported a meaningful improvement in anxiety and 44% in sleep quality. Around 14% reported side effects, most often fatigue, dry mouth and headache, and most of these were mild or moderate.
What the data can and cannot tell us
The registry is observational. There is no placebo group, so some changes may reflect expectation, the natural ups and downs of symptoms or other changes in someone’s life. People who stop treatment are also less likely to keep completing questionnaires, which can make average results look better than they are.
The registry’s authors set out these limits in every paper. They describe their findings as hypothesis-generating and supportive of clinical care, not as proof that a treatment works. That is precisely what makes the dataset useful: it points to which conditions to prioritise, which patients to recruit, which doses to test and which side effects to monitor in future trials.
A European perspective
The evidence question is not unique to the UK. Patients across Europe are prescribed cannabis-based medicines while the trial evidence is still developing, and the research challenges are shared: limited funds to trial plant-based products, the regulatory burden of working with a controlled drug, and stigma.
There are signs of progress. In 2025, researchers in Germany and Austria published two large randomised trials of a standardised full-spectrum cannabis extract in chronic low back pain, involving 820 and 384 adults. Results apply to that product in that condition, but they show that rigorous, long-term trials can be done.
What comes next
Medical cannabis is not one medicine. Products differ in their active ingredients, their strength and how they are taken. The next phase of research should test specific, well-characterised products in the conditions where the real-world signal is most consistent. Larger datasets and longer follow-up will help researchers ask sharper questions about who responds, at what dose and to which formulation.
“The first 50 papers have shown us where the questions are. The next step is independent, well-designed trials that test specific products in the conditions where the real-world signal is most consistent. Registry data can help design those trials, and then show how well their results hold up in everyday care,” said Dr Simon Erridge.
Full list of registry publications at ukmedicalcannabisregistry.com
This article is not medical advice. Anyone considering treatment should speak to their doctor.