Pain Severity Trajectories and Patient Profiles in Chronic Pain Patients Treated with Cannabis-Based Medicinal Products: A k-Means Clustering Analysis of the UK Medical Cannabis Registry.
Erridge S, Datta A, et al. • Journal of clinical pharmacology • 2026
Five distinct pain-severity trajectory clusters emerged among chronic pain patients treated with cannabis-based medicinal products, with baseline sleep quality, anxiety, and treatment indication more strongly associated with response than product- or dose-related factors.
Key Findings
Results
All patient-reported outcomes improved significantly from baseline to 24 months across the full cohort of chronic pain patients treated with CBMPs.
3744 patients were included in the cohort study from the UK Medical Cannabis Registry
BPI-SF pain severity improved with an effect size of d = 0.57 (P < .001)
BPI-SF pain interference improved with an effect size of d = 0.59 (P < .001)
EQ-5D-5L index value improved with an effect size of d = -0.76 (P < .001)
Assessments were conducted at baseline and up to 24 months using patient-reported outcome measures including BPI-SF, EQ-5D-5L, GAD-7, and Single-Item Sleep Quality Scale
Results
More than half of patients achieved a minimal clinically important difference in each pain measure at 24 months.
This finding applied to both BPI-SF pain severity and pain interference measures
The study assessed minimal clinically important difference thresholds across the 24-month follow-up period
The cohort consisted of patients whose primary indication was chronic pain
Results
Longitudinal k-means clustering identified five distinct pain-severity trajectory clusters among CBMP-treated chronic pain patients.
The number of clusters was chosen using the gap statistic
All five clusters ended below their baseline pain severity, but 'the magnitude and pattern of change varied markedly'
Trajectories were based on BPI-SF pain severity scores over up to 24 months
The clustering method used was longitudinal k-means clustering applied to pain-severity trajectories
Results
Baseline sleep quality, anxiety, and treatment indication were more strongly associated with cluster membership (response pattern) than product- or dose-related factors.
Predictors of cluster membership were examined by logistic regression
Sleep quality was assessed using the Single-Item Sleep Quality Scale at baseline
Anxiety was assessed using the Generalised Anxiety Disorder-7 (GAD-7) at baseline
Product-related and dose-related factors were comparatively less predictive of trajectory group membership
Results
Adverse events were reported by 14.37% of patients and were predominantly mild or moderate in severity.
538 patients out of 3744 reported adverse events (14.37%)
The adverse events were characterized as predominantly mild or moderate
Adverse event data were collected over the 24-month observation period
What This Means
This research suggests that when people with chronic pain are treated with cannabis-based medicinal products (CBMPs) in the UK, their pain levels and overall quality of life tend to improve over a 24-month period. The study followed nearly 3,750 patients registered in the UK Medical Cannabis Registry and found that, on average, pain severity, pain interference with daily life, and general health-related quality of life all showed meaningful improvements. More than half of patients reached what researchers consider a 'minimal clinically important difference' — meaning the improvement was large enough to be noticeable and meaningful to the patient.
Importantly, the study found that patients do not all respond in the same way. Using a statistical technique called k-means clustering, researchers identified five distinct patterns of how pain changed over time. While all groups ended up with less pain than when they started, the speed and degree of improvement differed considerably between groups. Notably, patients' baseline sleep quality, anxiety levels, and the specific type of chronic pain condition they had were better predictors of which response pattern they would follow than factors like which cannabis product they used or what dose they received.
This research matters because it moves beyond asking 'does cannabis work for chronic pain on average?' and instead tries to understand 'who responds in what way and why?' The finding that sleep and anxiety at the start of treatment are linked to different response patterns suggests that these factors may be important to consider when identifying which patients are most likely to benefit from CBMPs. The relatively low rate of adverse events (about 14%), most of which were mild or moderate, also provides information about the tolerability of these treatments in a real-world clinical setting.
Check Your Own Numbers
Upload your bloodwork. We'll cross-reference your results against this study and 4,700 others.
Erridge S, Datta A, Mukherjee A, Vasooja D, Coomber R, Beri S, et al.. (2026). Pain Severity Trajectories and Patient Profiles in Chronic Pain Patients Treated with Cannabis-Based Medicinal Products: A k-Means Clustering Analysis of the UK Medical Cannabis Registry.. Journal of clinical pharmacology. https://doi.org/10.1002/jcph.70313