Large Canadian Study Documents How Older Adults Use Medical Cannabis, But Cannot Prove It Works

Nearly 10,000 seniors reported improved pain, sleep, and mood after using CBD-rich cannabis oils, but the absence of a control group means perceived benefits cannot be attributed to cannabis itself.

Why This Matters

Older adults are the fastest-growing demographic of cannabis users in Canada and in many other countries where medical cannabis is legal, yet age-specific clinical evidence remains remarkably sparse. Pain, insomnia, and polypharmacy are pervasive in this population, and patients frequently ask clinicians whether cannabis might help. This study represents the largest dataset reported for seniors using medical cannabis, making its strengths and its considerable limitations directly relevant to clinicians fielding those questions today.

Clinical Summary

Chronic pain, sleep disruption, and mood disorders drive a growing number of older adults toward medical cannabis, yet controlled trials in this age group are nearly nonexistent. This retrospective descriptive study, published by researchers using data from the Canabo Medical Clinic network across Canada, examined 9,766 adults aged 65 and older who consulted between October 2014 and October 2020. The biological rationale for cannabis in pain management centers on endocannabinoid system modulation, with cannabidiol in particular proposed to influence inflammatory and nociceptive signaling without the psychoactive burden of THC, a consideration especially important in older adults vulnerable to cognitive side effects.

At approximately 90-day follow-up, 72.7% of respondents reported perceived improvement in pain, 64.5% in sleep, and 52.8% in mood. CBD-dominant cannabis oils were by far the most commonly used product (83.6% of oil users). Adverse effects were relatively infrequent, with dry mouth (12.8%), drowsiness (8.6%), and dizziness (4.0%) leading the list. Notably, 35.6% of respondents reported reducing opioid doses and 19.9% reported reducing benzodiazepine doses. However, almost half of the intake cohort did not complete follow-up, outcomes were measured by unvalidated self-report, and the complete absence of a control group means these perceived improvements cannot be distinguished from placebo response, regression to the mean, natural disease course, or survivorship bias. The authors themselves frame the work as descriptive and call explicitly for controlled research.

Dr. Caplan’s Take

What this study gets right is scale and honesty. Nearly 10,000 patients is a meaningful dataset for describing who uses medical cannabis and what products they prefer, and the authors do not overstate their findings. The gap, though, is the one that always haunts observational cannabis research: we still cannot tell patients that cannabis works for their pain based on data like these. When a 72-year-old with chronic low back pain asks me whether cannabis oil could replace her opioid, I have to explain that the thousands of seniors who reported feeling better in this study might have felt just as much better taking an inert oil, and we genuinely do not know.

In practice, I treat these data as permission to take the conversation seriously rather than as permission to prescribe confidently. For older patients already interested in trying CBD-dominant products, I discuss the favorable short-term side-effect profile suggested here, ensure there are no critical drug interactions with their existing medications, start at the lowest available dose, and schedule structured follow-up. I do not frame cannabis as a proven therapy, and I never encourage patients to reduce opioids or benzodiazepines based on descriptive data alone.

Clinical Perspective

This study sits squarely at the hypothesis-generating stage of the research arc. It confirms what smaller surveys have suggested: that older adults gravitate toward CBD-dominant oils, that pain is the primary driver, and that most who continue treatment report subjective improvement. What it does not and cannot do is confirm that cannabis caused those improvements. The roughly 52% attrition at follow-up is not merely a statistical inconvenience; it likely inflates apparent benefit, because patients who experienced no improvement or who had adverse effects were disproportionately unlikely to return. Clinicians should treat the self-reported symptom improvement figures as ceiling estimates at best.

From a pharmacological standpoint, CBD-dominant formulations carry meaningful drug interaction potential via CYP3A4 and CYP2C19 inhibition, which is especially relevant in older adults on warfarin, certain statins, or anticonvulsants. The adverse-effect rates reported here (dry mouth, drowsiness, dizziness) are plausible but almost certainly underestimated given the self-report methodology and loss to follow-up. The one actionable recommendation for clinicians right now is this: when an older patient discloses cannabis use or asks about it, conduct a thorough medication interaction review, document the conversation, and schedule a structured reassessment rather than deferring to the patient’s next routine visit.

Study at a Glance

Study Type

Retrospective descriptive study (two-timepoint, uncontrolled)

Population

Adults aged 65 and older attending Canabo Medical Clinic network, Canada

Intervention

Medical cannabis (predominantly CBD-dominant oils), clinician-authorized

Comparator

None

Primary Outcomes

Self-reported perceived changes in pain, sleep, and mood; adverse effects; concomitant medication dose changes

Sample Size

9,766 at intake; 4,673 at follow-up (~47.8% retention)

Follow-up Interval

Mean 90.6 days (SD 58 days)

Data Period

October 2014 to October 2020

Journal

Not specified in source data

Funding Source

Data sourced from Canabo Medical Clinic (commercial cannabis provider)

Ethics Approval

Sunnybrook Research Ethics Board

What Kind of Evidence Is This

This is a retrospective, uncontrolled descriptive study drawing on routinely collected self-report questionnaire data from a commercial medical cannabis clinic. It occupies a low tier in the evidence hierarchy, comparable to a case series with structured intake and follow-up. The single most important inference constraint is the absence of any control or comparator group, which means no perceived benefit documented here can be causally attributed to cannabis rather than to placebo response, expectation effects, natural symptom fluctuation, or selective follow-up among satisfied users.

How This Fits With the Broader Literature

This study extends a small body of observational research on cannabis use in older adults, including survey work by Minerbi and colleagues and the Israeli longitudinal cohort study by Abuhasira and colleagues, both of which similarly reported high rates of self-perceived benefit and modest adverse-effect profiles among seniors. The current dataset is substantially larger and provides more granular product-preference data. However, it shares the fundamental limitation of all prior work in this space: no randomized controlled trial of medical cannabis for chronic pain has been conducted specifically in adults over 65. The consistent pattern across descriptive studies, that older adults tend to prefer CBD-dominant products, tolerate them reasonably well in the short term, and report subjective benefit, is a coherent signal worth testing rigorously, but it remains an untested hypothesis rather than established evidence of efficacy.

Common Misreadings

The most likely overinterpretation is treating the high proportions of self-reported improvement (72.7% for pain, 64.5% for sleep) as evidence that cannabis oil is effective for these conditions in older adults. The large sample size makes these numbers feel authoritative, but sample size does not compensate for the absence of a comparator. Without knowing how many people in a similar population would report improvement over the same period without cannabis, the figures are uninterpretable as efficacy measures. Similarly, the reported reductions in opioid and benzodiazepine use, while clinically intriguing, were self-reported and unverified, and cannot be separated from broader prescribing trends during the study period.

Bottom Line

This study reliably characterizes how nearly 10,000 Canadian seniors use medical cannabis: predominantly CBD-dominant oils, primarily for pain. It documents that most who return for follow-up perceive benefit and report few short-term adverse effects. It does not and cannot establish that cannabis caused those improvements. These data are best understood as descriptive epidemiology that justifies, but does not substitute for, the controlled trials this population urgently needs.

References

1. Abuhasira R, Schleider LB, Mechoulam R, Novack V. Epidemiological characteristics, safety and efficacy of medical cannabis in the elderly. Eur J Intern Med. 2018;49:44-50.
2. Minerbi A, Hauser W, Fitzcharles MA. Medical cannabis for older patients. Drugs Aging. 2019;36(1):39-51.
3. Canabo Medical Clinic retrospective descriptive study of older adult medical cannabis users, Canada, 2014-2020. Ethics approval: Sunnybrook Research Ethics Board.