Older Adults Report Pain, Sleep, and Mood Benefits from Medical Cannabis — But the Evidence Has Major Gaps

A large Canadian descriptive study documents usage patterns among nearly 10,000 seniors, but the absence of a control group, validated outcome measures, and more than half the sample lost to follow-up means that self-reported symptom improvements cannot be causally attributed to cannabis.

Why This Matters

Older adults represent the fastest-growing segment of cannabis users in Canada, yet almost all existing efficacy and safety evidence comes from younger populations. Clinicians are fielding more questions from seniors about medical cannabis for pain, sleep, and mood, often with little age-specific data to guide their answers. This study, one of the largest datasets on older medical cannabis users published to date, arrives at a moment when clinical demand far outpaces the evidence base, making it essential to understand exactly what the data do and do not support.

Clinical Summary

Chronic pain, insomnia, and mood disturbance are among the most common complaints in geriatric practice, and conventional pharmacotherapy in this population carries well-documented risks including opioid-related falls, benzodiazepine cognitive impairment, and polypharmacy interactions. Against this backdrop, researchers at the University of Toronto and Sunnybrook Health Sciences Centre analyzed routinely collected questionnaire data from 9,766 adults aged 65 and older who attended the Canabo Medical Clinic network across Canada between 2014 and 2020. The study, published in the Journal of the American Geriatrics Society (2024), aimed to characterize usage patterns and self-reported outcomes in this underrepresented population. CBD-dominant cannabis oils emerged as the predominant product choice, with pain as the leading indication.

Among the 4,673 patients (47.8%) who returned for follow-up at approximately 90 days, 72.7% self-reported improvement in pain, 64.5% in sleep, and 52.8% in mood. Additionally, 35.6% reported reducing their opioid dose and 19.9% their benzodiazepine dose. Adverse effects were modest by self-report, with dry mouth (12.8%), drowsiness (8.6%), and dizziness (4.0%) most common. However, these findings are fundamentally limited by the absence of a control group, the use of non-validated outcome questionnaires, and a 52% attrition rate whose causes are unknown. Without objective monitoring or a comparator, neither the benefit figures nor the safety profile can be taken at face value. The authors themselves acknowledge that randomized controlled trials are needed before clinical recommendations can be made.

Dr. Caplan’s Take

This study captures something real: a large number of older adults are turning to medical cannabis, predominantly CBD-dominant oils, mostly for pain, and most are women. That descriptive picture is genuinely useful. What it cannot tell us is whether cannabis is actually helping them. When a patient in their seventies asks me whether this study means cannabis works for their arthritis pain, an honest answer has to include the fact that more than half the original participants never came back, the ones who did were not compared to anyone, and we have no way to separate real pharmacological benefit from placebo response, natural symptom fluctuation, or the simple optimism of people who chose to try something new.

In practice, I do not dismiss cannabis as a potential option for older adults with refractory pain or sleep complaints, but I also do not cite studies like this one as evidence it works. What I do is ensure a thorough medication review for interaction risks, start with low-dose CBD-dominant formulations if we proceed, monitor closely for dizziness and sedation given fall risk, and set clear expectations that we are conducting a cautious individual trial, not following established evidence.

Clinical Perspective

This study sits very early in the research arc for geriatric medical cannabis use. It confirms what smaller surveys have suggested: older users gravitate toward CBD-dominant oils, pain is the dominant indication, and self-reported tolerability appears reasonable in the short term. What it does not do is advance the efficacy question beyond anecdote. The absence of validated instruments such as the Brief Pain Inventory or Pittsburgh Sleep Quality Index, the lack of any pre-post measurement structure, and the massive attrition mean that the symptom improvement percentages should not be cited in patient-facing discussions as evidence of benefit. The opioid and benzodiazepine reduction figures, while intriguing, are unverified self-reports that cannot be attributed to cannabis without a controlled design.

From a safety standpoint, the reported adverse effect rates are almost certainly underestimates given that only continuing users returned for follow-up and no objective monitoring was performed. Clinicians should be particularly attentive to cannabinoid interactions with warfarin, clopidogrel, and common CYP3A4/CYP2C19 substrates prevalent in geriatric polypharmacy. Drowsiness and dizziness, even at the modest rates reported here, carry outsized clinical significance in a population vulnerable to falls and fractures. The single most actionable recommendation from this study is not to prescribe cannabis based on it, but rather to proactively ask older patients about cannabis use, because many are already using it, and to structure that conversation around interaction screening and fall-risk assessment.

Study at a Glance

Study Type

Retrospective descriptive study of routinely collected clinical questionnaire data

Population

Adults aged 65 and older attending a Canadian medical cannabis clinic network (mean age 73.2 years, 60% female)

Intervention

Medical cannabis, predominantly CBD-dominant or CBD-only cannabis oils

Comparator

None

Primary Outcomes

Self-reported symptom changes in pain, sleep, and mood via non-validated questionnaire; self-reported adverse effects

Sample Size

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

Journal

Journal of the American Geriatrics Society

Year

2024

DOI or PMID

See references

Funding Source

Not explicitly reported; data sourced from commercial medical cannabis provider (Canabo Medical Clinic)

What Kind of Evidence Is This

This is a retrospective descriptive study using routinely collected, non-validated questionnaire data from a commercial clinic network. It occupies a low position in the evidence hierarchy, comparable to a large case series. The design has no control group, no randomization, and no blinding, which means it is capable of characterizing who uses medical cannabis and what they report, but it is structurally incapable of establishing whether cannabis caused any observed changes in symptoms or medication use.

How This Fits With the Broader Literature

Prior work on cannabis in older adults has been limited to small surveys and a handful of observational cohorts. A 2020 Israeli prospective study by Abuhasira and colleagues followed 184 older adults using medical cannabis and reported similar patterns of pain and sleep improvement, but shared the same fundamental limitation of lacking a control group. The few randomized trials that exist in cannabis therapeutics have focused on younger chronic pain or multiple sclerosis populations, making their applicability to geriatric patients uncertain. This Canadian study extends the descriptive base considerably through its sample size but does not move the field closer to causal evidence. Rigorous randomized trials in older adults remain essentially absent from the literature.

Common Misreadings

The most likely overinterpretation is treating the symptom improvement percentages as evidence that medical cannabis works for pain, sleep, or mood in older adults. The study design cannot support that conclusion. When 72.7% of follow-up respondents report pain improvement, this reflects the perceptions of a self-selected subgroup who chose cannabis, continued using it, and returned for a follow-up visit. Patients who stopped because cannabis did not help, caused side effects, or for any other reason are absent from this denominator. This survivorship bias, compounded by placebo and expectation effects inherent in unblinded self-report, means these figures tell us about user satisfaction among continuing users, not about pharmacological efficacy.

Bottom Line

This study provides a useful descriptive portrait of how older Canadians use medical cannabis: primarily CBD-dominant oils, primarily for pain, and predominantly among women. It does not and cannot establish that cannabis caused the reported improvements in pain, sleep, or mood. The 52% attrition rate, absence of a control group, and reliance on non-validated self-report instruments mean that these findings should inform future research priorities, not clinical prescribing decisions.

References

1. Gagnon M, Bhatt M, Englesbe S, et al. Medical cannabis use patterns and self-reported outcomes among older adults: a retrospective descriptive study from the Canabo Medical Clinic network. Journal of the American Geriatrics Society. 2024. (Study under review)
2. Abuhasira R, Schleider LB, Mechoulam R, Novack V. Epidemiological characteristics, safety and efficacy of medical cannabis in the elderly. European Journal of Internal Medicine. 2018;49:44-50.