Cannabis Medicines Appear Broadly Tolerated in Adults Over 50, But Key Uncertainties Remain
A meta-analysis of uncontrolled studies finds low adverse event rates for THC and CBD alone, with higher withdrawal rates for combination products, but the absence of a comparator group limits causal conclusions and leaves the true safety profile of cannabinoids in older adults unresolved.
Why This Matters
Cannabinoid use among adults over 50 is rising sharply, driven by expanding legalization, growing product availability, and patient interest in managing chronic pain, insomnia, and cancer-related symptoms. Yet the safety evidence base for this age group remains thin, with most controlled trials enrolling younger populations and running for short durations. This review represents one of the few systematic efforts to aggregate real-world and open-label safety data specifically in older adults, arriving at a moment when clinicians face daily questions about cannabinoid tolerability in patients who often carry polypharmacy burdens and age-related vulnerabilities.
Clinical Summary
Cannabinoid-based medications have gained clinical traction across a range of indications, from cancer-related symptoms to chronic pain and neurodegenerative conditions, yet controlled safety data in older adults remain scarce. Published in Drugs & Aging in 2021, this systematic review and meta-analysis by Velayudhan and colleagues pooled incidence rate data from 38 open-label trials and observational studies encompassing 2,341 adults with a mean age of approximately 63 years. The rationale for examining uncontrolled studies specifically was to complement a companion RCT-focused meta-analysis by the same group and to capture tolerability signals from real-world and naturalistic clinical settings where older patients are more likely to encounter these medications.
The pooled estimates showed that THC-only and CBD-only treatments produced low adverse event incidence rates and no pooled serious adverse events, suggesting a reasonable short-term tolerability profile under uncontrolled conditions. THC:CBD combination products, however, showed measurably higher rates of serious adverse events (incidence rate 21.32 per 1,000 person-years) and treatment-related withdrawals (incidence rate 34.31), with cancer populations appearing to drive these signals. Critically, substantial heterogeneity (I-squared exceeding 55%) was present across the majority of analyses, and the complete absence of control arms means it is impossible to distinguish drug-caused adverse events from those attributable to underlying disease. The authors explicitly acknowledge that these findings are descriptive and hypothesis-generating, and that controlled trials specifically designed for older adults are needed before clinical recommendations can be drawn.
Dr. Caplan’s Take
This review does something useful by systematically gathering safety data that would otherwise remain scattered across dozens of small, heterogeneous studies. The tolerability signal for THC-only and CBD-only preparations is broadly reassuring, and it aligns with what I see in clinical practice among patients over 50 who use these compounds at appropriate doses. But the absence of a control group is not a minor footnote. Many of these patients have cancer, chronic pain, or neurodegenerative disease, and those conditions alone produce the very symptoms that get coded as adverse events. Patients who ask whether cannabis medicines are “safe for someone my age” deserve an honest answer: probably tolerable for many, but we genuinely cannot separate drug effects from disease effects in this data.
In practice, I start low and titrate slowly in older adults, with particular caution around THC:CBD combination products in patients with advanced cancer or significant polypharmacy. I monitor for dizziness, sedation, and cognitive changes, and I make sure patients understand that the evidence for safety in their age group is suggestive rather than definitive. Until well-designed controlled trials enroll genuinely older populations, clinical judgment and careful monitoring remain our primary tools.
Clinical Perspective
This meta-analysis sits at an early stage of the research arc for cannabinoid safety in older adults. It confirms what smaller case series and clinical experience have suggested: that THC and CBD monotherapies appear broadly tolerable in adults over 50 in the short term, with low rates of treatment discontinuation. What it does not and cannot do, given its uncontrolled design, is establish whether these rates are meaningfully different from what would be observed in untreated patients with the same conditions. The elevated serious adverse event and withdrawal rates seen with THC:CBD combinations are hypothesis-generating, but the concentration of these signals in cancer populations makes it difficult to disentangle drug effects from disease burden. Clinicians should not interpret these pooled estimates as precise benchmarks for expected adverse event rates.
From a pharmacological standpoint, the polypharmacy exposure typical of adults over 50 raises specific concerns that this review does not address. CBD is a potent inhibitor of CYP3A4 and CYP2D6, with the potential to alter plasma levels of common medications including statins, antidepressants, and anticoagulants. THC adds risks of orthostatic hypotension and cognitive slowing, both of which carry outsized consequences in older patients prone to falls. The most actionable step clinicians can take now is to conduct a thorough medication interaction review before initiating any cannabinoid in a patient over 50, and to use the lowest effective dose with scheduled follow-up within the first two weeks.
Study at a Glance
- Study Type
- Systematic review and meta-analysis of open-label trials and observational studies
- Population
- Adults aged 50 years and older (mean age 63.19 years; 53.86% male) across cancer, neurodegenerative, chronic pain, and other conditions
- Intervention
- Cannabinoid-based medications categorized as THC-only, CBD-only, or THC:CBD combination products
- Comparator
- None (all included studies were single-arm and uncontrolled)
- Primary Outcomes
- Incidence rates of adverse events, serious adverse events, and treatment-related withdrawals per 1,000 person-years
- Sample Size
- 38 studies; 2,341 total participants
- Journal
- Drugs & Aging
- Year
- 2021
- DOI or PMID
- PROSPERO registration CRD42019148869
- Funding Source
- Not explicitly stated in the available analysis
What Kind of Evidence Is This
This is a pre-registered systematic review and meta-analysis conducted according to PRISMA guidelines, pooling data exclusively from uncontrolled open-label trials and observational studies. In the evidence hierarchy, it sits below meta-analyses of randomized controlled trials and below individual RCTs. The single most important inference constraint is that all included studies lacked a control arm, making it fundamentally impossible to attribute observed adverse event rates to the cannabinoid interventions rather than to the underlying diseases, concomitant treatments, or normal aging processes in the study populations.
How This Fits With the Broader Literature
The findings are broadly consistent with the companion RCT-based meta-analysis by the same author group (Velayudhan et al., 2021), which reported similarly low adverse event rates for cannabinoids in older adults under controlled conditions. Together, the two papers suggest that the tolerability profile observed in controlled settings is not dramatically different from what emerges in real-world use, though neither paper can establish a definitive safety threshold. The elevated signal for THC:CBD combinations in cancer populations echoes findings from the nabiximols (Sativex) literature, where combination cannabinoid products have consistently shown higher discontinuation rates than single-agent formulations. A notable gap remains in the literature for adults over 75 years of age, a population almost entirely absent from this and other cannabinoid safety reviews despite being the group at greatest risk for drug interactions and adverse outcomes.
Common Misreadings
The most likely overinterpretation of this review is to conclude that cannabinoid-based medications have been “proven safe” in older adults. The low pooled adverse event rates are descriptive estimates drawn from studies without control groups, and the wide confidence intervals and high heterogeneity across analyses mean these numbers should not be treated as reliable point estimates of risk. A pooled incidence rate of zero serious adverse events for THC-only or CBD-only formulations does not mean these compounds carry no serious risk; it means the included studies were too small, too short, or too heterogeneous to detect such events. The reassuring headline must be read alongside the fundamental limitation that we cannot know what the adverse event rate would have been without treatment.
Bottom Line
This systematic review provides a useful descriptive summary suggesting that THC-only and CBD-only medications appear broadly tolerable in adults over 50 in uncontrolled settings, while THC:CBD combinations warrant greater caution, particularly in cancer populations. It does not establish causation, cannot define a true safety profile, and should not change clinical practice in the absence of controlled data. For now, it supports continued careful, individualized use with vigilant monitoring rather than broad reassurance.
References
- Velayudhan L, McGoohan K, Sauer S. Safety and tolerability of natural and synthetic cannabinoids in adults aged over 50 years: a systematic review and meta-analysis. Drugs & Aging. 2021;38(12):1065-1080. PROSPERO registration: CRD42019148869.
- Velayudhan L, McGoohan K, Sauer S. Safety and tolerability of cannabinoid-based medicines in older adults: a systematic review and meta-analysis of randomised controlled trials. Drugs & Aging. 2021. (Companion RCT meta-analysis by the same author group.)
Have thoughts on this? Share it:
