Cannabis Medicines Cause Dose-Dependent Side Effects in Older Adults, Meta-Analysis Finds

Pooled data from 58 randomized controlled trials show that THC-containing cannabinoid medicines raise the risk of dizziness, dry mouth, and cognitive effects in a dose-related pattern among adults aged 50 and over, but serious adverse events and deaths were not significantly increased compared to placebo.

Why This Matters

Adults over 50 represent the fastest-growing segment of medical cannabis users, yet most safety data have historically been derived from younger populations. This age group is uniquely vulnerable to the neurological and balance-related side effects that cannabinoids are known to produce, because dizziness, somnolence, and coordination impairment translate directly into fall risk, a leading cause of morbidity and mortality in older adults. Understanding the dose-dependent nature of these side effects is essential for clinicians navigating the tension between patient demand for cannabinoid therapies and the obligation to minimize harm in a population already managing polypharmacy and frailty.

Clinical Summary

Cannabis-based medicines (CBMs) are increasingly prescribed to older adults for chronic pain, spasticity, chemotherapy-induced nausea, and other conditions, but their adverse event profile in this age group has lacked systematic quantification. A 2024 systematic review and meta-analysis by Velayudhan and colleagues, published in Drugs & Aging and registered on PROSPERO, pooled data from 58 placebo-controlled RCTs encompassing 6,611 participants (mean ages 50 to 87 years) and 1,655.84 person-years of cannabinoid exposure. The investigators used incidence rate differences rather than relative risk ratios, providing clinicians with absolute risk estimates that are more directly interpretable in clinical decision-making. The mechanistic basis for concern centers on THC’s agonism at CB1 receptors in the central nervous system, which produces dose-dependent neurological and cognitive effects, while CBD’s modulatory role at multiple receptor targets may attenuate or alter some of these effects.

The pooled analysis found that THC alone significantly increased all-cause adverse event incidence rate differences (IRD 18.83 per 100 person-years; 95% CI 1.47 to 55.79) and treatment-related adverse events (IRD 16.35; 95% CI 1.25 to 48.56) compared to placebo. THC:CBD combination products showed comparable increases (all-cause IRD 19.37; treatment-related IRD 11.36). THC dose-dependently increased dry mouth, dizziness, balance and coordination problems, dissociative and perceptual symptoms, and somnolence. Critically, serious adverse events, withdrawals due to adverse events, and deaths were not statistically significantly elevated above placebo, though the authors acknowledged that pooled samples were likely underpowered to detect rare events. Confidence intervals for several estimates were notably wide, reflecting substantial heterogeneity across formulations, doses, durations, and clinical indications. The authors concluded that while CBMs appear broadly tolerable in this population, the dose-dependent side effect profile warrants careful dose titration and monitoring, and that larger, longer trials specifically enrolling older adults are needed before definitive safety conclusions can be drawn.

Dr. Caplan’s Take

This meta-analysis does something genuinely useful: it quantifies the absolute additional risk of common side effects in the age group that most needs this information. The finding that THC dose-dependently increases dizziness, balance problems, and somnolence is not surprising to anyone who prescribes cannabinoids to older patients, but having pooled numbers to reference is valuable. What concerns me is the gap between the reassuring headline that serious adverse events were not significantly elevated and the reality that 58 trials with roughly 6,600 participants are simply not powered to detect rare but consequential outcomes like hip fractures following cannabinoid-related falls. The patients who ask me about cannabis medicines are often already on multiple medications, and the side effects identified here interact with existing fall risk in ways that a meta-analysis cannot fully capture.

In practice, I approach THC-containing therapies in older adults with deliberate caution. I start at the lowest available dose, titrate slowly, and counsel patients and caregivers explicitly about fall risk, especially during the first weeks. I find the dose-response data here reinforcing of a principle I already follow: in this population, the therapeutic window is narrow, and the cost of overshooting it is a broken hip rather than just an unpleasant afternoon. When CBD-predominant options can address the clinical goal, I lean toward those first.

Clinical Perspective

This meta-analysis sits at a meaningful inflection point in the cannabinoid safety literature. Prior reviews either combined all adult age groups or focused on efficacy endpoints with safety as a secondary consideration. By restricting inclusion to trials with mean participant age of 50 years or older and reporting absolute incidence rate differences, this study provides the most age-specific adverse event quantification to date for cannabinoid medicines. The dose-response findings for neurological side effects confirm what smaller studies have suggested and add statistical structure to clinical intuition. However, the evidence does not support blanket reassurance. The null finding for serious adverse events should not be interpreted as evidence of absence, given the wide confidence intervals and the inherent difficulty of detecting rare events in pooled datasets of this size. Clinicians should not cite this study as establishing the safety of cannabinoid medicines in older adults; rather, it establishes a tolerability profile with clearly identified dose-dependent risks.

From a pharmacological standpoint, the interaction between weekly THC and CBD doses on neurological, psychiatric, and cardiac side effects is clinically noteworthy, as it suggests CBD may modulate certain THC-related adverse events, though the direction and magnitude of this modulation are not yet predictable at the individual patient level. Clinicians prescribing THC-containing products to older adults should be particularly alert to additive sedation with benzodiazepines, opioids, and anticholinergic medications, all of which are common in this population. The single most actionable recommendation from this evidence is to adopt an explicit low-and-slow titration protocol for any THC-containing cannabinoid medicine in patients over 50, with structured reassessment of balance, cognition, and fall risk at each dose escalation.

Study at a Glance

Study Type

Systematic review and meta-analysis of placebo-controlled RCTs

Population

Adults aged 50 and over (mean age range 50 to 87 years; n=6,611)

Intervention

THC-containing cannabinoid-based medicines (THC alone or THC:CBD combination)

Comparator

Placebo

Primary Outcomes

Incidence rate differences (IRD) for all-cause and treatment-related adverse events, serious adverse events, withdrawals, and deaths

Sample Size

58 RCTs; 6,611 participants (3,450 receiving cannabinoid medicines); 1,655.84 person-years of exposure

Journal

Drugs & Aging

Year

2024

DOI or PMID

PROSPERO registration CRD42019148869

Funding Source

Not specified in available data

What Kind of Evidence Is This

This is a PROSPERO-registered systematic review and meta-analysis of placebo-controlled randomized controlled trials, following PRISMA reporting guidelines. It sits near the top of the evidence hierarchy for questions about intervention-related harms. The single most important inference constraint is the substantial heterogeneity introduced by pooling across diverse clinical indications, cannabinoid formulations, dosing regimens, and trial durations, which means the average effect estimates may not accurately reflect risk in any specific clinical scenario or for any particular product.

How This Fits With the Broader Literature

This analysis extends prior systematic reviews, notably the 2019 Cochrane review by Defined and colleagues on cannabinoids for chronic neuropathic pain and the 2018 review by Whiting and colleagues covering broader indications, both of which identified similar side effect patterns but did not stratify by age or report absolute incidence rate differences. The dose-response findings are consistent with pharmacokinetic studies showing that older adults achieve higher THC plasma concentrations at equivalent doses due to altered body composition and hepatic metabolism. What is genuinely novel here is the age-restricted pooling and the use of IRD rather than odds ratios, which provides more clinically intuitive risk communication.

The finding that CBD may modulate THC-related adverse events aligns with preclinical and early clinical evidence of CBD’s partial antagonism at CB1 receptors and its effects on THC metabolism via CYP enzyme competition, though the clinical significance of this interaction remains incompletely characterized in older populations specifically.

Common Misreadings

The most likely overinterpretation of this meta-analysis is to read the null finding for serious adverse events and deaths as positive evidence that cannabinoid medicines are “safe” for older adults. This exceeds what the data support. The confidence intervals for serious adverse event comparisons were wide, and the pooled sample, while large by cannabinoid trial standards, remains underpowered to detect the rare but consequential outcomes that matter most in geriatric medicine, such as fall-related fractures, hospitalizations from drug interactions, or cardiovascular events. Absence of statistical significance for rare harms in a pooled analysis of this size is not the same as demonstrated safety, and clinicians should not present it as such to patients.

Bottom Line

This meta-analysis provides the strongest available age-specific evidence that THC-containing cannabinoid medicines produce dose-dependent increases in dizziness, balance impairment, somnolence, and dry mouth in adults over 50. It does not establish that these medicines are safe in this population, as the analysis was underpowered for rare serious events. For clinical practice today, the evidence supports cautious, low-dose initiation with structured monitoring of fall risk and cognition at every dose adjustment.

References

1. Velayudhan L, et al. Adverse events from cannabinoid-based medicines in middle-aged and older adults: a systematic review and meta-analysis of randomised controlled trials. Drugs & Aging. 2024. PROSPERO registration: CRD42019148869.
2. Whiting PF, Wolff RF, Deshpande S, et al. Cannabinoids for medical use: a systematic review and meta-analysis. JAMA. 2015;313(24):2456-2473. doi:10.1001/jama.2015.6358
3. Mucke M, Phillips T, Radbruch L, Petzke F, Hauser W. Cannabis-based medicines for chronic neuropathic pain in adults. Cochrane Database Syst Rev. 2018;3(3):CD012182. doi:10.1002/14651858.CD012182.pub2