Medical cannabis collection · Sources reviewed 11 October 2026
Age alone does not determine whether a medicine will help, but a treatment’s effect on alertness, balance and daily independence can become particularly important in later life. The evidence on cannabis in older adults needs to be read alongside the person’s condition, other medicines, priorities and level of frailty. A reassuring average from a trial may not describe every patient.
On this page
Who was actually studied?
A 2024 meta-analysis included 58 randomised trials involving 6,611 participants. Eligible trials had a mean participant age of at least 50, with study means ranging from 50 to 87. That is a broad population: a study whose average age is 52 is not equivalent to a study of people in their late eighties with multiple illnesses. Velayudhan and colleagues, 2024.
THC-containing medicines were associated with more adverse effects, including dizziness, sleepiness, balance or coordination difficulties and changes in perception or thinking. Several effects increased with THC exposure. The pooled analysis did not find statistically significant increases in serious adverse events or deaths, but this does not establish that rare or longer-term harms are absent.
“relatively short duration of treatment in included RCTs”
Velayudhan and colleagues, 2024The review identifies short treatment duration as a limitation. RCTs are randomised controlled trials.
Why ordinary daily activities matter
A symptom score is only one part of the outcome. Being less uncomfortable may be valuable, but so are getting to the bathroom safely, remaining able to read or converse, and carrying out a usual routine. A small change in steadiness can matter differently to someone who already needs assistance than to a trial participant with good mobility.
This is a reason to define the intended benefit before treatment is assessed. The same applies to alertness: more sleep is not automatically better restorative sleep, and less visible agitation is not automatically greater comfort. The person’s priorities should shape the interpretation.
Does it help poor appetite?
A 2025 triple-blind crossover trial tested a THC:CBD mouth spray in 17 adults aged at least 65 who had poor appetite. Food intake did not increase significantly: the estimated difference was 10 kilocalories, with a confidence interval extending from a decrease of 55 to an increase of 75. The study used two trial days separated by two weeks. The older-adult appetite trial.
This is useful negative evidence about the immediate food-intake outcome under those conditions. It is also a very small, short experiment. It cannot establish longer-term effects on body weight, nutrition, independence or falls. Feeling hungrier, eating more and maintaining muscle are related but distinct outcomes.
What about dementia-related symptoms?
A 2026 review pooled nine randomised trials involving 334 people with dementia. It did not establish statistically significant improvements in agitation, broader neuropsychiatric symptoms or cognition. Results were heterogeneous, and estimates for several outcomes were imprecise. Sleepiness was more frequent with cannabinoids. Dementia systematic review, 2026.
The appropriate conclusion is that a dependable benefit has not been established. It would be too strong to claim that every preparation is certainly ineffective: small studies and wide uncertainty leave questions unresolved. It would also be too strong to describe reduced agitation as an established effect on the basis of selected favourable trials.
Any symptom change should be interpreted in the context of comfort, communication and function. A medicine that mainly sedates someone has not thereby demonstrated an improvement in the underlying cognitive disorder.
What short trials cannot tell us
A systematic review of longer-term harms in chronic pain found that much of the evidence came from non-randomised studies and was of very low certainty. Such studies can reveal concerns missed by brief trials, but they can also be affected by incomplete reporting and selective follow-up. Longer-term harms review, 2022.
Someone who stops treatment because of difficulty may disappear from the group of continuing users. Looking only at those who remain can make long-term tolerability appear better than it is. Conversely, an adverse event occurring during treatment is not automatically caused by it. Both attribution and follow-up need care.
Make the review concrete
Medical cannabis in later life is a set of condition-specific treatment questions. Evidence about pain should not be used as proof of benefit for appetite or dementia, and evidence from relatively healthy middle-aged participants should not be presented as a complete safety assessment for frail older adults.
References and further reading
- Velayudhan et al. (2024). Adverse events caused by cannabinoids in middle aged and older adults for all indications: a meta-analysis of incidence rate difference.
- Randomised triple-blind crossover trial of THC:CBD spray for poor appetite in older adults (2025).
- Systematic review and meta-analysis of randomised cannabinoid trials for behavioural and psychological symptoms of dementia (2026).
- Long-term and serious harms of medical cannabis and cannabinoids for chronic pain: systematic review of non-randomised studies (2022).
Educational information, not an individual prescribing plan. Decisions about medicines belong with the treating clinician. External reference links open in a new tab.