Cannabis and Dementia: What Research Shows About THC and the Aging Brain

Dementia slowly takes memory, and in its later stages it often brings agitation: pacing, shouting, restless nights and sometimes aggression. Families watching that happen want something that helps, and cannabis keeps coming up.

The research on cannabis and dementia pulls in two directions. Small, measured doses of THC have calmed agitation in several trials of people who already have dementia, while heavy use shows up as a risk in at least one large population study. Both stories are smaller and messier than the headlines.

Agitation trials with THC

Agitation affects an estimated 40% of people with Alzheimer's, according to Johns Hopkins researchers, and the best evidence on THC and dementia agitation comes from care settings. In a trial from Johns Hopkins and Tufts presented in 2024, 75 people with severe Alzheimer's agitation took 5 milligrams of dronabinol, a synthetic THC pill, or a placebo twice a day for three weeks. Agitation fell by an average of 30% on the drug, and the trial's co-lead called agitation one of the most distressing symptoms of Alzheimer's dementia. In an earlier crossover trial presented in 2018, 39 patients with an average age of 87 took about 1.6 milligrams of nabilone, another synthetic cannabinoid, a day. Agitation improved compared with placebo, but sedation was more common, at 45% against 16%.

The biggest result so far arrived in July 2026. In the LiBBY trial, 120 people with dementia who were eligible for hospice care took an oil with 2 milligrams of THC and 100 milligrams of CBD twice a day, or a placebo. Agitation scores fell 6.27 points more than on placebo by week 2 and 8.23 points more by week 12. Serious adverse events were more common in the treatment group, 23.3% against 11.9%, though investigators judged none of them related to the medication. These are topline results, not yet peer reviewed.

Not every trial worked. A 2015 study in Neurology gave 50 people with dementia 1.5 milligrams of THC three times a day for three weeks and found no significant difference from placebo, with similar side effects in both groups.

A February 2026 review of 10 trials with 328 participants concluded that cannabinoids are efficacious for alleviating dementia-related agitation, with sedation about twice as common as on placebo, though overall side effects did not differ. Its own numbers carry a warning. Once the studies at high risk of bias were removed, the effect on agitation was no longer statistically significant, and the benefit showed up mainly at doses above 10 milligrams of THC equivalent and in severe dementia.

CBD for dementia, usually with some THC

CBD for dementia has a thinner record, and the clearest positive result came from an oil that also carried some THC. In a 2022 trial in Israel, 60 patients with dementia and agitation took an oil with 30% CBD and 1% THC or a placebo for 16 weeks. Agitation scores dropped by at least 4 points in 60% of the oil group against 30% on placebo, side effects were similar in both groups, and the authors concluded that the oil significantly reduced agitation over placebo.

Other results are mixed:

• Vascular dementia, Brazil. In 30 people, 300 milligrams of CBD a day for four weeks reduced behavioral symptoms without harming memory or daily function.
• Severe dementia, 2026. In a crossover trial of 25 patients with an average age of 83, an oil with up to 20 milligrams of THC and 40 milligrams of CBD a day did no better than placebo on the main agitation scale, though it was well tolerated.

Heavy use and dementia risk

So does weed cause dementia? The largest study so far found a link with heavy use, though it cannot prove cause. Researchers in Ontario followed 6 million individuals aged 45 years or older with no history of dementia. People who needed emergency or hospital care because of cannabis were diagnosed with dementia more often over the next five years: 5.0% of them, against 3.6% of people with hospital care for any reason and 1.3% of the general population. After adjustment, their risk was 23% higher than the first comparison group and 72% higher than the general population.

It also covers only use severe enough to put someone in a hospital. A February 2026 analysis of UK Biobank and US veterans data, covering up to 18,975 lifetime cannabis users and 12,222 people with cannabis use disorder, found that cannabis use was not linked to cognitive decline or dementia risk.

Long-term heavy use from a young age is a separate concern. In the Dunedin study, which follows 1,037 people born in one New Zealand city, long-term cannabis users lost an average of 5.5 IQ points between childhood and midlife and had a smaller hippocampus, the brain region central to memory.

Lab work on cannabis and Alzheimer's

The hopeful side of cannabis and Alzheimer's research comes from mice and lab dishes. The endocannabinoid system is the body's own cannabinoid signaling network. In 2017, researchers in Germany reported in Nature Medicine that the activity of the ECS declines during aging, and that a low daily dose of THC reversed the age-related decline in learning and memory tests in mice aged 12 and 18 months.

A year earlier, Salk Institute scientists found preliminary evidence that THC can help nerve cells clear amyloid beta, the toxic protein that builds up in Alzheimer's disease, in neurons grown in the lab. None of this has carried over to people yet. The Alzheimer Society of Canada's position is that there is currently no evidence that cannabis is useful for the treatment or prevention of Alzheimer's disease, though it notes that cannabis can help with behavioral symptoms such as agitation and aggression in some people with dementia.

The unexamined day is a wasted opportunity. Reflect on what you did, what you learned, and how you can improve.

John Dewey

Milligrams matter more than strain names

Every positive THC trial above used small, measured doses, such as 5 milligrams of dronabinol, about 1.6 milligrams of nabilone or 2 milligrams of THC with CBD. Dried flower works on a different scale, because the percentage on a strain page turns into milligrams per gram:

• Afghan Hash Plant, an Afghan indica from regular seeds at 21% THC, one of the lowest figures in our catalog, holds about 210 milligrams of THC in a gram of dried flower.
• Bruce Banner, a Strawberry Diesel and OG Kush cross at 32% THC, holds about 320 milligrams in a gram.

Even the milder of the two carries about 100 times the THC in one dose of the LiBBY oil. A homemade edible made from either can easily hold a dose no dementia trial has tested.

Storage matters as much as dose. Someone with memory loss can mistake an edible for an ordinary sweet, and after edibles became legal in Ontario, emergency visits by older adults for cannabis poisoning tripled, according to a 2024 study in JAMA Internal Medicine. The researchers also noted that most edible products give no dosing advice for older adults. Label anything homemade, and lock it away.

Older brains, bigger side effects

Age changes how THC lands. A 2024 meta-analysis of 58 randomized trials in adults 50 and older found that THC raised the rate of dizziness, dry mouth, drowsiness and problems with balance and coordination, and higher doses raised it more. In California, cannabis-related emergency visits by people 65 and older went from 366 in 2005 to 12,167 in 2019, and University of California researchers warned that slower reactions and weaker attention can lead to injuries and falls.

Cannabis also interacts with common prescriptions. A 2024 systematic review found that interactions between cannabis/cannabinoids and warfarin, valproate, tacrolimus and sirolimus were the most widely reported. Warfarin is a widely used blood thinner, valproate treats seizures and mood disorders, and tacrolimus and sirolimus protect transplanted organs. Before an older relative tries cannabis, have a pharmacist check the full medication list for clashes between weed and prescription drugs like these.

Calmer days, unproven cures

THC pills and THC with CBD have reduced agitation in several dementia trials, with drowsiness as the price, while other trials found no clear effect. CBD-rich oils show mixed results. Hospital-level cannabis use was linked to more dementia diagnoses in Ontario, while a large UK and US analysis found no link, and the promising Alzheimer's work is still in mice and cells. The doses that helped were a few milligrams, far below what a gram of modern flower holds.

Talk to the doctor before the first dose. Barney's Farm has been developing premium cannabis genetics since the 1980s, with over 40 Cannabis Cup wins. Explore our full seed catalog and find strains bred for every climate and skill level.