30 seconds summary
- Research is promising but still limited. Small studies suggest cannabinoid-based treatments may reduce agitation in some people with severe dementia, but evidence for CBD alone, particularly for anxiety, remains uncertain.
- A recent review also found that benefits weakened when lower-quality studies were excluded.
- CBD may cause drowsiness and interact with other medicines, so it should only be used under medical supervision not as a cure or proven treatment for dementia.
Anxiety and agitation are among the most challenging symptoms experienced by people living with dementia. A person who was previously calm may become restless, frightened, verbally aggressive or physically resistant to care. They may pace continuously, call out repeatedly, refuse medication, strike caregivers or become distressed when completing ordinary activities such as bathing and dressing. These behaviours can be deeply upsetting for the individual, exhausting for family members and difficult for professional caregivers to manage.
Because conventional medications do not always work and may produce serious side effects, interest has grown in cannabidiol, commonly known as CBD. CBD is a substance found in the cannabis plant. Unlike tetrahydrocannabinol, or THC, CBD does not usually cause intoxication or the “high” associated with cannabis. It is often promoted as a natural treatment for anxiety, sleep problems, pain and behavioural disturbances.
The central question is whether CBD can genuinely reduce anxiety and agitation in people with dementia. Current research offers some promising findings, particularly for certain CBD-rich products and combinations of CBD and THC. However, the evidence remains limited. Many studies have been small, short and based on products containing both CBD and THC rather than CBD alone.
The most accurate conclusion is that CBD may help reduce agitation in some dementia patients, but it is not yet a proven or routine treatment. Evidence for treating anxiety specifically is even less certain. CBD should not be given to a person with dementia without medical supervision because it can cause sedation, dizziness, liver abnormalities and interactions with other medicines.
Understanding Agitation in Dementia
Agitation is not a single symptom. It is a group of behaviours associated with emotional distress, excessive movement, verbal tension or physical aggression. It may include pacing, wandering, restlessness, shouting, repetitive questioning, resisting care, pushing, hitting or destroying objects.
Anxiety may appear separately or contribute to agitation. A person with dementia may feel unsafe because they no longer recognise their surroundings, cannot understand what is happening or cannot communicate what they need. Memory loss can make each unfamiliar face, sound or activity seem like a new threat.
Behavioural changes are partly caused by deterioration in areas of the brain that regulate judgement, emotion, impulse control and communication. However, agitation should not automatically be regarded as an unavoidable result of dementia. It is frequently a sign that something else is wrong.
Pain, constipation, urinary retention, dehydration, hunger, infection, medication side effects, poor sleep, hearing loss and vision problems can all trigger behavioural disturbances. Environmental causes include noise, unfamiliar caregivers, overcrowding, changes in routine, uncomfortable temperatures and being asked to perform a task the person no longer understands. A sudden change in behaviour requires medical assessment because delirium, infection or another acute illness may be responsible.
Agitation can also be a form of communication. A person who can no longer say “I am in pain,” “I am frightened” or “I do not understand what you want” may express the same message by shouting, pacing or pushing someone away. Treatment must therefore begin by investigating what the behaviour might be communicating.
Why CBD Has Attracted Interest
The human body has an endocannabinoid system involved in mood, stress responses, appetite, sleep, pain perception and immune activity. Cannabinoids from the cannabis plant may influence this system.
THC acts directly on cannabinoid receptors and can cause relaxation, altered perception, impaired memory, euphoria, anxiety or psychotic symptoms, depending on the person and dose. CBD has a more complicated and indirect pattern of activity. Researchers have proposed that it may affect serotonin signalling, inflammatory processes and the body’s own cannabinoid chemicals.
Laboratory and animal research has produced theories that CBD might reduce anxiety, inflammation and abnormal stress responses. These mechanisms could be relevant to dementia-related agitation. Nevertheless, biological plausibility is not proof of clinical effectiveness. A substance can produce promising effects in laboratory experiments without becoming a safe or successful treatment for people.
CBD is often described as non-intoxicating, but “non-intoxicating” does not mean harmless. At sufficient doses it can cause sleepiness, dizziness, diarrhoea, reduced appetite and changes in liver enzymes. Its effects can also be altered by other medicines.
CBD Is Not the Same as Medical Cannabis
One of the most important issues in interpreting the evidence is the difference between pure CBD and cannabinoid medicines containing THC.
Many positive studies described in the media as “CBD research” have actually tested:
- CBD-rich cannabis oil containing a small quantity of THC;
- balanced CBD–THC products;
- synthetic THC medicines such as dronabinol or nabilone; or
- whole-plant preparations containing several cannabinoids.
These treatments are not interchangeable. A finding involving a product containing both CBD and THC does not prove that an over-the-counter bottle of CBD oil will produce the same result.
The proportions of CBD and THC, total dose, method of delivery and manufacturing quality can all influence the effect. Prescription-grade or research-grade preparations are produced according to defined standards. Consumer products may contain different amounts from those shown on the label, residual THC or contaminants.
Evidence from a CBD-Rich Oil Trial
One of the most frequently discussed studies was a 2022 randomised, double-blind, placebo-controlled trial conducted in Israel. It included 60 people with dementia and behavioural disturbances, with an average age of approximately 79 years.
Participants received either placebo or a broad-spectrum cannabis oil three times daily for 16 weeks. The active oil contained a high concentration of CBD but also contained THC—approximately 295 milligrams of CBD and 12.5 milligrams of THC per millilitre.
At 16 weeks, 60% of participants assigned to the cannabis oil achieved a reduction of at least four points on the Cohen-Mansfield Agitation Inventory, compared with 30% receiving placebo. A reduction of at least eight points occurred in 50% of the treatment group and 15% of the placebo group. Improvements were also reported in measures of sleep disturbance and aggression.
These results suggest that a CBD-rich, low-THC product may reduce dementia-related agitation. However, the trial had significant limitations. It involved only 60 participants at one research centre. Eight participants discontinued the study, all from the active-treatment group. The product was not pure CBD, so the relative contribution of CBD, THC or their interaction cannot be determined. The authors appropriately concluded that larger studies were required.
Evidence for CBD Without THC
Research involving relatively pure CBD is much more limited.
The CANBiS-AD trial, reported in 2024, was a small phase 2a study of CBD in people with Alzheimer’s disease and behavioural or psychological symptoms. Participants received CBD capsules beginning at 200 milligrams per day, with possible increases to 600 milligrams per day, or placebo.
Only 15 participants received study treatment: eight received CBD and seven received placebo. The trial was designed mainly to determine whether treatment was acceptable and feasible, not to provide definitive proof of effectiveness.
Researchers observed greater numerical improvements in several symptoms—including anxiety, agitation, hallucinations, irritability and caregiver distress—in the CBD group. However, the sample was far too small to establish reliable effectiveness. Baseline differences between the groups also made comparison difficult.
Dizziness occurred in 63% of CBD-treated participants and none of the placebo-treated participants. Falls were reported in both groups. Although treatment adherence and retention were good, the results should be regarded as preliminary evidence supporting a larger trial rather than confirmation that pure CBD works.
This study illustrates an important difference between a promising signal and a proven treatment. A few patients improving during a small pilot study may justify further investigation, but it does not establish how often the treatment works, which patients are most likely to benefit or whether the benefits outweigh the risks.
What Systematic Reviews Show
A systematic review with meta-analysis published in 2026 examined 10 randomised controlled studies involving 328 participants with dementia. The studies tested various cannabinoid preparations rather than CBD alone.
When the results were combined, cannabinoids appeared to produce a moderate reduction in agitation compared with placebo. However, there was substantial variation among the trials. They used different products, doses, treatment periods and measurement scales.
More importantly, when studies judged to have a high risk of bias were removed, the reduction in agitation was no longer statistically significant. The review found no clear overall improvement in total neuropsychiatric symptoms. Sedation was approximately twice as common with cannabinoid treatment as with placebo.
The review therefore supports cautious optimism, not certainty. Cannabinoids may reduce agitation, particularly in people with severe dementia, but the estimated benefit depends heavily on a small number of studies and is weakened by methodological limitations.
A separate 2026 analysis of cannabinoid treatments for agitation in Alzheimer’s disease also reported reductions in several agitation and neuropsychiatric symptom scores. However, it included only six randomised trials, with substantial differences between the products studied. Somnolence was again the clearest safety signal.
Emerging Evidence in Late-Stage Dementia
In July 2026, investigators presented topline findings from the phase 2 LiBBY trial involving 120 people with late-stage dementia who were receiving or eligible for hospice care. Participants had clinically significant agitation and were given either placebo or a purified combination containing 100 milligrams of CBD and 2 milligrams of THC twice daily.
After two weeks, the treatment group showed a 6.27-point greater reduction in agitation than the placebo group. Benefits continued through 12 weeks, when the difference was 8.23 points. Clinicians rated 87.2% of treated participants as improved at 12 weeks, compared with 23.6% receiving placebo.
Overall adverse-event rates were similar, although serious adverse events occurred more often in the active-treatment group. Investigators judged that the serious events were not caused by the medication.
These are encouraging results because people with advanced dementia are often excluded from clinical trials. Nevertheless, they were announced as conference topline results and require full peer-reviewed publication. The product also contained both CBD and THC, so the findings do not demonstrate that CBD alone would have the same effect. They apply primarily to people with severe, end-of-life dementia rather than every person experiencing mild or moderate agitation.
Can CBD Reduce Anxiety in Dementia?
The evidence for anxiety is less developed than the evidence for agitation.
Some dementia trials have measured anxiety as part of a broader neuropsychiatric symptom scale. The CBD-only CANBiS-AD pilot observed a numerical improvement in anxiety, while the CBD-rich oil trial found wider behavioural benefits. However, neither provides strong, independent proof that CBD reliably treats anxiety in dementia.
Anxiety and agitation also overlap. A person who becomes less restless may appear less anxious, even when anxiety has not been assessed directly. Improved sleep or reduced pain could indirectly reduce both symptoms.
It is therefore reasonable to say that CBD might reduce anxiety for some patients, but this possibility is based on indirect and preliminary evidence. Dedicated, adequately powered trials focused specifically on anxiety in dementia are still needed.
Safety Concerns in Older Adults
Safety is especially important because people with dementia are often physically frail and take several medicines.
Sedation and falls
Sleepiness is one of the most consistent adverse effects associated with cannabinoids. Sedation may look like successful treatment because the person becomes quieter. However, suppressing activity through excessive sedation is not the same as relieving fear or distress.
Drowsiness can reduce mobility, impair swallowing, worsen confusion and increase the risk of falls. Dizziness and changes in blood pressure may further increase that risk.
Liver effects
CBD can raise liver-enzyme levels. In a controlled FDA study of healthy adults taking approximately 250 to 550 milligrams daily for four weeks, 5.6% developed alanine aminotransferase levels more than three times the upper limit of normal. Liver abnormalities may not initially cause noticeable symptoms, making laboratory monitoring important when higher-dose CBD is used.
Drug interactions
CBD can affect liver enzymes that process many prescription medicines. It may change the concentration or effects of blood thinners, antiseizure drugs, antidepressants, antipsychotics and sedatives. Combining CBD with medicines that slow brain activity can intensify drowsiness.
The FDA warns that CBD may cause liver injury, interact with other drugs and increase sedation when used with alcohol or medicines prescribed for anxiety, sleep and stress. Other possible effects include diarrhoea, reduced appetite, irritability and changes in alertness.
Product quality
Commercial CBD oils should not be assumed to match the products used in clinical studies. A preparation may contain less or more CBD than stated, varying amounts of THC, pesticides, solvents or other contaminants.
In the UK, the Medicines and Healthcare products Regulatory Agency states that a CBD product promoted for treating a medical condition is considered a medicine and normally requires marketing authorisation demonstrating appropriate safety, quality and effectiveness.
Where CBD Might Fit into Dementia Care
CBD should not be the first response to new or worsening agitation.
The person should first be assessed for pain, infection, constipation, urinary problems, dehydration, medication effects, poor sleep and sensory impairment. Caregivers should also look at what happened immediately before the episode. A bath, change of caregiver, loud room or misunderstood instruction may be the trigger.
Non-drug approaches should generally be attempted first. These include maintaining a predictable routine, reducing noise and glare, responding to emotion rather than arguing about facts, offering food and fluids, providing meaningful activities, using familiar music, encouraging safe movement and speaking slowly in a reassuring tone.
When symptoms cause severe distress or danger and these strategies are insufficient, medication may be considered as part of an individualised plan. The risks of conventional treatments must also be recognised. Antipsychotics, for example, can sometimes reduce severe agitation but are associated with sedation, falls, stroke and increased mortality in older people with dementia. NICE recommends using them only when a person is severely distressed or risks harming themselves or others, at the lowest effective dose and for the shortest appropriate period.
A medically supervised cannabinoid treatment might eventually become an option for selected patients, particularly those with severe, treatment-resistant agitation or those receiving palliative care. Current evidence does not support replacing established assessment and care practices with unsupervised CBD.
Any proposed treatment should have a clearly defined goal, such as reducing aggressive episodes or enabling essential personal care. Caregivers and clinicians should record the frequency and severity of symptoms before treatment, monitor for adverse effects and stop the product when meaningful improvement does not occur.
Ethical Considerations
People with moderate or severe dementia may be unable to provide informed consent. Decisions may therefore involve relatives, legal representatives and healthcare professionals.
The objective should be to reduce suffering while preserving alertness, mobility, communication and dignity. A treatment that makes someone continually sleepy may reduce visible disruption but worsen quality of life.
Caregivers should ask whether the person appears more comfortable and engaged, rather than simply whether they are easier to manage. Treatment should never be used primarily as a chemical restraint.
Conclusion
CBD is a promising but still experimental approach to anxiety and agitation in dementia.
Small clinical trials have reported improvements, and a recent meta-analysis suggests that cannabinoid treatments may reduce agitation. Particularly encouraging results have come from a CBD-rich oil trial and a larger study of combined CBD and low-dose THC in people with late-stage dementia.
However, most positive evidence relates to products containing THC as well as CBD. Research on pure CBD remains extremely limited. Existing studies are generally small, use different doses and products, and provide little information about long-term safety. Evidence that CBD specifically treats anxiety in dementia is weaker than the evidence concerning agitation.
CBD can also cause dizziness, sedation, liver-enzyme abnormalities and clinically important drug interactions. These concerns are particularly relevant for frail older adults who take multiple medicines.
Consequently, CBD cannot currently be recommended as a standard treatment for dementia-related anxiety or agitation. It may eventually have a carefully supervised role for selected patients, especially when severe symptoms have not responded to other measures. Until stronger evidence is available, assessment of unmet needs, treatment of underlying medical problems and person-centred non-drug interventions should remain the foundation of care.
Families considering CBD should discuss it with a doctor or pharmacist who can examine the person’s complete medication list, consider the risks of falls and sedation, evaluate liver health and establish a structured monitoring plan. Consumer CBD products should not be treated as substitutes for research-grade or prescribed cannabinoid medicines.
The developing evidence provides a legitimate reason for hope, but not a reason to abandon caution. The key question is not simply whether CBD can make a person quieter. It is whether a carefully selected and monitored treatment can safely reduce fear, discomfort and distress while protecting the person’s dignity and remaining abilities.
About Author
