Cannabinoids show promise for dementia-related agitation, but evidence remains limited

Neuropsychiatric symptoms (NPS) are highly prevalent in persons with dementia and have been associated with adverse health outcomes. Neuropsychiatric symptoms can impose major physical, psychosocial, and financial burdens on caregivers while contributing to additional pressure on healthcare systems. Although atypical antipsychotics have received regulatory approval for treating specific NPS, such as agitation, their use in older adults has been linked to higher risks of mortality, cardiovascular events, and falls. Cannabinoids have emerged as promising pharmacological treatments for NPS in dementia, including agitation, anxiety, and depressive symptoms, owing to their behavior-modulating effects. The objective of this study was to provide a comprehensive assessment of the efficacy and safety of cannabinoids for NPS in dementia, which may help refine evidence-based guidelines for their use. We systematically reviewed studies from MEDLINE, Embase, PsycINFO, CINAHL, Cochrane Library, and ClinicalTrials.gov from inception to August 2025. Search terms relevant to cannabinoids and dementia were used. Randomized controlled trials on the use of cannabinoids for treating NPS in dementia with published results were included. Participant characteristics, including age, sex, and baseline cognition, were collected. Random-effects meta-analyses were conducted to examine standardized mean differences in NPS scores between cannabinoid and placebo groups. Risk of bias was assessed using the Cochrane Risk of Bias 2 tool (RoB 2) for randomized trials. Heterogeneity between studies was analyzed through I2 statistics. Additionally, subgroup analyses and meta-regressions were performed for variables of interest. Last, risk ratios for outcomes related to adverse events were calculated to evaluate the safety of cannabinoids in this population. Of the ten included studies (328 participants), eight assessed total NPS and nine reported on agitation in persons with dementia. Our results indicated that cannabinoids did not decrease total NPS (standardized mean difference [SMD]: - 0.18, 95% confidence interval [CI] - 0.48 to 0.12; p = 0.2) compared to placebo, but reduced agitation (SMD: - 0.52, 95% CI - 1.00 to - 0.05, p = 0.03) with high heterogeneity (I2 = 77.2%). However, after removing studies rated as high risk (RoB 2) for a sensitivity analysis, the result was no longer statistically significant (SMD: - 0.35, 95% CI - 0.79 to 0.10, p = 0.1). Subgroup analyses demonstrated that cannabinoids were beneficial with doses higher than 10 mg of tetrahydrocannabinol-equivalent (SMD: - 0.63, 95% CI - 0.98 to - 0.28, p < 0.01), and in patients with severe dementia (SMD: - 0.96, 95% CI - 1.75 to - 0.16, p < 0.01). The risk of overall adverse events did not differ significantly between cannabinoid and placebo treatments, though cannabinoids were associated with an increased risk of sedation (risk ratio = 2.09, 95% CI = 1.22 to 3.57, p < 0.01). This review provides up-to-date evidence that cannabinoids are efficacious for alleviating dementia-related agitation and are generally well tolerated in this population, though sedation was more commonly reported in the cannabinoid group. However, the data remained scarce for other NPS and requires further research.