Discussion

In this exploratory study of older adults attending a Canadian cognitive assessment clinic, approximately 1 in 10 patients with information collected on cannabis use status reported using cannabis. This is similar to general population data showing 8% of Canadians aged 65 and older used cannabis products in 2020³, which is higher than the 2.9% prevalence observed in older Americans.^{Reference Maxwell, Jesdale and Lapane28}

There are few data on the use of legal, recreational or medicinal use of cannabis among community-dwelling persons living with cognitive disorders or dementia. In a focus group study conducted in Los Angeles, USA, in 2019, at which time cannabis use was legal in California for medicinal and recreational use, 3 out of 26 caregivers indicated that they provided cannabis to a person with dementia.^{Reference Kaskie, Bobitt and Herrera20} However, systematic surveys of clinic patients or persons with cognitive disorders in the general population are lacking.

Two-thirds of those who used cannabis in our study reported recreational use, contrasting with previous studies where older adults typically framed their cannabis use as medicinal.^{Reference Baumbusch and Sloan Yip8} Factors contributing to cannabis use in this demographic may include the positive effects of cannabis on reducing agitation and aggression in those with dementia as well as its broader benefits for sleep, pain and mood symptoms.^{Reference Walther, Schüpbach, Seifritz, Homan and Strik11,Reference Herrmann, Ruthirakuhan and Gallagher14,Reference Kaskie, Bobitt and Herrera20,Reference Outen, Burhanullah and Vandrey29,Reference Liu, Chau, Ruthirakuhan, Lanctôt and Herrmann30} This may also reflect societal trends toward greater acceptance of recreational cannabis use.^{Reference Lau, Sales, Averill, Murphy, Sato and Murphy31} This underscores the relevance of investigating the factors contributing to the sustained use of cannabis in this age group, particularly within the context of cognitive concerns.

We found that certain characteristics distinguished cannabis users from non-users. Cannabis users were younger, which reflects a generational shift in the acceptance of cannabis use.^{Reference Lloyd and Striley7} Cannabis users were also more likely to be former or current cigarette smokers, which corresponds with smoking being a risk factor for cannabis use and vice versa.^{Reference Weinberger, Wyka and Goodwin32} Cannabis use is associated with increased initiation of cigarette smoking, less smoking cessation and higher risk of smoking relapse.^{Reference Weinberger, Delnevo and Wyka33} The mechanisms underlying this association involve nicotine and cannabis both acting on the brain’s reward system, potentially making individuals who smoke cigarettes more susceptible to using cannabis.^{Reference Martin-Soelch, Kobel and Stoecklin34}

The higher prevalence of mood disorders (56.5% vs 26%, p < 0.05) among those who reported cannabis use in our study raises questions about the potential bidirectional relationship between cannabis use and mood disorders in older adults. Although our cross-sectional study does not allow for causal inferences, this finding suggests that cannabis could either be used as a form of self-medication for mood symptoms or be a contributing factor to the development or exacerbation of mood disorders through various biological and psychosocial pathways.^{Reference Khantzian35,Reference Karila, Roux and Rolland36} A systematic review by Sourkhou et al. found that cannabis use is associated with increased depressive and manic symptoms, as well as a higher likelihood of developing major depressive disorder and bipolar disorder.^{Reference Sorkhou, Dent and George37} Cannabis use was linked to an unfavorable prognosis in both mood disorders, suggesting a negative influence on the course and outcomes of these disorders.^{Reference Sorkhou, Dent and George37} Kuhns et al. report a bidirectional relationship between cannabis use and the onset of depression, with worse prognoses for both depression and bipolar disorder, including a higher risk of suicidality.^{Reference Kuhns, Kroon, Colyer-Patel and Cousijn38} Evidence suggests an association between mood disorders and cannabis use across different age groups.^{Reference Mann, Cheung and Stoduto39} This prompts the need for longitudinal research to elucidate the temporal dynamics of this relationship, specifically within the demographic of older adults with cognitive impairments, as this could have implications for mental health interventions and screening strategies.

Multiple small clinical trials have explored the use of medical cannabis^{Reference Peprah and McCormack40} or synthetic cannabinoids (drugs that stimulate cannabis receptors)^{Reference Bahji, Meyyappan and Hawken23} for the treatment of neuropsychiatric symptoms of dementia, including agitation. However, when interpreting results of cannabinoid research, it is important to recognize that there are chemical differences between phytocannabinoids (derived from the cannabis plant), endocannabinoids (endogenously produced metabolites that have activity on cannabinoid CB1 and CB2 receptors)^{Reference Lu and Mackie41} and synthetic cannabinoids (chemically synthesized derivatives of cannabinoids for prescription or illegal use) that may affect their medicinal properties. Canadian guidelines for medical management of dementia do not address cannabis use,^{Reference Ismail, Black and Camicioli42} while Canadian guidelines for managing behavioral and psychiatric symptoms of dementia recommend the use of nabilone (a synthetic cannabinoid) for the treatment of refractory agitation with low-quality evidence, without addressing the use of phytocannabinoids.^{Reference Hatch, Seitz and Bruneau43}

The impact of cannabis on cognition is controversial. A systematic review of cannabis use in older adults found that, while conclusions were limited by variability in the products used and quality of the research methods, modest reductions in cognitive performance were seen in users taking higher doses and among those with heavier lifetime use.^{Reference Scott, Brennan and Benitez44} On the other hand, cannabis has been reported to reduce neuroinflammation and prevent aggregation of amyloid and tau in preclinical studies^{Reference Watt and Karl16–Reference Ferrer19} and therefore might reduce the rate of cognitive decline in Alzheimer’s disease. However, this hypothesis has not been tested in adequately powered clinical trials. Instead, most trials have tested the effects of cannabis on neuropsychiatric symptoms, and surprisingly, most of them did not report the effects on cognitive test scores. A systematic review^{Reference Scott, Brennan and Benitez44} found evidence from only two small (50 or fewer subjects each) randomized controlled trials: one trial found no effect of THC on episodic memory,^{Reference van den Elsen, Ahmed and Verkes12} but another found that the synthetic cannabinoid nabilone improved MMSE over 6 weeks of treatment.^{Reference Herrmann, Ruthirakuhan and Gallagher14}

A limitation of this study is that many patients did not have their cannabis use status collected. Participants with cannabis information were more likely to be older and have a history of current or former smoking, alcohol use, insomnia, OSA, mood disorders, anxiety disorders, stroke and agitation (Table 1). Participants included also had a higher prevalence of dementia, MCI and SCD, whereas those excluded had a higher prevalence of other cognitive diagnoses, including psychiatric disorders, neurodevelopmental disorders and normal cognition. The higher prevalence of these conditions among included participants may indicate an underlying bias in data collection, where clinicians might have been more diligent in questioning patients on cannabis use if they presented with more complex health profiles and more pronounced health issues. Patients and family members may have also underreported cannabis use due to stigma or concerns about provider perceptions, leading to an underestimation of cannabis use. The limited sample size results in an underpowered study, and these findings need confirmation through future larger studies. However, even if all patients with missing cannabis information were non-users, the prevalence of use (∼5%) would still be appreciable, indicating that clinicians should systematically ascertain cannabis use in all patients attending a cognitive assessment clinic.

Additional limitations of our study that impact the interpretation and generalizability of the findings should be considered. First, our data collection form only accounts for participants who currently use cannabis and does not account for those who may have used cannabis in the past. Additionally, among current users, we did not collect information on dosing and method of cannabis consumption. Due to the limited sample size, we could not perform multivariable logistic regression to identify characteristics independently associated with cannabis use. Additionally, the duration of cognitive symptoms and cannabis use is unclear, leaving open the question of whether cannabis contributes to cognitive decline and presentation to the memory clinic or whether it is used to manage symptoms associated with cognitive decline.

These limitations underscore the need for larger longitudinal studies with prospectively collected data on cannabis use. These studies should collect information on lifetime as well as current use, the reasons for use including symptoms that prompted use, whether the decision to use cannabis was made by the person affected by the cognitive disorder or by their care partner and symptoms of dependence or abuse. Determining the exposure to THC and CBD is very complex due to the variety of products available in Canada, which may include solid concentrates, hash, dried flower, edibles, vape oils, tinctures, topicals, oral oils in capsules, oral oils in drops and drinks.^{Reference Wadsworth, Rynard and Driezen45} While Health Canada requires product labels that report the quantity of THC and CBD, the accuracy of these labels has been called into question, as the products are not manufactured to drug grade. For example, one study found that 40% of cannabis products in Ontario had assayed levels of THC that were outside variability limits^{Reference Doggett, Fein, Campbell, Henriquez, Busse and MacKillop46}; most commonly, the product was labeled as being more potent than it actually was.

Overall, this exploratory study provides preliminary insights into cannabis use among older adults in Canada with concerns or diagnoses of cognitive impairment, revealing that a notable 10% use cannabis, primarily for recreational purposes. This underscores the importance of regularly screening for cannabis use to minimize potential adverse effects from interactions with other medications or other side effects, such as dizziness or falling.^{Reference Bertram, Porath and Seitz47} As societal attitudes toward cannabis evolve and its accessibility increases, understanding its effects on older adults with cognitive disorders becomes increasingly critical. Future research should aim to include larger sample sizes and longitudinal designs to confirm these preliminary findings and should investigate patients’ motivations for use, the relationship of cannabis use with mood disorders and the role of cannabis in cognitive decline. Understanding these dynamics is important for healthcare providers to offer informed guidance and support to older adults with cognitive concerns who use or are considering using cannabis.