This study provides real‐world information about medical cannabis use by rheumatology patients in the setting of cannabis recreational legalization, with information on patient characteristics, perceived effects, method and amount of use, access to product, and reasons for discontinuation. We report that 12.6% of 1000 unselected patients with a rheumatologist‐confirmed diagnosis of a spectrum of rheumatic diseases had ever used medical cannabis, with just over half continuing use. Discontinuation was mostly due to lack of effect, side effects, and cost. Ever medical users were mostly women in their 60s, with about half having previous experience with recreational cannabis. Pain relief was the most prevalent reason for use, but side effects were reported by 61% of users. Only 20% of medical users accessed cannabis entirely by the legal medical route, with most obtaining cannabis by various nonmedical routes, including the black market. It is therefore understandable and alarming that only a small minority of patients had knowledge of the molecular content or the daily amount of product used. Inhalation was the method of use for almost half of users, with two‐thirds using multiple methods of administration. The treating physician was aware of any cannabis use in only about one‐third of patients.

Several issues arising from this study deserve attention. Firstly, the method of access to cannabis is concerning. Most (80%) medical users were purchasing all or at least a portion of cannabis via a nonmedical route, with many reporting purchase via a personal contact, store/outlet, or the black market. Reasons for access via the nonmedical route may be multiple. The legal process to obtain medical cannabis can be cumbersome. A medical document authorizing use is submitted to a cannabis grower, who may then advise about strain and dosing and who arranges delivery of the product by postal service to a home address. Storefront assess is clearly simpler. Cost is also a plausible reason for nonmedical access. Medical cannabis is not reimbursed by public or private insurers for rheumatic complaints and is almost twice the price of black‐market cannabis. The average cost of medical cannabis is about $9.5/g, with patients reporting a monthly cost of between 100 and 300 Canadian dollars. Nonmedical cannabis, especially cannabis that is illegally bought, has risks associated with nonregulated products, with unknown molecular content and potential for contamination by other substances or infecting organisms.

Secondly, familiarity with smoking cigarettes or cannabis may have played a role for many patients in choosing inhalation as the method of administration. This is concerning in view of the large body of literature specifically addressing the added risks of smoking for patients with inflammatory rheumatic diseases (6). Vaping has, to date, been considered safer than smoking, but this study was conducted just prior to the publication on the emergence of severe respiratory illness related to vaping (7). Therefore, it is possible that patients perceived vaping as less harmful than smoking but preferred the more familiar method of inhalation. Furthermore, the effect of cannabis occurs within a few minutes of inhalation, which would give an immediate effect on symptoms in the moment.

Thirdly, the treating physician was unaware of any cannabis use for two‐thirds of the patients. Stigma may still be a factor in nondisclosure, but the widespread media coverage of cannabis may have prompted patients to self‐medicate without medical advice. Recreational cannabis legalization has allowed a cultural change, with cannabis no longer taboo, less risk of prosecution, and easier access. Furthermore, over half of nonusers would consider future medical cannabis use. Cannabis is now a Canadian household commodity and is offered by friends or family as a suggested therapy for medical illness.

Our current findings differ considerably from our earlier study in 2014. As anticipated, there has been a tripling in the numbers of patients using medical cannabis to almost 13%, and characteristics of patients have shifted toward an older, mostly female population with more prevalent inflammatory arthritis. We previously observed users to be mostly younger, unemployed men with osteoarthritis. Another notable change is that recreational cannabis experience was reported by about half of ever medical users in 2019, whereas this percentage was 80% in 2014. According to Statistics Canada's National Cannabis survey for the fourth quarter of 2019, cannabis for any reason had been used in the past 3 months by 16.7% of those surveyed, with half (about 8%) reporting use for medical reasons (8, 9). In contrast, only 1% of the Canadian population were registered with Health Canada as users of medical cannabis at that same time, emphasizing that most medical cannabis is accessed via the nonmedical route (10). Our findings that 6.5% of patients were current medical cannabis users align with the Canadian survey data. We also perceive that there is less of a blur between medicinal and recreational use, with some patients clearly identifying their use as one or the other or both. It could be argued that patients in our original study underreported medical cannabis use. As the prevalence of recreational ever use was similar in the two studies, we contend that the findings of the present study are reliable.

Previous surveys reporting on the efficacy of medical cannabis in chronic pain conditions generally report patient satisfaction with use, improved pain and psychological well‐being, and only limited side effects (11, 12, 13, 14, 15, 16, 17, 18). It is noteworthy that almost half of users in this present cohort had discontinued cannabis use, giving lack of effect or side effects as their reason for discontinuation. Those continuing use perceived benefits in multiple domains, but there were no differences in PGA, PtGA, or pain VAS between those continuing and discontinuing use. Furthermore, over 60% of current medical users reported some side effects. Our present finding of discontinuation of use is more likely reflective of real‐world experience because many studies reporting favorable effects may be subject to bias, with study settings often in designated cannabis clinics or with data from cannabis dispensaries. Alternately, compliance and success with use may be better when a patient is followed by a qualified health care professional with guidance about products and dosing, such as reported by clinics in Israel (18).

We believe that Canadian cannabis recreational legalization has influenced medical cannabis use with increased tolerance for use, social acceptability, removal of stigma, and perception of safety. Widespread media coverage, easier access, less risk of prosecution, and more affordable product on the illegal market are also influential factors. Cannabis is not innocuous, and use has implications for both patient and societal safety. Lacking long‐term studies in patients populations with comorbidities and concomitant medication use, safety issues have been recently highlighted by the National Academies of Sciences, Engineering, and Medicine and extend from the immediate risks of impaired cognition, psychomotor control, and memory to the long‐term risks for psychological and pulmonary health (19, 20).

Limitations must be considered. Although the study was anonymous, some patients may have refrained from acknowledging cannabis use. The information provided about cannabis used, including molecular content and daily amount used, was limited, likely because most cannabis was accessed via a nonmedical route. Additional information that is pertinent but unavailable included the baseline levels of symptoms prior to initiating cannabis, duration of cannabis use for all medical users, and treatment adjustments related to cannabis use. This is a single‐setting study that included only patients seen in a specialist rheumatology clinic; therefore results may not be reflective of a broader patient population and cannot be extrapolated to other countries where recreational cannabis is illegal or where medical cannabis is prescribed by physicians and/or is reimbursed by public or private insurers (21). A strength is that we have repeated a previous study, using the same methods, in a similar setting at two distinct time points, namely before and after recreational cannabis legalization. This information can be seen as a snapshot of the real‐world experience of patient use of medical cannabis, using a natural experimental setting.

We report a tripling of medical cannabis use in a cohort of rheumatology patients compared with 5 years ago, confirming our hypothesis that cannabis recreational legalization increases medical use. Almost 1 in 10 patients had tried cannabis as a therapy. Unlike many reports attesting to efficacy, we found that almost half of patients discontinued use after a mostly self‐administered trial. There are concerns about prevalent inhalation of cannabis, access via the nonmedical route, health risks for immunocompromised patients, and poor patient knowledge of daily amount or molecular content of the cannabis. Physicians must be aware that many patients may be self‐medicating with cannabis and that they are at risk for acute and chronic adverse effects that may be amplified by the underlying disease and prescribed medications. Our findings suggest that physicians may no longer be the gatekeepers for patients regarding cannabis use, with prevalent use slipping outside the bounds of ideal clinical care.