The rise in CBD accessibility is in part due to the passage of the Hemp Farming Act of 2018, which removed hemp with less than 0.3% THC from its previous Schedule I controlled substance designation (Comer, 2018). Marketing efforts have largely contributed to the rising popularity of CBD, and it has been touted as a treatment for numerous ailments. Cannabinoids offer an interesting future and relatively novel modality for pain reduction in musculoskeletal disease. The use of CBD for arthritic conditions is beginning to garner attention and may provide a therapeutic avenue with a favorable side effect profile.

In the present study, we found a high prevalence of CBD use to treat joint pain secondary to arthritis. More than half of individuals (60.5%) reported a reduction or discontinuation of opioids, acetaminophen, and/or anti-inflammatories. In terms of the perceived effects of CBD on pain, physical function, and sleep quality, many patients using CBD reported symptomatic improvements. Furthermore, we sought to determine if participants with different types of arthritis had differences in perceived benefits of CBD. Subgroup analysis based on arthritis etiology failed to show differences between groups in terms of pain improvement and sleep quality. The osteoarthritis group had improved subjective physical function scores compared to the autoimmune arthritis group; however, it is unknown if this statistical finding is clinically meaningful. In terms of pain improvement, participants reported a 44% reduction in pain and a 2.58-point difference in the NRS after using CBD. The magnitude of pain alleviation in our study participants, as measured by percentage reduction in pain or point difference, both surpass the clinically important differences reported in literature (2-point reduction or 30% reduction in pain, respectively) (Farrar et al., 2001, Salaffi et al., 2004). Subgroup analysis demonstrated that patients with osteoarthritis reported greater reductions in NRS score than RA and other autoimmune arthritis groups, which may suggest that the efficacy of CBD may be better in patients with osteoarthritis. Research involving CBD for arthritis pain is lacking; however, there have been clinical studies involving CBD and chronic pain. Gulbransen et al. found improvements in quality of life, pain, depression, and sleep quality in 400 patients with non-cancer-related chronic pain. Patients with arthritic conditions were included in their chronic pain cohort, but they did not stratify by reason for chronic pain in their analysis (Gulbransen et al., 2020a, b).

Two of the more commonly utilized medications that participants used for pain relief for arthritis were non-steroidal anti-inflammatory (NSAIDs) and acetaminophen. Boureau et al. compared ibuprofen and paracetamol in patients with arthritis of the knee or hip and found superior pain reduction with NSAID use. Treatment with NSAIDs resulted in a 2.3-point reduction whereas paracetamol resulted in a 1.5-point reduction (Boureau et al., 2004). A systematic review that investigated the use of acetaminophen for pain reduction reported only minimal reductions in pain for patients suffering from hip and knee arthritis, questioning the true efficacy (Leopoldino et al., 2019). Additionally, they found that when comparing acetaminophen to placebo, there was only a 3% absolute reduction in pain and relative reduction in pain of 5% (Leopoldino et al., 2019). The use of topical NSAIDs has also been studied as an alternative to oral NSAIDs, due to the unwanted systemic side effects of the latter. A large meta-analysis found only a minimal pain reduction for topical NSAIDs versus placebo and importantly it was noted that at no time point did topical NSAIDs reach the MCID (Zeng et al., 2018). These conflicting findings support the need for safe, alternative treatments for patients to manage their pain.

One of the particularly interesting findings of this investigation was the proportion of respondents who reduced or discontinued the use of other medications (60.5%). Medications frequently prescribed to manage the symptoms of arthritis can have unwanted side effects. The use of anti-inflammatories has been associated with numerous complications, including the risk of serious gastrointestinal bleeding, ulcerations, and cardiovascular events (Wongrakpanich et al., 2018). Furthermore, it has been reported that up to 50% of patients with arthritis are dissatisfied with the efficacy of anti-inflammatories to treat their pain (Taylor et al., 2013). In our study cohort, of those patients that reduced or discontinued medications after starting CBD, 31.1% reduced anti-inflammatory medications and 17.8% discontinued them altogether.

Despite a lack of evidence supporting long-term opioid use for arthritic pain, a recent study evaluating prescribing practices found that 12.8% of patients were prescribed opioid medications for knee arthritis (Gwam et al., 2021). The U.S. Centers for Disease Control and Prevention (CDC) reported nearly 450,000 opioid-related deaths in the past decade. Given the severity of the current opioid epidemic, finding alternative treatments for pain is vital to public health (Mattson et al., 2021). In our cohort, 8.6% of patients reduced and 18.9% discontinued the use of opioids. These findings suggest that CBD could be an alternative to opioids for the treatment of arthritic pain. Cannabinoids have been shown effective in human trials for chronic pain (Campbell et al., 2018; Moeller-Bertram et al., 2019; Wiese & Wilson-Poe, 2018). In a recent open-label study involving patients with chronic pain and more than 2 years of opioid use, 53.2% of participants were able to reduce opioid medications, lessen their pain, and improve sleep quality after 8 weeks of consuming oral CBD hemp extract (Capano et al., 2020). These findings are congruent with what we observed in our cohort.

Between the different arthritis types, a larger percentage of patients with osteoarthritis (66.2%) reported reductions in other medications used compared to RA (56.3%) and autoimmune arthritis (53.7%). Additionally, we noted that increased dosing frequency and length of use of CBD led to higher rates of other medication reduction and cessation. When controlling for age, dose, gender, the type of arthritis was not found to be an independent predictor. Patients who used CBD more than once a day (OR: 16.9, p < 0.001) and once a day (OR: 5.89, p = 0.006) were more likely to reduce or stop other medications, when compared to those who took the medication less than once a month. Similarly, patients taking CBD for a period of more than three years were 3.6 times more likely to reduce or cease other medications, when compared to those taking CBD for less than 30 days. One important consideration is that patients who reduce or stop other medications may also be more likely to use CBD regularly and for longer durations as a replacement for the alternative medications.

Despite all of the study participants having a diagnosis that was made by a physician, only 45.6% had spoken to a health care professional about CBD use for their condition. This may be because CBD is readily accessible and does not require a prescription. Another possibility is that healthcare professionals may not be discussing CBD use with their patients. Although CBD is available for purchase without consultation with a doctor, the authors recommend that patients discuss this with their doctor before use. Cannabidiol could have unwanted side effects and/or interactions with other medications. Most of the literature regarding the adverse events of CBD use has come from clinical trials for the FDA-approved indication to treat seizure disorders (Corroon & Kight, 2018). Recently, the World Health Organization released a critical review on CBD and found it is generally well tolerated with a good safety profile and does not have abuse potential (Expert Committee on Drug Dependence, 2018). Previous cross-sectional analyses have found at least one reported side effect in 29–59% of patients using CBD for different indications which is in concordance with our findings (41%) (Corroon & Phillips, 2018; Devitt, 2019; Moltke & Hindocha, 2021). Furthermore, most of the reported side effects in our study were mild (84%), which echoes the findings in clinical trials and cross-sectional studies.

The study has limitations that must be considered when interpreting these results. First, the survey used was created by the authors and was based on clinical expertise as well as other published studies (Corroon and Phillips, 2018, Bhamra et al. 2021, Moltke et al, Schilling et al. 2021, Moeller-Bertram et al. 2019, Devitt 2019). Due to this novel survey design, there is concern for the validity of the survey. In order to improve validity and accuracy, the survey underwent multiple rounds of internal review to reduce errors and to monitor the quality of questions as described by Passmore et al. 2002. Importantly, the structure of the survey was based on the Likert scale and NRS, which have been validated in the literature and therefore improve the validity of the custom survey used in the investigation (Safikhani et al. 2018, Passmore et al. 2002). Second, the study population was a self-selected sampling of convenience and hence might not be representative of the general population. Of the respondents, 176 (30%) patients with arthritis had not tried CBD, whereas 428 (70%) endorsed CBD use. Two recent studies found a prevalence of CBD use in patients with arthritis to be between 24 and 57% which is lower than our findings (Deckey et al., 2021, Nowell, 2019). The authors feel it is unlikely that 70% of patients with arthritis in the general population have tried CBD. This higher prevalence may be due to a selection bias secondary to patients who have used CBD being more likely to participate in the study. Third, patients with polarizing views on experiences using cannabinoids—whether they are positive or negative—may be more likely to have participated in the study, which could lead to either positive or negative bias. Additionally, because the survey was distributed online through social media posting and patient representative groups, patients with arthritis who have limited connectivity to the internet are potentially underrepresented. Fourth, patients submitted subjective reports of their symptoms, and the survey was not distributed in a controlled manner immediately after CBD use, making the results susceptible to recall bias. Designing a prospective study may allow assessment of pain before and after CBD use in real time to circumvent recall bias limitation. Lastly, a large proportion of patients did not know the type of CBD product (e.g., isolate, broad spectrum, full spectrum) they consumed, and therefore, the authors were unable to evaluate associations between patient-perceived effects of CBD and product type. Different types of CBD and the various modes of administration could have possible varying effects, which we were not able to capture in this investigation.