Self-Medication Paths

Abstract

Objectives:

Cannabis is used by one-third of people living with chronic pain to alleviate their symptoms despite warnings from several organizations regarding its efficacy and safety. We currently know little about self-medication practices (use of cannabis for therapeutic purposes without guidance), mainly since the legalization of recreational cannabis in countries such as Canada has expanded the scope of this phenomenon. This study aimed to describe legal cannabis self-medication for pain relief in people living with chronic pain and to explore perceptions of the effectiveness and safety of cannabis.

Methods:

A cross-sectional descriptive study was performed among 73 individuals living with chronic pain and using cannabis (Quebec, Canada). Data collection using telephone interviews occurred in early 2023.

Results:

Results indicated that 61.6% of participants reported using cannabis without the guidance of a health care professional (self-medication). Surprisingly, among those, 40.0% held a medical authorization. Overall, 20.6% of study participants were using both medical and legal nonmedical cannabis. Different pathways to self-medication were revealed. Proportion of women versus men participants self-medicating were 58.2% versus 70.6% (P=0.284). In terms of perceptions, 90.4% of the sample perceived cannabis to be effective for pain management; 72.6% estimated that it posed no or minimal health risk.

Discussion:

Cannabis research is often organized around medical versus nonmedical cannabis but in the real-world, those 2 vessels are connected. Interested parties, including researchers, health care professionals, and funding agencies, need to consider this. Patients using cannabis feel confident in the safety of cannabis, and many of them self-medicate, which calls for action.

Despite chronic pain being a prevalent condition with numerous biopsychosocial influences, its management remains fraught with challenges.¹ While it is acknowledged that its treatment should be multimodal² and with the collaboration from various health care professionals (HCPs),³ several barriers to optimal pain management persist.⁴ Cannabinoids (cannabis/synthetic cannabinoids) are recommended as a third-line treatment for neuropathic pain in Canadian guidelines.⁵ They, however, do not feature among the recommended treatments for other pain-related conditions such as low back pain and fibromyalgia.^6–8 Given the limited efficacy of pharmacological pain treatments^9–12 and their side effects,¹³ a growing number of studies focus on the therapeutic potential of cannabinoids.^14–16 Although they are sometimes seen as an alternative to opioids,^17,18 a lot of work remains to be done to better understand cannabis use patterns among patients with pain. In 2021, the International Association for the Study of Pain claimed it could not endorse cannabinoids use for pain management.^19,20 This position echoes the Canadian Rheumatology Association and National Institute for Clinical Excellence’s^21,22 and is driven by the lack of strong evidence on the efficacy and safety of cannabis.^19,20,23,24 Interestingly, however, one-third of people living with chronic pain use cannabis,²⁵ most people using cannabis in the general population do so to relieve their pain^26,27 and their perception toward cannabis appears positive.^28,29 In addition, some expert panels are in favor of cannabis use for pain management with some careful follow-up.⁸

In Canada, medical cannabis, which must be authorized by a physician or a nurse practitioner,³⁰ has been legal since 2001 (Marihuana for Medical Purposes Regulations, followed by the Access to Cannabis for Medical Purposes Regulations). In the United States, medical cannabis is legal in 37 states.³¹ Nonmedical cannabis bought through retail stores (often designated as “recreational cannabis”) is legal in countries such as Canada since 2018 and in some US states. Illegal sources (black market) still remain a means to obtain cannabis.³² Under the Canadian framework, a cannabis user could use cannabis obtained from the medical program for recreational purposes, while another individual could use cannabis from recreational retail stores to treat a symptom. Cannabis legalization paired to the difficulty in finding an effective treatment for chronic pain leads some individualss to use nonmedical cannabis as a treatment without prior consultation with health care professionals.^17,33–35 These individuals, therefore, find themselves self-medicating,^36–39 which can be defined as obtaining and consuming one or more drugs without the intervention of a physician either for the diagnosis, the prescription, or the treatment monitoring.^38,40 In other words, in our context, it refers to the use of cannabis without the guidance of an HCP, whether it is authorized (medical cannabis) or not (nonmedical cannabis) by a HCP. In fact, self-medication can also occur in the context of medical cannabis, considering that according to the Cannabis Act in Canada,⁴¹ an initial authorization for medical cannabis does not guarantee that patients will receive the following support from a HCP, as the law does not mandate this medical guidance.⁴² Considering that strong evidence on the efficacy and safety of cannabis for pain relief are currently limited,^23,24,33 it is relevant to better understand self-medication practices. In fact, using cannabis without the guidance of a HCP can carry important risks due to the lack of oversight, such as incorrect dosages, which can lead to either under-treatment or adverse effects from overuse.^43,44 In addition, users may not be aware of potential interactions between cannabis and other medications they are taking. Self-medicating without support regarding legal and safety issues also has the potential to lead to problems and exposure to unsafe products.^44–46 Using cannabis under professional guidance can help mitigate these risks through proper dosage, monitoring, and comprehensive care.

The review report on the Cannabis Act produced by the Government of Canada mentions the importance of continuing research on cannabis.⁴⁷ To date, few studies have addressed self-medication with cannabis, and these have been conducted in countries where the health care system is not comparable to Canada (Germany;⁴⁸ USA⁴⁹). In Canada, a recent study explored self-medication with cannabis, but was not specific to a chronic pain population and did not include nonmedical cannabis users.⁴⁵ However, 53% of participants reported engaging in self-medication with cannabis to alleviate pain.⁴⁵ To enhance health care information, mitigate risks associated with cannabis-related harm and support a patient-centered approach, the present study aimed to describe legal cannabis self-medication for pain relief in people living with chronic pain. In addition, their perceptions toward effectiveness and safety of cannabis were explored.

METHODS

Study Design and Population

This cross-sectional descriptive study was performed in a convenience sample of adults living with chronic pain. More specifically, to be eligible, participants had to: (1) live with pain for more than 3 months (ICD-11 definition⁵⁰); (2) currently use medical and/or nonmedical cannabis or have used cannabis in the last year (all reasons for use considered); (3) be at least 18 years old; (4) be able to answer questions over the phone in French. No exclusion criteria were applied. The study received ethics approval from the Université du Québec en Abitibi-Témiscamingue’s research ethics committee (#2022-11–Audet, C.). In terms of patient engagement, a person with lived experience of chronic pain (CB), was a full-fledged team member and participated in the project conceptualization, questionnaire development, and interpretation of results.

Recruitment and Data Collection Methods

As a potential participant pool, the list of individuals included in a previous study, that is, the Chronic Pain Treatment (COPE) Cohort,⁵¹ was utilized. This cohort comprises 1935 French-speaking adults living with pain for more than 3 months, recruited through the web in all regions of the province of Quebec (Canada) between June and October 2019. COPE Cohort participants were previously found to be comparable to random (representative) samples of Canadians living with chronic pain in terms of age, employment status, level of education, pain duration, mean pain intensity, and most common pain locations.⁵¹ Most (n=1114) expressed their interest in being approached by e-mail for future studies conducted by our research team.

For the present study, e-mail invitations were sent to potential participants (convenience sample) until we obtained a sample size of 73 individuals. This sample served as a well-balanced compromise, allowing for study completion within an acceptable timeframe and precise estimation of descriptive statistics, all while ensuring the feasibility of conducting telephone interviews. As compared with a web-based questionnaire, this data collection approach was chosen for several reasons, including the novelty and anticipated complexity of the phenomenon of self-medication with medical and/or nonmedical cannabis, and the possibility to minimize missing data. In fact, telephone interviews allowed the research team the opportunity to clarify misunderstandings and provide examples. At this level, we consider this choice to be a strength of the study. Furthermore, participants of the COPE Cohort had expressed a desire for opportunities to have more personal contact with the research team (as opposed to web-based self-administered questionnaires). As compared with a web-based questionnaire, administering the interviews was thus deemed to be an adequate methodological decision and the research team was able to recommend adequate resources to many participants in case of psychological distress.

From January 10 to April 25, 2023, e-mail invitations were sent to potential participants by the principal investigator of the COPE Cohort (above-mentioned pool of 1114 participants who expressed their interest in being approached for future studies; until the reach of our sample size). These e-mails included a description of the study, along with an information and consent letter in an attachment. If interested, individuals were invited to respond to the e-mail, providing their name, phone number, and the best time for a telephone interview. Participants were generally contacted within a week of their response. Informed consent was confirmed by responding to the invitation e-mail and initiating the phone interview. The interviews, lasting ∼60 minutes, were conducted by a Master’s degree student and registered social worker, and were computer-assisted (LimeSurvey). A standardized interview guide (questionnaire—Appendix 1, Supplemental Digital Content 1, http://links.lww.com/CJP/B146) was developed by our multidisciplinary research team (a patient, a social worker, a pain clinic anesthesiologist, a nursing research scientist, an opioid misuse and addiction scientist, and an epidemiologist) and pretested with 6 individuals with lived experience of chronic pain, including men and women with different levels of education. These pretests were used to estimate completion times and to determine any modifications required, which led to specify minor elements, such as adding examples to statements.

Study Variable

Cannabis Self-Medication for Pain Relief

There is currently no consensus on the definition of self-medication.^36,38,39,52 However, we retained the concept of individual action,⁵² without the guidance of a health care professional. In this study, cannabis use for pain relief and the conditions under which cannabis was used were explored using questions developed by the research team to cover 3 topics: (1) medical cannabis authorization (yes/no); (2) use of legal nonmedical cannabis (yes/no); (3) if they received guidance from a registered HCP for their cannabis use (yes/no). Before answering these questions, participants were carefully explained the difference between medical (medical authorization under the Cannabis Act) and legal nonmedical cannabis (accessed through the recreational route; Government-operated in-person and online stores in the province of Quebec). They were presented with examples of registered HCPs, such as physicians, nurses, and pharmacists, who can play a role in supporting cannabis use. Details were provided to operationalize “guidance,” that is, accompaniment, expert advice, recommendations, insights, and assistance. Since it is known that patients who receive authorization through a medical cannabis program do not necessarily have cannabis-related medical follow-up,⁴⁴ participants could report using cannabis without the guidance of a health care professional (self-medication), whether it was authorized (medical cannabis) or not (legal nonmedical cannabis) by a HCP.

Self-Perceived Cannabis Effectiveness and Safety

Self-perceived cannabis effectiveness and safety were measured using 5-points Likert scales, as used in the Medical Cannabis Access Survey, to facilitate data comparison:⁵³ (1) “In your experience, what is (or has been) the effectiveness of cannabis in managing your pain? (not at all effective/slightly effective/moderately effective/very effective/extremely effective/don’t know); (2) In your opinion, is the use of cannabis for pain management a risky practice for your health?” (no risk/minimal risk/moderate risk/high risk/don’t know). Using closed-ended questions, participants were also asked if they perceived that cannabis could lead to addiction, if consuming cannabis was riskier, less risky, or equivalent to consuming prescription opioids (eg, morphine, fentanyl, hydromorphone were provided), and if consuming cannabis was riskier, less risky, or equal to consuming illicit drugs in general (examples that resonate with patients such as speed, ecstasy, GHB, cocaine, magic mushrooms were provided).

Cannabis Use Characteristics

The questionnaire administered over the phone contained items related to cannabis use, including reasons for use, products used, methods of use (semiclosed-ended questions, allowing participants to select all options that applied), concentration of their products (THC-dominant, CBD-dominant, balanced products, multiple products with different combinations of THC and CBD), and frequency of use (times per day). These items were based on the 2022 Canadian Cannabis Survey,⁵⁴ the 2022 Quebec Cannabis Survey,⁵⁵ and the Medical Cannabis Access Survey⁵³ to facilitate data comparison. A question about cannabis use before the age of 24⁵⁶ (yes/no) was also added.

Pain and Psychological Variables

The pain profile section of the questionnaire covered pain location (1 item), duration (1 item), and intensity (11-point numeric rating scale about average pain intensity in the last 7 days).⁵⁷ Neuropathic component to the participant’s pain was evaluated using the 4-item DN4 (Douleur Neuropathique en 4 Questions)—Interview part (a score >3/7 indicates a likely presence of a neuropathic component).⁵⁸ The DN4 is one of the most used and validated screeners of neuropathic pain.⁵⁹ Pain interference was measured using the Brief Pain Inventory (BPI) 7-item interference scale.⁶⁰ Items include general activity, mood, walking ability, normal work, relations with others, sleep, enjoyment of life, personal care, recreational activities, and social activities in the past 7 days. The BPI is one of the most commonly used validated pain measures.⁶¹ Pain catastrophizing was assessed using the 4-item Brief Pain Catastrophizing Scale (BriefPCS),⁶² an abbreviated version of the PCS validated for a quick screen of exaggerated negative orientation toward pain. Psychological distress was measured using the 4-item Patient Health Questionnaire (PHQ),⁶³ an ultrabrief screener for anxiety and depression validated in a great diversity of clinical and nonclinical populations.⁶⁴

Sociodemographic Variables

Information was collected on participants’ age, sex at birth, sex identity and country of birth (as social determinants of health⁶⁵), region of residence (remote vs. nonremote), employment status, family annual income, and education.

Statistical Analysis

Descriptive statistics, such as counts, proportions, means, SD, medians, minimums, and maximums were used to compile participant characteristics and cannabis product utilization. To facilitate the comparison of our results with those of the 2022 Quebec Cannabis Survey,⁵⁵ common variables measured in both data collections were depicted using dual bar charts (comparisons of the proportions without statistical testing). To address the first objective about cannabis self-medication, the 3 above-mentioned questions (medical authorization yes/no, use of legal nonmedical cannabis, and guidance from a registered HCP) were combined to create a tree diagram (conceptual map). Proportions of individuals self-medicating, with medical authorization, and/or using legal nonmedical cannabis were computed for the entire sample, and then stratified by sex and age groups (women vs. men; individuals 65 years or older vs. below 65 years old; χ² tests). 95% CI were computed to assess the precision of the estimation of our primary statistics of interest. In an exploratory analysis (the sample size was not planned for an analytical study design/multivariable analyses), characteristics of individuals self-medicating versus not self-medicating were explored using bivariate comparisons (t tests and χ² tests). To address the second objective, self-perceived cannabis effectiveness and safety were analyzed using descriptive statistics. All data were analyzed using SPSS Statistics 19 (IBM Corp., Armonk, NY).

RESULTS

Sample Characteristics

Table 1 presents the characteristics of the 73 participants included in the study. Most participants were females (sex at birth; 76.7%) and identified as women (sex identity; 75.3%). The mean age of the participants was 54.36±12.05 years old (range: 30 to 79). The most reported specific pain condition was fibromyalgia (n=23, 31.5%). When looking at pain locations, most participants experienced back pain (76.7%) and 67 participants had multisite pain (91.8%). Twenty-seven participants presented moderate to severe psychological distress (38.0%), and most participants presented moderate to high levels of pain catastrophizing (n=51, 71.8%). A total of 62 participants (84.9%) were current cannabis users and 10 participants (13.70%) stopped using it in the last 12 months (1 missing value). The majority (82.2%) of participants used cannabis at least once a day in the previous 12 months.

TABLE 1.

Sample Demographics and Cannabis Use Characteristics

Sociodemographic profile	Participants (n=73), n (%)
Age (y), mean±SD	54.4±12.1
Range	(30-79)
Median	55
Sex at birth
Female	56 (76.7)
Male	17 (23.3)
Sex identity*
Woman	55 (75.3)
Man	17 (23.3)
Questioning	1 (1.4)
Country of birth
Canada	71 (97.3)
Other	2 (2.7)
Region of residence†
Remote	15 (20.8)
Nonremote	57 (79.2)
Employed (full-time or part-time)	27 (37.0)
Family annual income
Under $25,000	13 (17.8)
$25,000-$49,999	20 (27.4)
$50,000-$74,999	15 (20.6)
$75,000-$99,999	9 (12.3)
$100,000 and over	15 (20.6)
Prefers not to answer	1 (1.4)
Postsecondary education	58 (79.5)
Pain profile
Most common self-reported pain locations‡
Back	56 (76.7)
Legs	18 (24.7)
Shoulders	17 (23.3)
Neck	15 (20.5)
Knees	14 (19.2)
Generalized pain (yes vs. no)	27 (37.0)
Multisite pain (≥2 sites)	67 (91.8)
Most common self-reported pain conditions‡
Fibromyalgia	23 (31.5)
Osteoarthritis	18 (24.7)
Herniated disc	13 (17.8)
Pain duration (≥10 years)	59 (83.1)
Average pain intensity in the past 7 days (0 to 10 NRS), mean±SD
Range	5.2±1.7
Median	(1-10) 5
Recoded
Mild (1-4)	25 (35.2)
Moderate (5-7)	42 (59.2)
Severe (8-10)	4 (5.6)
Presence of neuropathic pain (DN4)	43 (58.9)
Pain interference (BPI score, 0-10), mean±SD	5.2±2.1
Catastrophizing (0-16)
Low: 0-5	20 (28.2)
Moderate: 6-8	16 (22.5)
High: 9-16	35 (49.3)
Psychological distress (PHQ-4)
None: 0-2	17 (23.9)
Mild: 3-5	27 (38.0)
Moderate: 6-8	16 (22.5)
Severe: 9-12	11 (15.5)
Cannabis use profile
Reasons for use‡
Pain	60 (96.8)
Sleep	49 (79.0)
Stress and anxiety	26 (41.9)
For pleasure	21 (33.9)
Mood	18 (29.0)
Appetite	9 (14.5)
Cannabis products used in the last 12 months‡
Liquid extracts or concentrates (eg, oil)	49 (67.1)
Dried flowers or leaves, buds	36 (49.3)
Pills and capsules	24 (32.9)
Edibles	19 (26.0)
Hashish (resin or pollen)	12 (16.4)
Beverages	5 (6.9)
Extracts or solid concentrates	2 (2.7)
Concentrations
THC-dominant products	13 (17.8)
CBD-dominant products	20 (27.4)
Balanced products	4 (5.5)
Several products with different combinations	33 (45.2)
Do not know	3 (4.1)
Frequency of use (number of times/day), mean±SD	2.0±1.6
Cannabis use before the age of 24 (yes vs. no)	41 (56.2)

0% missing values for all presented variables except for the region of residence (1 missing value).

BPI indicates Brief Pain Inventory; CBD, Cannabidiol; NRS, Numeric Rating Scale; PHQ-4, Patient Health Questionnaire; THC, Δ-9-thetrahydrocannabinol.

^*Sex identity choices: woman, man, nonbinary, genderqueer, transgender, 2-spirit, questioning, none of the above, other, and I prefer not to answer.

^†Six remote resource regions as defined by Revenu Quebec (ie, the provincial revenue agency) are: Bas-Saint-Laurent, Saguenay–Lac-Saint-Jean, Abitibi-Témiscamingue, Côte-Nord, Nord-du-Québec, Gaspésie–Îles-de-la-Madeleine. Nonremote regions are near a major urban center.

^‡Nonmutually exclusive categories.

Figure 1 compares the methods of cannabis use between the present study population living with chronic pain and the general population (2022 Quebec Cannabis Survey⁵⁵) that includes individuals with and without chronic pain (comparisons of the proportions without statistical testing). While smoked cannabis predominated in the general population (81.6% vs. 42.5% in our sample), people of the community were less likely to opt for oil-based cannabis products (29.5% vs. 67.1% in our sample).

FIGURE 1.

Methods of cannabis use. Oral drops and sprays include all oil-based products. Dabbing: inhalation with a hot knife or nail.

Cannabis Self-Medication for Pain Relief

Figure 2 presents the tree diagram showing the distribution of participants according to their user profiles (authorization for medical cannabis, legal nonmedical cannabis use, and guidance by a health care professional). Up to 61.6% (n=45) of our sample reported using cannabis for pain relief without the guidance of a HCP, that is, were self-medicating with cannabis (95% CI: 49.52-72.79). Among those, 40.0% (n=18) held a medical authorization. In the whole sample, this represented a total of 24.7% who reported holding an authorization for medical cannabis and being self-medicating.

FIGURE 2.

Tree diagram (conceptual map) of the possible uses of cannabis. Medical cannabis: cannabis used by individuals holding medical authorization under the Cannabis Act in Canada. Nonmedical cannabis: cannabis purchased from a legal recreational cannabis store. Professional guidance: expert advice, recommendations, insights, assistance from a health care professional.

None of the sex identity-stratified or age-stratified results yielded statistically significant differences. Sex identity-stratified results revealed no statistically significant differences between the proportion of women and men self-medicating (58.2% vs. 70.6%; P=0.284). As much as 60.00% of women in our sample held authorization for medical cannabis, whereas 41.2% of men did, but this difference was not statistically significant (χ² P=0.198). A similar proportion of women and men participants used legal nonmedical cannabis (65.5% vs. 64.7%; χ² P=0.917). Those proportions are not mutually exclusive as participants can use both. Participants aged 65 and over (vs. below 65) had similar profiles of self-medication with cannabis (63.2% vs. 61.1%; χ² P=0.875). When comparing older and younger participants, no statistically significant differences were found, that is, proportions using legal nonmedical cannabis were 73.7% vs. 63.0% (Fisher exact test P=0.290); proportions using medical cannabis were 47.4% vs. 57.4% (χ² P=0.450).

Table 2 shows the characteristics of participants depending on whether they were self-medicating with cannabis or using it under the guidance of a registered HCP. As compared with individuals not self-medicating, self-medicating participants were younger (t test P=0.027), more often started the use of cannabis before the age of 24 (χ² P=0.022), and more often used dried cannabis (χ² P=0.001). There was no significant difference found between participants who self-medicated and those who did not in terms of the frequency of moderate to severe psychological distress (χ² P=0.954) or high levels of pain catastrophizing (χ² P=0.179). Participants with mild pain were mostly self-medicating (68.0% vs. 32.0%), as were those with moderate (57.1% vs. 42.9%) and severe (100.0% vs. 0.0%) pain.

TABLE 2.

Sociodemographic Profiles According to Self-Medication Status

	Cannabis self-medication (n=45), n (%)	Using cannabis under professional guidance (n=28), n (%)	P
Age (y), mean±SD	52.22±12.08	57.79±11.40	0.027
Sex at birth
Female	33 (73.3)	23 (82.1)	0.284
Male	12 (26.7)	5 (17.9)
Sex identity*
Woman	32 (71.1)	23 (82.1)	0.478
Man	12 (26.7)	5 (17.9)
Questioning	1 (2.2)	0
Region of residence†
Remote	11 (24.4)	4 (14.8)	0.253
Nonremote	34 (75.6)	23 (85.2)
Employed (full-time or part-time)	20 (44.5)	7 (25.0)	0.094
Family annual income
Under $25,000	9 (20.0)	4 (14.3)	0.889
$25,000-$49,999	12 (26.7)	8 (28.6)
$50,000-$74,999	10 (22.2)	5 (17.9)
$75,000-$99,999	5 (11.1)	4 (14.3)
$100,000 and over	8 (17.8)	7 (25.0)
Prefers not to answer	1 (2.2)	0
Postsecondary education	36 (80.0)	22 (78.6)	0.883
Cannabis use before the age of 24 (yes vs. no)	30 (66.7)	11 (39.3)	0.022
Cannabis products used in the last 12 months‡	28 (62.2)	21 (75.0)	0.258
Liquid extracts or concentrates (eg, oil)	29 (64.4)	7 (25.0)	0.001
Dried flowers or leaves, buds Pills and capsules	13 (28.9)	11 (39.3)	0.358
Frequency of use (number of times/day)	1.92±1.67	2.13±1.57	0.343

^* Sex identity choices: woman, man, nonbinary, genderqueer, transgender, 2-spirited, questioning, none of the above, other, and I prefer not to answer.

^† One missing value. Remote resource regions as defined by Revenu Quebec (ie, the provincial revenue agency) are: Bas-Saint-Laurent, Saguenay–Lac-Saint-Jean, Abitibi- Témiscamingue, Côte-Nord, Nord-du-Québec, Gaspésie–Îles-de-la-Madeleine. Nonremote regions are near a major urban center.

^‡ Nonmutually exclusive categories.

Self-Perceived Cannabis Effectiveness and Safety

Figure 3 presents the self-perceived effectiveness of cannabis in reducing pain. These results show that 90% of the participants perceived cannabis to be at least slightly effective, and 49% perceived cannabis to be “very” or “extremely” effective. The 10 participants of 73 (13.7%) who stopped using cannabis it in the last 12 months reported various reasons for stopping including no effectiveness (n=7), adverse side effects (n=6), and too expensive (n=2).

FIGURE 3.

Self-perceived cannabis effectiveness in reducing pain. Cannabis refers to all cannabis-based products, including laboratory-synthesized cannabinoids (eg, nabilone).

Most participants (72.6%) estimated that cannabis use posed no or minimal health risk, whereas 27.4% believed it presented a moderate to high health risk; 86.3% of participants believed that cannabis use could lead to an addiction problem, 5.5% did not know, and 8.2% believed that it could not lead to addiction. When compared with opioids, 83.6% of participants felt that cannabis was safer, and when compared with other illicit drugs, 100% they also felt that cannabis was safer. When comparing self-medicating and nonself-medicating participants, 37.8% versus 10.7% perceived cannabis to pose moderate to high health risk (Fisher exact test P=0.10) and 93.3% versus 85.7% perceived cannabis to be effective in some way in reducing pain (Fisher exact test P=0.249).

DISCUSSION

The present descriptive study described cannabis self-medication for pain relief in people living with chronic pain. Their perceptions toward the effectiveness and safety of cannabis were also explored. To our knowledge, this is the first postlegalization study to investigate cannabis self-medication in people living with chronic pain in Canada that covers both medical and nonmedical users. Positive perceptions regarding the effectiveness and safety of cannabis, as well as the widespread practice of self-medication (61.6%), should be a matter of concern. Surprising pathways to self-medication were revealed.

Our sample has demonstrated similar characteristics (age, postsecondary education, pain intensity, presence of neuropathic pain, reasons for cannabis use, and concentrations) when compared with other chronic pain populations studies,^28,66–69 except regarding the proportion of female participants, pain duration, and the methods of cannabis use. The proportion of female participants in the current study (77%) was observed to be higher than that reported in chronic pain random samples (55% to 65%).⁵¹ When compared with studies on cannabis users, this difference is even more important with proportions of females ranging from 45% to 55%.²⁸ First, this could be explained by the COPE Cohort⁵¹ and the present spin-off study web-based recruitment methods. Given that females are more likely to work in an online environment⁷⁰ and use social media,⁷¹ this could have influenced the proportion of females in the present sample (which is greater than random samples of individuals living with chronic pain). It is probably not sex differences in cannabis use that explain this trend in our sample, as the prevalence of cannabis use among women and men living with chronic pain was found to be similar.²⁵ This over-representation of women thus calls for the stratification of prevalence estimates to determine if results are genuinely extrapolatable. In our case, no statistically significant difference was found between women and men regarding the prevalence of self-medication, minimizing the possibility of sampling bias. About duration of pain, the proportion of participants living with pain ≥10 years was higher (83.1%) in this study than what is found in other random and nonrandom samples of people living with chronic pain (47% to 53%^51,66,67,72). We could hypothesize that cannabis is not the first treatment attempted by patients, which is consistent with other findings.³³ In terms of methods of use, sprays and drops were preferred by people living with chronic pain in our sample, a trend markedly less prevalent within the general population using cannabis.^54,55 In other studies involving medical and nonmedical cannabis users with chronic pain, sprays and drops methods were less prevalent (smoked cannabis was preferred).²⁸ These differences between the present sample and the general population or previous studies in individuals with chronic pain could possibly be explained by the proportion of participants (55%) using medical cannabis. Two recent studies revealed that oil-based cannabis products were the most commonly used among individuals authorized for medical cannabis.^73,74 All things considered, these differences underscore the importance of specific studies on the reality of people living with chronic pain to guide the establishment of policies, preventive efforts, and interventions promoting optimal use. This subgroup is present within the general population (1 of 5 individuals lives with chronic pain), but our study suggests that it does not present the same profile as found in large federal and provincial surveys (eg, less pulmonary risk associated with smoked cannabis).

Cannabis Self-Medication for Pain Relief

Our study revealed that 61.6% of individuals living with chronic pain and using cannabis are self-medicating. Legalization of nonmedical (recreational) cannabis in Canada in 2018 has facilitated access to cannabis, opening the door to self-medication. Before the legalization (when considering medical cannabis only), 15% of people living with chronic pain in Canada self-medicated with cannabis.²⁶ To the best of our knowledge, no study has yet estimated this prevalence among individuals living with chronic pain since the legalization. A study on self-medication with cannabis (for any health problems) conducted in Quebec reported that 53% of the participants (n=489) self-medicated with cannabis for reducing pain.⁴⁵

An important proportion of individuals living with chronic pain and using cannabis reported using both medical and legal nonmedical cannabis (21%). Similar results showing an overlap between medical and nonmedical cannabis have been reported in a previous Canadian study, where ∼80% of the medical cannabis users were also using nonmedical cannabis.⁷⁵ This poses a challenge for the therapeutic application of cannabis and its medical oversight,⁷⁶ as people who use both medical and nonmedical cannabis are more likely to exhibit cannabis use problems.⁷⁵ As compared with medical cannabis, using nonmedical cannabis is associated with various disadvantages and risks such as uncertainty about strains, doses, and purity, risk of using more cannabis as well as the lack of insurance coverage.^43,44,75 Although it was expected that using medical cannabis would have the advantage of receiving professional guidance, interestingly, participants in the present study who held medical cannabis authorization were not necessarily accompanied or supervised by a HCP. This highlights one of the most important limitations of the Cannabis Act in Canada,⁴¹ in that it does not ensure that patients consuming cannabis for medical purposes will receive support from a HCP, as the law does not mandate this medical guidance.⁴² Furthermore, HCPs who would be interested in getting involved in follow-up care currently have limited guidelines to rely upon.⁷⁷ Globally, in a real-life context, medical cannabis and legal nonmedical cannabis are interconnected since a large proportion of patients utilize both.⁷⁵ The introduction of the Cannabis Act has enabled the procurement of products from diverse sources for the treatment of medical conditions according to the patients’ preferences (eg, patients perceive legal nonmedical products as being more cost-effective).⁷⁸ In addition to this reality, there is the presence of self-medication, which complicates the consumption profiles of users and exposes them to various health risks (eg, dosing challenges, no evaluation of potential interaction with medications, no support regarding occupational hazards).^43,44 This raises concerns regarding the risks of self-medication and highlights the need for medical supervision related to cannabis use. We should note that those are potential risks as we did not find any evidence comparing the actual risks between patients self-medicating and those who are not. Moreover, in the present study, all 4 participants who reported severe pain were self-medicating. Globally, our descriptive study represents a first attempt to describe self-medication practices, but further studies should be conducted at this level.

Different unexpected pathways to self-medication were revealed in the present study. This opens the door to a variety of new research questions that could be explored, for example, through qualitative studies: What motivates individuals to self-medicate? What are the unmet needs that could lead to self-medications? Why do some holders of medical cannabis authorization lack the follow-up support? How do HCPs operationalize support for nonmedical cannabis use? How do illegal sources of nonmedical cannabis come into play? While frameworks exist for self-medication with over-the-counter medications,³⁸ new frameworks should be established to guide research specifically on cannabis use among people living with chronic pain.

Self-Perceived Cannabis Effectiveness and Safety

The majority of participants felt that cannabis was effective in some ways in alleviating pain. It is worth mentioning that this assessment reflects the overall impression of pain improvement felt by participants, rather than a quantifiable measure of pain intensity before and after cannabis use. These results are consistent with previous findings showing that 78% of patients using cannabis to alleviate their pain perceived a moderate improvement in their condition.²⁶ From a qualitative perspective, the literature suggests that patients often report a sense of relief when using cannabis, even if a decrease in pain intensity is not observed on the visual analog scale or other measurement tools⁷⁹ underlining the importance of diversification of outcome measures in cannabis efficacy and effectiveness studies.

As for the perceived safety of cannabis, participants in this study felt somewhat confident, with less than a third of them reporting a moderate to high health risk linked to its utilization. This proportion is notably small compared with the general populations, where three-quarter of individuals perceive cannabis as posing a moderate to high health risk.^54,55 Cannabis may mitigate experiences of pain and benefit mental health for those with chronic pain,⁸⁰ without causing related cannabis problems.⁸¹ Han et al⁸² also suggested that the risk perception among patients with disabling illnesses may decrease considering changing priorities in the setting of chronic or advanced disease.⁸² However, the large majority of participants in the present study (86%) believed that cannabis use presented a risk of addiction. This proportion is higher than what was found in a recent study from Goodman and Hammond,⁸³ where 51% to 62% of the participants perceived that cannabis use presented a risk of addiction. In that study, which included participants from Canada and the United States, the Canadian participants had a higher perception of risk compared with their American counterparts. This disparity among Canadian participants and the participants in our study could be attributed to the numerous cannabis consumption awareness campaigns broadcasted in Canada,^84,85 particularly in Quebec.⁸⁶ In the present study, it was surprising to see that so many people perceive this risk but still engage in self-medication with cannabis. In fact, when comparing self-medicating and nonself-medicating participants, those who self-medicated more often perceived cannabis to pose a moderate to high health risk. Florimbio et al⁸⁷ suggest that the risk perception vary by cannabis consumption patterns and method. Therefore, the self-medicating participants in our study were more likely to use dried flower and smoking methods, which may have influenced their risk perception. Another hypothesis is that individuals self-medicating with cannabis recognize their limited knowledge of dosage, interactions with medications, and overall safety. Without health care guidance, reliance on anecdotal information may heighten concerns about potential adverse effects and health risks. In addition, the lack of professional oversight may lead to uncertainties about the quality and purity of cannabis products, further increasing perceived health risks among self-medicating users. Such hypotheses should be further explored in future studies.

Strengths and Limitations

This study has several strengths, including the utilization of telephone interviews, a standardized questionnaire, and questions derived from existing surveys and recognized validated measurement scales. The study incorporated a community sample of participants from 16 of the 17 administrative regions of the province of Quebec, ensuring diversity, particularly among individuals and regions that are often underrepresented in studies where recruitment is conducted in university-affiliated clinics. However, some limitations are worth mentioning. Much like all studies involving cannabis users, individuals who are well informed about or have a positive attitude or tolerance toward cannabis might have been more inclined to participate in the study, thereby creating the potential for selection bias. While telephone interviews led to a number of strengths, they were not anonymous (possibility of social desirability bias). In fact, cannabis still carries an illicit connotation, even though it has been legal in Canada for 5 years.⁵⁵ The present descriptive study is, however, a first step toward understanding self-medication among individuals living with chronic pain who use cannabis. Future studies with larger sample sizes and different designs (eg, case-control studies, observational studies) will be needed to deepen the understanding of the biopsychosocial determinants of self-medication using multivariable analyses. One should also keep in mind that generalizing results is difficult outside the legislative context in Canada regarding cannabis use. Finally, this study is also limited as it focuses on medical and legal nonmedical cannabis use. Nonmedical cannabis obtained from illegal sources should be the focus of further investigations.

CONCLUSION

Our study showed that among adults living with chronic pain and reporting cannabis use, 6 of 10 (61.6%) endorsed self-medicating with cannabis, and many of them did so despite having authorization for medical cannabis. Unexpected pathways to self-medication have been uncovered and deserve further exploration through more extensive research. Moreover, 1 participant of 5 used both medical and legal nonmedical cannabis. Since cannabis research is often organized around medical versus legal nonmedical cannabis, it is appropriate to say that these 2 silos need to be broken, considering that patients can use cannabis obtained through recreational channels for therapeutic purposes. This descriptive study showed that in the real world, medical and legal nonmedical cannabis use are 2 connected vessels. This highlights the urgent need for HCPs to become more involved in the supervision of cannabis use to mitigate patient risks. Interested parties, including governments, researchers, HCPs, and funding agencies, need to take this into account. In a context where several countries are considering loosening their laws regarding the use of cannabis, our results are timely and will contribute to the scientific knowledge.