Gender Differences in Medical Cannabis Use: Symptoms Treated, Physician Support for Use, and Prescription Medication Discontinuation

Introduction

As legalization and utilization of medical cannabis (MC) continues to expand in the United States, a small but growing body of evidence supports the use of cannabis and cannabinoids for their analgesic, anticonvulsant, and anti-inflammatory properties in the treatment of a range of chronic conditions.^1,2 The integration of MC into health care delivery systems in the United States continues to be stymied by its federal legal status and disagreement over its usage.³

In the absence of established clinical guidelines that stem from the paucity of clinical trial testing of MC products on the market, the adoption of MC as a therapy to manage symptoms associated with various chronic conditions continues to proliferate as an alternative or complementary method vis-à-vis prescription medications.^4–6 To date, gender-related differences in MC use are not widely reported, and little is known about physicians' support of patients' use of MC to address symptoms associated with chronic conditions.

There are well-established differences in health care utilization between women and men.⁷ Previous research indicates that, compared with men, women are more likely to have a general practitioner, contact their general practitioner and specialists, and obtain a higher number of outpatient consultations.^8,9 Women have higher utilization rates for physical, mental, and emergency health services than their male counterparts,^10,11 as well as higher rates of unmet health care needs.^9,11 Women also utilize preventative care more than men,¹² and women exhibit higher utilization rates of mental health services than men even when controlling for prevalence of mental health disorders.¹³

Women's adoption of complimentary or alternative medicine (CAM) also differs from men.¹⁴ Women access CAM services more often than men,^15–17 and pain is a leading motive for patients to seek CAM.¹⁸ Specifically, women also report using CAM more often than men for headaches and migraines, and preventative services.^17,19 Among the U.S. cancer population, women are significantly more likely to use CAM, particularly for pain, depression, and insomnia.¹⁶ Physician support for patients' use of CAM has been investigated chiefly among cancer physicians, and previous research has found support among oncologists.²⁰ A systematic review of cancer studies and CAM found that significant numbers of cancer patients do not believe that their physicians were aware of their CAM use, but that patient–provider discussion of CAM use enhanced patients' satisfaction with care.²¹

Gender differences within some chronic conditions and symptomology may be particularly relevant to the utilization of MC. A meta-analysis of gender differences in pain indicated that women tended to have higher prevalence of neuropathic, postprocedural, and musculoskeletal pain, specifically osteoarthrosis, abdominal, back, and headache/migraine pain.²² Research also indicates that, for women, depression is associated with more intense activity-related pain than for men.²³ Hunt et al. found that despite the increased prevalence of back pain and higher health care utilization by women, they were equally likely (not more likely) to consult for back pain, but they were more likely to consult for headache-related pain.²⁴

Women may be more likely to be dual diagnosed with an anxiety disorder, bulimia nervosa, or major depressive disorder and, therefore, may have more significant and disabling disease burden.²⁵ Symptoms such as pain, anxiety, depression, and insomnia often interact and may mutually reinforce one another,^26–28 and recent findings suggest that persons using MC may perceive greater efficacy in treating co-occurring symptoms.²⁹

Across jurisdictions where MC is legal, research assessing provider attitudes toward and support for MC use has produced mixed findings. In two U.S. states with relatively recent MC legalization, providers have expressed support for its treatment of selected conditions, yet a lack of knowledge about MC's broader applications. A recent survey in Minnesota reported that a majority of primary care providers expressed support for MC use as an adjunct for cancer patients and in the treatment of terminal illness and intractable pain, but most did not know of MC efficacy in treating the symptomology of other qualifying conditions in the state's MC program.³⁰

In New York, a majority of physicians reported willingness to discuss MC with patients and supported its use for the treatment of neuropathy or pain, but were equivocal on their professed knowledge of the endocannabinoid system.³¹ By contrast, a 2013 survey of general practitioners in Colorado (where MC had been legalized in 2000) found only a small minority supporting MC use.³² Research from Israel, where MC was made available in 1990, has found physicians currently split in their attitudes toward MC: some perceive its potential to treat conditions while acknowledging the limits of conventional medicine, whereas others view cannabis as incompatible with biomedicine and with the potential for dependency.³³ Patients in Canada have reported different levels of communication regarding MC with physicians depending on their conditions, as patients with arthritis and HIV exhibit greater likelihood of discussing MC with their physician than do patients with depression or anxiety.³⁴

Gendered patterns of cannabis use in the United States includes similar prevalence of recreational cannabis use in adolescence among males and females that increases among males during young adulthood.^35,36 This trend appears to persist even after MC laws have been enacted.³⁷ Among MC users, although men are more likely to be experienced cannabis users before initiation of MC,³⁸ women are more likely to report MC use for treatment of nausea, anxiety, and migraines than men.³⁹ Additional gender-based differences exist for prescription drug use, as women are more likely than men to use antidepressants, antianxiety, and pain medications,^40–42 despite there being no differences in the prevalence of mental health conditions by gender.¹³

With MC emerging as a possible CAM option for treating a range of symptoms, it is important to better understand correlates of patients' decision making in using MC and discontinuing prescription medications, yet little research has examined gender-related differences in MC use vis-à-vis prescription medications. In this exploratory study of MC users in Illinois, we sought to understand gender differences in MC utilization, symptoms treated by MC, physician support for patients' MC use, and discontinuation of prescription medication use subsequent to MC adoption.

The Compassionate Use of Medical Cannabis Program Act (410 ILCS 130) currently allows a person with one of 52 qualifying conditions with written certification from a recommending physician to obtain an MC registry card. Registered participants in the program may purchase MC from a chosen dispensary within the state and are not required to receive prescriptions from physicians.⁴³

Materials and Methods

Procedures

We conducted a cross-sectional online survey of persons with Illinois state MC cards recruited from licensed MC dispensaries across the state. Participants were recruited through flyers mailed to dispensaries located in Illinois and through online networks. Persons interested in participating went to the study screening URL to determine eligibility. Inclusion criteria: (1) registered MC user in Illinois, (2) self-reported qualifying condition for MC use in Illinois, (3) age ≥18 years, and (4) current (past month) cannabis use. A total of 22 respondents were determined ineligible at screening. Eligible participants proceeded to the study consent URL and completed an online informed consent process. Those who consented to participate proceeded to the online survey.

Measures

Survey questions included self-reported participant demographic information (e.g., current gender, race/ethnicity, age, and employment status), conditions that qualified participants for the state MC program (e.g., severe fibromyalgia, cancer, and multiple sclerosis), symptoms treated with MC (e.g., pain, spasticity, nausea, and anxiety), and cannabis use before qualification for MC use (yes/no). For those reporting prior use, we asked if cannabis use had increased or decreased after qualification for MC use. We assessed discontinuation and reduction of prescription medications by having participants report if they had ever used MC to discontinue prescription medications, or reduce but not discontinue prescription medications, respectively. Perceived support from primary care providers and specialists was rated using a 4-point scale (1 = very unsupportive, 2 = somewhat unsupportive, 3 = somewhat supportive, 4 = very supportive). We asked participants which type of physician provided certification for their qualifying condition: primary care provider, specialist, or another physician willing to provide certification (e.g., “medical cannabis practice”).

Data collection

The survey was administered online July–August 2017 using Qualtrics survey software. All data were collected anonymously and could not be linked back to individual participants. To avoid multiple survey completions by an individual, Internet protocol (IP) addresses were temporarily stored during data collection to exclude individuals who had already taken the survey. Twelve repeat respondents were deleted for a final sample of 367 respondents. For this article, we analyzed data from 201 women and 160 men (n = 361); four participants were excluded as they did not report their current gender, and two participants identifying as transgender were not included as they constituted <1% of the final sample. All IP addresses were deleted from the database when the survey was closed. Password-protected data were stored on protected servers. Upon survey completion, participants were sent by e-mail a $20 electronic gift certificate. The study was approved by the Institutional Review Board of DePaul University.

Data analysis

Statistical analyses proceeded in two steps. First, we compared participants' qualifying conditions, symptoms treated with MC, perceived physician support for MC use, use of MC, and prescription medications by participant gender using chi square for analysis of frequencies on the nominal variables and t-tests for scores rating perceived support from provider for MC use. Then, we included gender as a covariate with other variables of interest into logistic regression models predicting discontinuation of prescription medications. Using logistical regression, we tested the relationship of the independent variables (1) gender, (2) physician support for MC use, and (3) number of symptoms treated with MC as predictive of (4) discontinuation of prescription medications.

Results

Participants identified as white (82.5%), Latinx/Hispanic (8.9%), black/African American (3.9%), and multiracial/other (4.7%). Almost one third reported being employed full time (31.8%), with the remaining participants employed part time (21.5%), disabled (25.4%), unemployed (16.8%), or retired (4.5%). Severe fibromyalgia (27.75) and post-traumatic stress disorder (21.3%) were the most commonly reported qualifying conditions, followed by rheumatoid arthritis (8.9%), spinal cord injury (8.9%), cancer (8.0%), and multiple sclerosis (7.5%). Pain (75.1%) was the most frequently reported symptom treated by MC, followed by anxiety (65.4%), inflammation (59.6%), and insomnia (56.2%). Our sample broadly aligned with demographics of the larger population of MC users in Illinois at the time of our data collection (52% women; 48% men) and in terms of the most frequently cited conditions for qualifications for the program, although our sample had proportionally higher percentages of rheumatoid arthritis and lower percentages of cancer.⁴³

In the bivariate analyses, women were significantly more likely to report using MC for pain (X² = 13.69, p < 0.001), anxiety (X² = 6.67, p < 0.01), inflammation (X² = 27.50, p < 0.001), and nausea (X² = 8.33, p < 0.01). There were no significant gender-related differences in MC use in the treatment of depression, insomnia, or muscle spasms. Regarding support, we found that men report marginally higher levels of support for MC use from primary care physicians and significantly higher levels of support for MC use from specialist physicians (F = 4.54, p < 0.05).

Women more frequently received qualifying documentation for an MC card from an MC practice (X² = 4.51, p < 0.05) compared with men. Among participants who reported prior cannabis use, women were significantly more likely to increase use of cannabis after acquiring an MC card (X² = 11.04, p < 0.05). Women were proportionately more likely to report reductions in prescription medications through MC use (X² = 3.87, p < 0.05), and discontinuation of prescription medications through MC use (X² = 13.65, p < 0.001) (Table 1).

Table 2 presents the results of the multivariable analyses that indicate being a woman (odds ratio [OR] = 2.09, 95% confidence interval [CI]: 1.11–3.90), number of symptoms treated with MC (OR = 1.42, 95% CI: 1.23–1.65), and reporting higher levels of support for MC use from a primary care physician (OR = 1.45, 95% CI: 1.10–2.03) increased the likelihood of discontinuing prescription medication through MC use.

Given the relatively large number of participants and disproportion numbers of women qualifying for MC use with fibromyalgia, we conducted a supplementary analysis excluding fibromyalgia patients. The supplemental adjusted model excluding fibromyalgia patients (n = 261) resulted in similar parameters as the full model: woman (OR = 2.07, 95% CI: 1.01–4.25), perceived support from primary care provider (OR = 1.77, 95% CI: 1.25–2.51), and number of symptoms treated (OR = 1.45, 95% CI: 1.10–1.91) increased the likelihood of discontinuing prescription medication through MC use.

Discussion

The results from our cross-sectional study describe a number of gender-associated patterns within the use and outcomes of MC among patients with chronic conditions. Women appear to be more likely than men to use MC for a range of symptoms (specifically, pain, anxiety, inflammation, and nausea), to have increased use of cannabis since qualifying for MC, and to subsequently have reduced or completely discontinued their prescription medications.

In addition, the women in our sample reported marginally lower levels of support from their primary care provider, and significantly less support from specialist physicians than the men in our sample, and significantly more of them received certification for their state MC card from MC practices. What is noteworthy for health care providers is that a majority of women in this study successfully integrated MC into their chronic condition management and were able to reduce or discontinue prescription medication, despite reporting (as a group) only moderate support from primary care practitioners and specialists.

A closer examination of the impact of MC on patients' medications, using multivariable analysis revealed that being female, treating more symptoms with MC, and having a higher level of support from a primary care physician is significantly associated with discontinuation of prescription medication use. Our results align with recent findings that report men are more likely to be more experienced users of cannabis before onset of medical use,³⁸ but women are more likely to substitute MC for prescription medications.⁴⁴ Population-based studies have shown that men are more likely to report higher levels of cannabis use from middle adolescence through their mid-30s.^35,36 Utilization studies have found that although MC was predominately used by men, among states with relatively early legalization that gender difference has narrowed over time.^45,46 In our study, male participants were more likely to report use before qualification for MC, whereas among prior users female participants were more likely to report their use of cannabis to increase after initiation of MC. Such differences align not only with the epidemiology of recreational and MC in the United States, but also have implications for how men and women may approach the use of MC and recreational cannabis differently.

Whereas cannabis use was once viewed largely as illicit drug use, the advancement of MC legalization across states in the United States and other international jurisdictions may be shifting attitudes toward the complementary and alternative medical potential of cannabis.

Given that previous research indicates that women engage in CAM services more frequently than men^15,16 the lack of a requirement for a prescription to acquire MC in Illinois may also contribute to it being viewed by many patients as an alternative/complementary treatment for chronic conditions. The implication that women may conceptualize MC as CAM merits further research, as this finding may be viewed as a consequence of patient experience with cannabis, and not only of shifting public attitudes toward it.

Men and women may experience pharmacological effects of MC use differently, as research has shown gender differences in analgesic effects among cannabinoids across animal and human studies.⁴⁷ In addition, pharmacological research has identified drug-metabolizing enzymes and drug transporters that may be inhibited by certain compounds of MC.⁴⁸ More patient-centered studies on MC are needed to better understand differences in dosing, outcomes, beliefs, attitudes, formulations, pharmacology, and metabolism between men and women.

MC users report a range of motivations for discontinuation of prescription medications, including concerns regarding addiction and toxicities, as well as better management of symptoms and side effects through MC use.^4,6,49 To the extent that such reduced dependence on prescription medication is associated with greater symptom management and higher quality of life, one implication of this study is that primary care practitioners could encourage patients who use MC to track their usage and openly discuss their experiences during office visits. Although it is possible that providers are discontinuing prescriptions after noticing reductions in usage, discontinuation of prescription medications coupled with neutral attitudes of primary care physicians toward MC may be perceived as support for continued MC usage.

Furthermore, given findings from previous studies,³³ there may be differences in provider support for discontinuing prescriptions depending on medication and patient condition. We did not ask participants to characterize physician responses to the lack of their prescription refills, but the interaction between patients and providers in this circumstance warrants further research. We also did not ask participants their provider's gender, but research has shown that gender concordant patient–physician dyads result in higher levels of patient-centered care.^50,51 More detailed data that elucidate decision making between patient and provider would deepen our understanding of how patients and providers conceptualize MC use as alternative, complementary, or integrative health care.

Several characteristics of our sample should be noted in relation to interpreting our results. First, because the participants were recruited through MC dispensaries, we obtained results from patients who almost certainly have experienced success in accessing MC and in using it to manage a range of symptoms related to their chronic conditions. We do not know how the results of the study might be different if we had accessed large numbers of patients who have tried MC but found it ineffective compared with their prescription medication, or patients who are using nonmedical-grade cannabis for therapeutic purposes. Nonetheless, given the growing availability of MC as a therapeutic option for a range of chronic conditions, we believe these results add useful details to the growing picture of patient experiences with MC.

Our study has several limitations that merit consideration. The cross-sectional design precludes any causal assumptions among the correlates of prescription medication discontinuation. All data were self-reported and potentially subject to recall bias. We did not utilize a probability sample, although our participants proportionately resemble the approved MC users in Illinois in terms of qualifying conditions and gender at the time our study was conducted.⁴³ Despite these limitations, our study provides important data on gender-related differences in MC utilization, symptoms treated, and perceived physician support for use. Future research on gender differences in this population may benefit from more detailed and nuanced data related to symptomology, utilization, dosing, and outcomes associated with MC, and interactions with physicians and others in the health care system to further extend these findings.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

Support for this study was provided through the Provost's Collaborative Research Fellowship, DePaul University.