ABSTRACT
Purpose
Sexual minority (SM; vs. heterosexual) young adults (YAs) report disproportionately high cannabis use rates. SMYAs may use medical cannabis for unaddressed healthcare needs, especially in rural areas, which may have greater anti‐SM stigma and less healthcare access.
Methods
SMYAs (N = 549; ages 18–25, M age = 21.75 [SD = 2.11]; 66.7% female; 45.5% gender minority; 32.6% rural‐residing; 44.6% racial and/or ethnic minority) in Oklahoma and surrounding states completed two waves of surveys (2023–2024). Regression‐based models examined: (1) direct associations of rural vs. urban residence with healthcare access (i.e., unmet healthcare needs, healthcare availability) and follow‐up cannabis use outcomes (i.e., any past‐month cannabis use, medical cannabis use); (2) direct associations of healthcare access with cannabis use outcomes; and (3) indirect associations of rural vs. urban residence with cannabis use outcomes through healthcare access. Models controlled for participant sociodemographics.
Findings
Rural (vs. urban) residence was directly associated with greater likelihood of having unmet healthcare needs and less healthcare availability. Unmet healthcare needs were associated with greater likelihood of any past‐month cannabis use and medical cannabis use, whereas greater healthcare availability was associated with less likelihood of any past‐month cannabis use and medical cannabis use. Rural (vs. urban) residence was indirectly associated with greater likelihood of medical cannabis use through greater likelihood of having unmet healthcare needs and having less healthcare availability.
Conclusions
Cannabis use among SMYAs may be a response to unaddressed medical needs, especially in rural areas. Public health efforts should prioritize improving healthcare infrastructure in rural areas and creating affirming, accessible healthcare environments for SMYAs.
Keywords: cannabis, medical cannabis, rurality, sexual minority, unmet healthcare needs
1. Introduction
Sexual minority young adults (SMYAs), defined as individuals who identify as lesbian, gay, bisexual, or another sexual orientation outside of heterosexual [1], face significant unmet healthcare needs compared to their heterosexual peers [2]. In the United States, studies indicate that SM adults experience barriers to accessing adequate healthcare approximately 2–3 times more frequently than heterosexual adults [3]. This problem is further exacerbated by rural–urban healthcare disparities, where rural communities often face reduced access to healthcare providers, limited availability of specialists, and a shortage of healthcare facilities [4]. As a result, unmet healthcare needs can prompt individuals to seek alternative forms of care, such as medical cannabis use [5]. Medical cannabis may be widely available in rural areas, even when other healthcare is not [6]. While medical cannabis may offer benefits for certain conditions, its use can also be harmful in the absence of appropriate medical guidance, creating both potential benefits and risks [7]. This makes the intersection of sexual minority identity, healthcare access, and the use of medical cannabis an urgent public health issue that requires further investigation.
Minority stress theory provides a useful framework for understanding why unmet healthcare needs are particularly prevalent among SMYAs. According to this theory, individuals who belong to marginalized groups, such as sexual minorities, experience unique stressors related to stigma, discrimination, and social exclusion, which can have detrimental effects on both mental and physical health [8, 9]. These stressors, whether in daily life or within a healthcare setting, can contribute to negative health outcomes and may create barriers to receiving equitable healthcare. SM individuals may also avoid or delay care due to anticipated discrimination or previous negative experiences [10]. Consequently, the impact of minority stress on healthcare access contributes to the unmet healthcare needs of SMYAs.
Rural SMYAs may be particularly vulnerable to unmet healthcare needs due to several compounding factors. Rural areas often suffer from a lack of healthcare infrastructure, including limited access to physicians and specialists [11]. Financial barriers are also more prevalent in rural regions, where lower incomes and limited insurance coverage may make healthcare unaffordable [11]. Furthermore, transportation challenges often prevent rural residents from traveling to needed healthcare services, making it even more difficult for SM individuals in these areas to access appropriate care [11]. Additionally, rural areas may experience heightened levels of anti‐SM stigma, which can discourage SM individuals from seeking care and further contribute to healthcare disparities [12].
As unmet healthcare needs continue to affect SMYAs, some may turn to medical cannabis as an alternative to traditional treatments. In rural environments, cannabis dispensaries can be more accessible than standard healthcare facilities [6]. While cannabis is increasingly being prescribed for a variety of medical conditions, the lack of physician oversight and clear instructions regarding its use can lead to riskier practices and adverse health outcomes [7]. Given these uncertainties, understanding patterns of medical cannabis use among individuals with unmet healthcare needs—especially among populations who may experience stigma in traditional healthcare settings, like rural SMYAs—is critical.
This study aims to explore the relationship between rurality, unmet healthcare needs, and medical cannabis use among SMYAs. Specifically, we seek to examine whether residing in rural (vs. urban) areas and unmet healthcare needs are associated with cannabis use (i.e., any cannabis use, medical cannabis use) among SMYAs and whether residing in rural (vs. urban) areas is associated with cannabis use through unmet healthcare needs as a mediator. By addressing these gaps in research, we hope to inform future public health interventions targeting sexual minority populations, particularly those residing in rural areas, to reduce healthcare disparities and improve overall health outcomes.
2. Methods
2.1. Participants and Procedures
The current study analyzed 2 waves of survey data among YAs (aged 18–25) who report an SM identity (i.e., gay, lesbian, bisexual, pansexual, queer, or another non‐heterosexual identity) and reside in Oklahoma (OK) or surrounding states (i.e., Arkansas [AR], Colorado [CO], Kansas [KS], Louisiana [LA], Mississippi [MS], Missouri [MO], New Mexico [NM], Texas [TX]). This study launched in June 2023–February 2024, involved 2 survey assessments 6 months apart, and was approved by the University of Oklahoma Health Campus Institutional Review Board.
Study ads were posted on social media (e.g., Facebook, Instagram, YouTube) and targeted individuals ages 18–25, residing in eligible states, and identifying as SM. Purposive, quota‐based sampling ensured variability in rural versus urban residence (∼30% rural), tobacco use (∼50% reporting past‐month use), and race and ethnicity (≥40% identifying as racial and/or ethnic minority). After clicking on an ad, interested individuals were directed to a webpage with a study description and screening questions assessing their age, zip code, sexual identity, race, ethnicity, and current tobacco use status. Those deemed preliminarily eligible were sent a link via email to the full study description and consent form in Qualtrics. They then completed a screener in Qualtrics to confirm eligibility and were administered the baseline survey, which took ∼20 min to complete. After completing the survey, participants were emailed a $10 Amazon e‐gift card incentive. Fraud prevention efforts included withholding details of eligibility criteria before screening, examining data validity (e.g., duplicate internet provider addresses, e‐mail addresses, phone numbers; illogical responses; survey completion time), and confirming validity of contact information before providing incentives [13, 14].
Of the 8113 individuals who clicked on ads, 1631 (20.1%) completed the pre‐screening and 1148 (70.4% of those who completed the pre‐screening) were preliminarily eligible and were provided a screener and survey link. Of the 728 (63.4%) who consented, 689 were eligible (94.6%). Ineligibility reasons included falling outside of the eligible age range (n = 15), not residing in one of the 9 states of interest (n = 9), and reporting a heterosexual identity (n = 9). Additionally, 6 responses were deemed duplicates. The W1 survey was completed by 616 participants (89.4%) and partially completed by 73 participants (10.6%). The current analyses use baseline survey data assessing sociodemographic factors, unmet healthcare needs, and healthcare availability, as well as 6‐month follow‐up survey data assessing cannabis use (n = 549, 89.1% retention).
2.2. Measures
2.2.1. Healthcare Access
Unmet healthcare needs were assessed at baseline with 2 items adapted from the National Survey on Drug Use and Health: “In the past 12 months was there ever a time you needed [to see a medical specialist about a health issue/mental health treatment or counseling], but did not get it?” (yes/no) [15]. Participants indicating yes to either item were coded as reporting unmet healthcare needs. Healthcare availability was assessed at baseline by asking participants to rate the availability of healthcare services in their community on a 5‐point scale (1 = Poor to 5 = Excellent) [16].
2.2.2. Rural–Urban Residence
Rural–urban residence was assessed at baseline by asking participants “What is your current ZIP Code of residence (i.e., ZIP Code where you currently live; If you attend college and live on campus, please provide your campus ZIP Code.)?” The Department of Agriculture Rural–Urban Commuting Area (RUCA) codes were used to dichotomize participants’ self‐reported ZIP Codes as “rural” (RUCA code of micropolitan, small town, or rural) or “urban” (RUCA code of metropolitan) based on RUCA coding recommendations [17, 18].
2.2.3. Cannabis Use Outcomes
At follow‐up, participants indicated the number of days they used cannabis in the past month (0–30 days). Days of past‐month cannabis use were dichotomized as no use (0 days) or past‐month use (1–30 days) based on a positively skewed distribution. Participants who reported past‐month cannabis use were then asked to indicate whether they use cannabis for medical purposes, recreational purposes, or both. Those who indicated using cannabis for medical purposes or medical and recreational purposes were coded as reporting past‐month medical cannabis use; those reporting no past‐month cannabis use or using cannabis for only recreational purposes were coded as not reporting past‐month medical cannabis use.
2.2.4. Sociodemographic Covariates
Covariates included participant age (continuous), sex (female, male, another sex), gender identity (cisgender [cisgender man, cisgender woman], gender minority [transgender, gender nonbinary]), sexual identity (monosexual [gay, lesbian], bisexual+ [bisexual, pansexual], another sexual identity [asexual, queer]), race and ethnicity (non‐Hispanic [NH] White, racial and/or ethnic minority [Black, Asian, Native American or American Indian, multiple races, Hispanic]), education (high school or less, some college, Bachelor's degree or higher), and state cannabis legalization at the time of baseline data collection (living in a state with legalized cannabis for medical purposes [OK, AR, KS, LA, MS, TX] vs. living in a state with legalized cannabis for medical and recreational purposes [CO, TX]) [19].
2.3. Data Analysis
Descriptive statistics and bivariate analyses were conducted with SPSS v29 and primary regression‐based analyses were conducted with Mplus 8.11. First, descriptive statistics characterized participants and examined response distributions. Next, bivariate analyses (i.e., Pearson's correlations for associations between continuous variables, independent‐samples t‐tests for associations between continuous and dichotomous variables, one‐way ANOVAs for associations between continuous and categorical variables, Chi‐square tests for associations between categorical variables) examined associations of baseline rural–urban residence and sociodemographic covariates with healthcare access (i.e., unmet healthcare needs, healthcare availability) and follow‐up cannabis use outcomes (i.e., any past‐month cannabis use, medical cannabis use), as well as associations of healthcare access with cannabis use. Finally, regression‐based models were conducted separately for each cannabis use outcome. Each of the 2 regression‐based models examined: (1) direct associations of rural versus urban residence with potential mediators (i.e., unmet healthcare needs, healthcare availability) and the cannabis use outcome; (2) direct associations of potential mediators with the cannabis use outcome; and (3) indirect associations of rural versus urban residence on the cannabis use outcome via each potential mediator. Regression‐based models were run using the MODEL INDIRECT command using weighted least square mean and variance (WLSMV) estimation to account for the dichotomous nature of both cannabis outcome variables and one mediator (i.e., unmet healthcare needs) [20, 21, 22]. We obtained probit regression estimates for all effects on dichotomous outcomes, which were transformed into probabilities. Acceptable model fit values are CFI values of ≥ 0.90 and RMSEA and SRMR values of ≤ 0.08 [23]. Regression‐based models controlled for all sociodemographic covariates.
3. Results
3.1. Participant Characteristics
Shown in Table 1, in this sample of SMYAs, participants were 21.75 years old on average (SD = 2.11); 32.6% resided in a rural area; 66.7% reported female sex; 45.5% identified as gender minority; 40.3% reported a monosexual identity (gay or lesbian), 42.3% a bisexual+ identity (bisexual or pansexual), and 17.5% another sexual identity (asexual or queer); 44.6% identified as racial and/or ethnic minority; 22.2% reported a high school education, 43.5% some college, and 34.2% a Bachelor's degree or higher; and 33.5% resided in a state with legal recreational cannabis. Over half of participants (59.4%) reported unmet healthcare needs and on average, participants rated the availability of healthcare services in their community as “average.” Regarding cannabis use at follow‐up, almost half of participants reported using cannabis in the past month (49.4%) with around one‐third reporting medical cannabis use (32.2%).
TABLE 1.
Participant characteristics and bivariate associations among rural–urban residence, healthcare access, and cannabis use.
| Total | Unmet healthcare need | Healthcare availability | Any current cannabis use | Current medical cannabis use | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Variables | N = 549 (100.0%) | No, N = 223 (40.6%) |
Yes, N = 326 (59.4%) |
M = 3.34 (SD = 1.14) |
No, N = 278 (50.6%) |
Yes, N = 271 (49.4%) |
No, N = 372 (67.8%) |
Yes, N = 177 (32.2%) |
||||
|
N (%) or M (SD) |
N (%) or M (SD) |
N (%) or M (SD) |
p |
M (SD) or r |
p |
N (%) or M (SD) |
N (%) or M (SD) |
p |
N (%) or M (SD) |
N (%) or M (SD) |
p | |
| Covariates | ||||||||||||
| Age, M (SD) | 21.75 (2.11) | 21.67 (2.25) | 21.80 (2.00) | 0.505 | 0.06 | 0.162 | 21.61 (2.12) | 21.89 (2.09) | 0.117 | 21.76 (2.12) | 21.73 (2.07) | 0.890 |
| Sex, N (%)^ | 0.004 | <0.001 | 0.395 | <0.001 | ||||||||
| Female | 366 (66.7) | 135 (36.9)a | 231 (63.1)a | 3.19 (1.12)a | 190 (51.9) | 176 (48.1) | 231 (63.1)a | 135 (36.9)a | ||||
| Male | 177 (32.2) | 88 (49.7)b | 89 (50.3)b | 3.69 (1.09)b | 85 (48.0) | 92 (52.0) | 138 (78.0)b | 39 (22.0)b | ||||
| Gender identity, N (%) | <0.001 | <0.001 | 0.338 | 0.003 | ||||||||
| Cisgender | 299 (54.5) | 148 (49.5)a | 151 (50.5)a | 3.54 (1.10)a | 157 (52.5) | 142 (47.5) | 219 (73.2)a | 80 (26.8)a | ||||
| Gender minority | 250 (45.5) | 75 (30.0)b | 175 (70.0)b | 3.09 (1.14)b | 121 (48.4) | 129 (51.6) | 153 (61.2)b | 97 (38.8)b | ||||
| Sexual identity, N (%) | 0.008 | 0.027 | 0.480 | 0.574 | ||||||||
| Monosexual | 221 (40.3) | 107 (48.4)a | 114 (51.6)a | 3.49 (1.11)a | 105 (47.5) | 116 (52.5) | 155 (70.1) | 66 (29.9) | ||||
| Bisexual+ | 232 (42.3) | 84 (36.2)b | 148 (63.8)b | 3.22 (1.13)b | 123 (53.0) | 109 (47.0) | 155 (66.8) | 77 (33.2) | ||||
| Another sexual identity | 96 (17.5) | 32 (33.3)b | 64 (66.7)b | 3.26 (1.20)a,b | 50 (52.1) | 46 (47.9) | 62 (64.6) | 34 (35.4) | ||||
| Race and ethnicity, N (%) | 0.543 | 0.263 | 0.084 | 0.099 | ||||||||
| NH White | 304 (55.4) | 120 (39.5) | 184 (60.5) | 3.29 (1.12) | 164 (53.9) | 140 (46.1) | 197 (64.8) | 107 (35.2) | ||||
| Racial and/or ethnic minority | 245 (44.6) | 103 (42.0) | 142 (58.0) | 3.40 (1.16) | 114 (46.5) | 131 (53.5) | 175 (71.4) | 70 (28.6) | ||||
| Education, N (%) | 0.402 | 0.142 | 0.982 | 0.168 | ||||||||
| High school or less | 122 (22.2) | 56 (45.9) | 66 (54.1) | 3.28 (1.24) | 61 (50.0) | 61 (50.0) | 81 (66.4) | 41 (33.6) | ||||
| Some college | 239 (43.5) | 93 (38.9) | 146 (61.1) | 3.26 (1.13) | 122 (51.0) | 117 (49.0) | 154 (64.4) | 85 (35.6) | ||||
| Bachelor's or higher | 188 (34.2) | 74 (39.4) | 114 (60.6) | 3.47 (1.07) | 95 (50.5) | 92 (48.9) | 137 (72.9) | 51 (27.1) | ||||
| State cannabis legalization | 0.548 | 0.264 | 0.303 | |||||||||
| Medical only | 365 (66.5) | 145 (39.7) | 220 (60.3) | 191 (52.3) | 174 (47.7) | 242 (66.3) | 123 (33.7) | |||||
| Medical and recreational | 184 (33.5) | 78 (42.4) | 106 (57.6) | 87 (47.3) | 97 (52.7) | 130 (70.7) | 54 (29.3) | |||||
| Rural–urban residence, N (%) | 0.004 | <0.001 | 0.507 | <0.001 | ||||||||
| Urban | 370 (67.4) | 166 (44.9)a | 204 (55.1)a | 3.50 (1.10)a | 191 (51.6) | 179 (48.4) | 270 (73.0)a | 100 (27.0)a | ||||
| Rural | 179 (32.6) | 57 (31.8)b | 122 (68.2)b | 2.99 (1.55)b | 87 (48.6) | 92 (51.4) | 102 (57.0)b | 77 (43.0)b | ||||
| Healthcare access | ||||||||||||
| Unmet healthcare need, N (%) | — | — | 0.023 | <0.001 | ||||||||
| No | 223 (40.6) | — | — | — | 126 (56.5)a | 97 (43.5)a | 177 (79.4)a | 46 (20.6)a | ||||
| Yes | 326 (59.4) | — | — | — | 152 (46.6)b | 174 (53.4)b | 195 (59.8)b | 131 (40.2)b | ||||
| Healthcare availability, M (SD) | 3.34 (1.14) | — | — | — | — | — | 3.43 (1.12) | 3.24 (1.15) | 0.053 | 3.48 (1.09)a | 3.04 (1.18)b | <0.001 |
Note: Healthcare availability assessed on a scale of 1 (Poor) to 5 (Excellent). Some participant responses were aggregated due to small sample sizes: cisgender = cisgender women (N = 171) and cisgender men (N = 128) aggregated; gender minority = gender nonbinary (N = 194), transgender men (N = 48), and transgender women (N = 8) aggregated; another sexual identity = asexual (N = 30) and queer (N = 66) aggregated. Bolded values denote statistical significance at p < 0.05. Different superscripts denote statistically significant differences at p < 0.05.
^Six participants reported another sex.
3.2. Associations of Rural–Urban Residence With Healthcare Access
Bivariate analyses examining associations of rural–urban residence and covariates with healthcare access are shown in Table 1. A greater proportion of participants with unmet healthcare needs resided in rural (vs. urban) areas. Additionally, participants in rural areas reported significantly less healthcare availability than those in urban areas.
Direct associations of rural–urban residence and covariates with healthcare access are shown in Table 2. Accounting for sociodemographic covariates, rural (vs. urban) residence was associated with greater likelihood of reporting unmet healthcare needs (probability = 0.45) and with less healthcare availability.
TABLE 2.
Direct effects of rural–urban residence on healthcare access and rural–urban residence and healthcare access on cannabis use.
| Unmet healthcare need | Healthcare availability | Any current cannabis use | Current medical cannabis use | |||||
|---|---|---|---|---|---|---|---|---|
| Variables | B (SE) | p | B (SE) | p | B (SE) | p | B (SE) | p |
| Covariates | ||||||||
| Age | 0.03 (0.03) | 0.301 | −0.01 (0.02) | 0.926 | 0.05 (0.03) | 0.096 | 0.03 (0.03) | 0.306 |
| Sex | ||||||||
| Female | REF | REF | REF | REF | REF | REF | REF | REF |
| Male | −0.16 (0.13) | 0.214 | 0.37 (0.11) | <0.001 | 0.13 (0.13) | 0.305 | −0.27 (0.14) | 0.051 |
| Gender identity | ||||||||
| Cisgender | REF | REF | REF | REF | REF | REF | REF | REF |
| Gender minority | 0.45 (0.12) | <0.001 | −0.30 (0.10) | 0.002 | 0.07 (0.12) | 0.596 | 0.06 (0.13) | 0.630 |
| Sexual Identity | ||||||||
| Monosexual | −0.20 (0.13) | 0.114 | 0.08 (0.11) | 0.435 | 0.19 (0.12) | 0.131 | 0.13 (0.13) | 0.343 |
| Bisexual+ | REF | REF | REF | REF | REF | REF | REF | REF |
| Another sexual identity | −0.04 (0.16) | 0.798 | 0.14 (0.13) | 0.300 | 0.04 (0.16) | 0.816 | 0.03 (0.17) | 0.838 |
| Race and ethnicity | ||||||||
| NH White | REF | REF | REF | REF | REF | REF | REF | REF |
|
Racial and/or ethnic minority |
0.05 (0.12) | 0.697 | −0.04 (0.10) | 0.701 | 0.22 (0.11) | 0.048 | −0.07 (0.12) | 0.549 |
| Education | ||||||||
| High school or less | −0.19 (0.15) | 0.186 | 0.01 (0.12) | 0.966 | 0.06 (0.14) | 0.682 | 0.06 (0.15) | 0.673 |
| Some college | REF | REF | REF | REF | REF | REF | REF | REF |
| Bachelor's or higher | 0.06 (0.14) | 0.668 | 0.05 (0.12) | 0.670 | −0.09 (0.14) | 0.525 | −0.18 (0.14) | 0.204 |
| State cannabis legalization | ||||||||
| Medical only | REF | REF | REF | REF | REF | REF | REF | REF |
| Medical and recreational | −0.03 (0.12) | 0.820 | 0.11 (0.10) | 0.258 | 0.19 (0.12) | 0.105 | −0.02 (0.13) | 0.868 |
| Rural–urban residence | ||||||||
| Urban | REF | REF | REF | REF | REF | REF | REF | REF |
| Rural | 0.32 (0.13) | 0.013 | −0.42 (0.10) | <0.001 | 0.12 (0.12) | 0.509 | 0.24 (0.13) | 0.061 |
| Healthcare access | ||||||||
| Unmet healthcare need | ||||||||
| No | — | — | — | — | REF | REF | REF | REF |
| Yes | — | — | — | — | 0.15 (0.07) | 0.029 | 0.29 (0.07) | <0.001 |
| Healthcare availability | — | — | — | — | −0.10 (0.05) | 0.048 | −0.13 (0.05) | 0.010 |
Note: Model fit: CFI = 0.98–0.99; RMSEA = 0.01–0.05; SRMR = 0.01–0.05. Bolded values denote statistical significance at p < 0.05. Different superscripts denote statistically significant differences between groups at p < .05. Groups sharing a superscript are not significantly different from one another.
3.3. Associations of Rural–Urban Residence and Healthcare Access With Cannabis Use
Bivariate associations of rural–urban residence, healthcare access, and covariates with cannabis use are shown in Table 1. A greater proportion of participants reporting current medical cannabis use resided in rural (vs. urban) areas. Additionally, a greater proportion of participants reporting current cannabis use and current medical cannabis use reported unmet healthcare needs. Those reporting current medical cannabis use also reported significantly less healthcare availability than those reporting no current medical use.
Direct associations of rural–urban residence, healthcare access, and covariates with cannabis use outcomes are shown in Table 2. Unmet healthcare needs (vs. no unmet healthcare needs) were associated with greater likelihood of any cannabis use (probability = 0.12) and medical cannabis use (probability = 0.37). Greater healthcare availability was associated with lower likelihood of any cannabis use (probability = 0.05) and medical cannabis use (probability = 0.14).
Regarding indirect effects of rural–urban residence on cannabis use outcomes via healthcare access (not shown in tables), rural (vs. urban) residence was indirectly associated with greater likelihood of medical cannabis use through greater likelihood of unmet healthcare needs (B [SE] = 0.09[0.04], p = 0.032) and less healthcare availability (B [SE] = 0.06[0.03], p = 0.030). Rural–urban residence was not indirectly associated with any cannabis use through either unmet healthcare needs (B [SE] = 0.05[0.03], p = 0.100) or healthcare availability (B [SE] = 0.04[0.02], p = 0.075).
4. Discussion
In a survey of SMYAs residing in Oklahoma and surrounding states, rural participants were significantly more likely to experience unmet healthcare needs and report lower healthcare availability than their urban counterparts. These healthcare access disparities were consequential, as associations of rural (vs. urban) residence with greater likelihood of medical cannabis use were mediated by greater unmet healthcare needs and lower healthcare availability. While rurality did not predict general cannabis use, it was indirectly linked to medical cannabis use through its relationship with limited healthcare access. Together, these findings suggest that SMYAs in rural areas may be turning to cannabis use, particularly for medical purposes, as a response to systemic gaps in traditional healthcare services.
Previous research has documented disparities in healthcare access for both SM populations and rural residents individually [4, 5], but few studies have examined the intersection of SM and rural‐residing identities. Stigma‐related barriers to healthcare experienced by SM individuals [24, 25, 26, 27] may be compounded by geographic barriers to care in rural areas [10, 11]. These dual stressors are consistent with minority stress theory, which posits that marginalized groups face unique psychosocial burdens that impact their health and healthcare‐seeking behaviors [7]. Moreover, our data align with recent studies suggesting that medical cannabis may serve as a coping mechanism for individuals experiencing barriers to formal healthcare, especially in high‐stigma environments [3]. That said, our study also adds nuance: while prior research suggests that cannabis use prevalence may increase with rurality broadly, our findings indicate that this relationship is specific to medical cannabis use and is likely driven by unmet needs rather than recreational motives.
Several limitations should be acknowledged. First, our sample was geographically limited to the South–Central United States and was not representative of all SMYAs nationwide, which may limit generalizability. Second, data were self‐reported, and both cannabis use and healthcare avoidance can be underreported due to stigma, especially in regions where cannabis is not fully legalized. However, our use of online recruitment may have reduced some social desirability bias, and high retention (89.1%) strengthens our longitudinal findings. Nonetheless, the broader study design lends stronger support for directional associations than cross‐sectional studies. Importantly, research on rural SMYAs remains limited, and our study helps fill this gap by identifying both risks and behavioral responses within this underserved group. Future research should explore longitudinal trajectories of medical cannabis use among SMYAs to better understand whether use increases over time as a sustained response to unmet healthcare needs. Longitudinal studies with more than two waves of data are also needed to better elucidate temporal ordering of key constructs, as rural–urban residence and healthcare access were both assessed at baseline in the current study. Additionally, qualitative studies could provide deeper insight into the motivations, perceived risks, and lived experiences behind medical cannabis use in rural SM populations.
The results of this study highlight the pressing need to address healthcare disparities among rural SMYAs. The indirect association between rurality and medical cannabis use via healthcare access points to systemic shortcomings in the availability and inclusivity of health services. These findings underscore that cannabis use in this population may not simply be a lifestyle choice but a response to unaddressed medical needs. Public health efforts should prioritize improving healthcare infrastructure in rural areas and creating affirming, accessible healthcare environments for SMYAs. Without such interventions, vulnerable populations will have few options and little medical guidance in addressing their healthcare needs. Recognizing and addressing these gaps is essential to promoting health equity across geographic and identity‐based lines.
Funding
This work was supported by the American Cancer Society (134128‐IRG‐19‐142), with additional support provided by the National Institute on Drug Abuse (K01DA055073) and the Oklahoma Tobacco Settlement Endowment Trust (R22‐03). Funders had no role in the study design, collection, analysis or interpretation of the data, writing the manuscript, or the decision to submit the paper for publication.
Conflicts of Interest
The authors declare no conflicts of interest.
Data Availability Statement
The data that support the findings of this study are available from the University of Oklahoma Health Sciences. Restrictions apply to the availability of these data, which were used under license for this study. Data are available from the authors with the permission of the University of Oklahoma Health Sciences.
References
- 1. National Institute on Minority Health and Health Disparities , Minority Health and Health Disparities: Definitions and Parameters (National Institutes of Health, 2023), https://www.nimhd.nih.gov/about/strategic‐plan/nih‐strategic‐plan‐definitions‐and‐parameters.html. [Google Scholar]
- 2. Mallory C. and Sears B., “Discrimination Against LGBT People in Oklahoma,” 2019, https://williamsinstitute.law.ucla.edu/publications/lgbt‐discrim‐ok/.
- 3. Vogel E. A., McQuoid J., Romm K. F., Kendzor D. E., and Cohn A. M., “Unmet Healthcare Needs and Medical Cannabis Use Among Sexual and Gender Minoritized Adults in a High‐Stigma Environment,” Journal of Homosexuality 72, no. 1 (2024): 1–19, 10.1080/00918369.2024.2302427. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Tabaac A. R., Solazzo A. L., Gordon A. R., Austin S. B., Guss C., and Charlton B. M., “Sexual Orientation‐Related Disparities in Healthcare Access in Three Cohorts of U.S. Adults,” Preventive Medicine 132 (2020): 105999, 10.1016/j.ypmed.2020.105999. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Reilly M., “Health Disparities and Access to Healthcare in Rural vs. Urban Areas,” Theory in Action 14, no. 2 (2021): 6–27, 10.3798/tia.1937-0237.2109. [DOI] [Google Scholar]
- 6. Cohn A., Sedani A., Niznik T., et al., “Population and Neighborhood Correlates of Cannabis Dispensary Locations in Oklahoma,” Cannabis 6, no. 1 (2023): 99–113, 10.26828/cannabis/2023.01.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Fischer B., Robinson T., Bullen C., et al., “Lower‐Risk Cannabis Use Guidelines (LRCUG) for Reducing Health Harms From Non‐Medical Cannabis Use: A Comprehensive Evidence and Recommendations Update,” International Journal of Drug Policy 99 (2022): 103381, 10.1016/j.drugpo.2021.103381. [DOI] [PubMed] [Google Scholar]
- 8. Hatzenbuehler M. L., “How Does Sexual Minority Stigma ‘Get Under the Skin’? A Psychological Mediation Framework,” Psychological Bulletin 135, no. 5 (2009): 707–730, 10.1037/a0016441. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Meyer I. H., “Prejudice, Social Stress, and Mental Health in Lesbian, Gay, and Bisexual Populations: Conceptual Issues and Research Evidence,” Psychological Bulletin 129, no. 5 (2003): 674–697, 10.1037/0033-2909.129.5.674. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Joudeh L., Harris O. O., Johnstone E., Heavner‐Sullivan S., and Propst S. K., “‘Little Red Flags’: Barriers to Accessing Health Care as a Sexual or Gender Minority Individual in the Rural Southern United States—A Qualitative Intersectional Approach,” Journal of the Association of Nurses in AIDS Care 32, no. 4 (2021): 467–480, 10.1097/JNC.0000000000000271. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Douthit N., Kiv S., Dwolatzky T., and Biswas S., “Exposing Some Important Barriers to Health Care Access in the Rural USA,” Public Health 129, no. 6 (2015): 611–620, 10.1016/j.puhe.2015.04.001. [DOI] [PubMed] [Google Scholar]
- 12. Movement Advancement Project , “Where We Call Home: LGBT People in Rural America,” 2019, www.lgbtmap.org/rural‐lgbt. [Google Scholar]
- 13. Heffner J. L., Watson N. L., Dahne J., et al., “Recognizing and Preventing Participant Deception in Online Nicotine and Tobacco Research Studies: Suggested Tactics and a Call to Action,” Nicotine & Tobacco Research 23, no. 10 (2021): 1810–1812, 10.1093/ntr/ntab077. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Pratt‐Chapman M., Moses J., and Arem H., “Strategies for the Identification and Prevention of Survey Fraud: Data Analysis of a Web‐Based Survey,” JMIR Cancer 7, no. 3 (2021): e30730, 10.2196/30730. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Walker E. R., Cummings J. R., Hockenberry J. M., and Druss B. G., “Insurance Status, Use of Mental Health Services, and Unmet Need for Mental Health Care in the United States,” Psychiatric Services 66, no. 6 (2015): 578–584, 10.1176/appi.ps.201400248. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Statistics Canada , “Canadian Community Health Survey—Annual Component (CCHS),” Accessed 2 July 2025, https://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvVariableList&Id=1531795.
- 17. U.S. Department of Agriculture Economic Research Service , “Rural‐Urban Commuting Area Codes,” 2024, https://www.ers.usda.gov/data‐products/rural‐urban‐commuting‐area‐codes/. [Google Scholar]
- 18. Washington Rural Health Research Center , “Using RUCA Data,” 2025, https://depts.washington.edu/uwruca/ruca‐uses.php. [Google Scholar]
- 19. DISA Global Solutions Inc , “Marijuana Legality Map,” 2026, https://disa.com/marijuana‐legality‐by‐state/. [Google Scholar]
- 20. Muthén L. K. and Muthén B. O., Mplus User's Guide. Eighth Edition (Muthén & Muthén, 1998. –2017). [Google Scholar]
- 21. Nguyen T. Q., Webb‐Vargas Y., Koning I. M., and Stuart E. A., “Causal Mediation Analysis With a Binary Outcome and Multiple Continuous or Ordinal Mediators: Simulations and Application to an Alcohol Intervention,” Structural Equation Modeling: A Multidisciplinary Journal 23, no. 3 (2016): 368–383, 10.1080/10705511.2015.1062730. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Muthén B., “Applications of Causally Defined Direct and Indirect Effects in Mediation Analysis SEM in Mplus,” (2011), https://www.statmodel.com/download/causalmediation.pdf.
- 23. Kline R. B., Principles and Practice of Structural Equation Modeling (Guilford Publications, 2015). [Google Scholar]
- 24. Burton C. W., Lee J. A., Waalen A., and Gibbs L. M., ““Things Are Different Now but”: Older LGBT Adults' Experiences and Unmet Needs in Health Care,” Journal of Transcultural Nursing 31, no. 5 (2020): 492–501, 10.1177/1043659619895099. [DOI] [PubMed] [Google Scholar]
- 25. Power R., Ussher J. M., Perz J., Allison K., and Hawkey A. J., The Out With Cancer Study Team , “‘Surviving Discrimination by Pulling Together’: LGBTQI Cancer Patient and Carer Experiences of Minority Stress and Social Support,” Frontiers in Oncology 12 (2022): 918016, 10.3389/fonc.2022.918016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Le Forestier J. M., Page‐Gould E., and Chasteen A., “Identity Concealment May Discourage Health‐Seeking Behaviors: Evidence From Sexual‐Minority Men During the 2022 Global Mpox Outbreak,” Psychological Science 35, no. 2 (2024): 126–136, 10.1177/09567976231217416. [DOI] [PubMed] [Google Scholar]
- 27. Huang A. K., Nguyen M. T., German D., Alexander K., Charlton B. M., and Glick J. L., “Health Care Utilization and Care‐Seeking Behavior Among Vulnerabilized Sexual Minority Women: A Social‐Ecological Approach,” Journal of Health Care for the Poor and Underserved 35, no. 2 (2024): 583–604, https://muse.jhu.edu/article/928634. [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study are available from the University of Oklahoma Health Sciences. Restrictions apply to the availability of these data, which were used under license for this study. Data are available from the authors with the permission of the University of Oklahoma Health Sciences.