Use of Primary Care and Emergency Departments for Substance Use Treatment: The Rural and Urban Divide

Abstract

Background: Alcohol and illicit substance use remain significant public health issues in the United States. In this analysis, we assessed differences in the use of primary care and emergency departments (EDs) for treatment of substance use among rural and urban sexual minorities (SMs).

Methods: Data come from the National Survey on Drug Use and Health (NSDUH, 2015-2019). Survey-weighted multivariable linear and logistic regression analyses were used to assess the relationship between sexual identity and the use of primary care settings or EDs for treatment of substance use, stratified by urbanicity of residence.

Results: Among the entire sample, 7.9% reported residing in rural environments with slightly more SMs living in urban (7.3%) relative to rural (5.4%) locales. Both rural (β=-0.20; 95% CI: −0.29, −0.10) and urban SMs (β=-0.13; 95% CI: −0.16, −0.11) self-reported worse overall health. Urban SMs, but not rural SMs, had significantly higher odds of reporting use of primary care treatment for substance use (aOR 2.80; 95% CI: 2.13, 3.68). ED treatment for substance use was greater among both rural (aOR = 2.99; 95% CI: 1.01, 8.87) and urban SMs (aOR = 3.02; 95% CI: 2.12, 4.30) as was overall number of ED visits among both rural (β = 0.48; 95% CI: 0.24, 0.72) and urban SMs (β = 0.23; 95% CI: 0.19, 0.28) .

Conclusion: These findings suggest increased reliance on EDs for treatment of alcohol or substance use among rural SMs. Future research should examine whether increasing culturally competent primary care services for SMs in rural areas may be a key intervention point for reducing health disparities.

Keywords:

• emergency department
• sexual minorities
• rural
• urban

Introduction

In the United States (U.S.), substance use disorders (SUD) remain a significant public health issue, with nearly 1 in 3 individuals developing an SUD in their lifetime (Substance Abuse & Mental Health Services Administration, 2020). With the recent influx of potent opioids such as fentanyl, the risk of death by overdose in the U.S. has risen to over 100,000 deaths annually in 2022 (Ahmad et al., 2022). Alongside illicit substance use, nearly 1 in 4 individuals in the US also reported an episode of heavy drinking in 2019 (e.g., more than 5 drinks per single occasion for men and more than 4 drinks per single occasion for women) with nearly 79,000 deaths in the US attributable to alcohol misuse annually (Ahmad et al., 2022; Centers for Disease Control & Prevention, 2008; Substance Abuse & Mental Health Services Administration, 2020). However, rates of substance use and related disorders vary by geographic location and sexual minority (e.g., gay, lesbian & bisexual)(SM) status. Few studies have specifically examined rural sexual minority populations relative to urban sexual minorities.

One key difference between urban and rural populations is their concentration and distribution of SM populations (Substance Abuse & Mental Health Services Administration, 2020). The population of SMs in the US continues to grow and now accounts for 7.1% of the overall US population (Gallup, 2022). SM populations are at not only at elevated risk of SUD relative to heterosexuals (Boyd et al., 2019; Kann et al., 2016; Kerridge et al., 2017; Krueger et al., 2020; McCabe et al., 2009) but are also at elevated risk for binge drinking (Fish, 2019) and cannabis use (Boyd et al., 2019; Gruskin et al., 2001; Kann et al., 2016; Parent et al., 2019). Many sexual minorities reside in urban settings but there is noted growth in rural populations with rural sexual minorities accounting for nearly 5% of the rural population (Movement Advancement Project, 2019). However, there is a dearth of research that focuses on differences between rural and urban sexual minority populations.

One key difference between rural and urban populations more broadly is the availability of treatment services for substance use and its related disorders. For example, 20% of Americans live in rural environments; yet, less than a tenth of primary care physicians practice in these locations (Gudbranson et al., 2017; Nielsen et al., 2017), resulting in emergency departments becoming the default location for seeking treatment not only for everyday maladies but also treatment of substance misuse and its related disorders. In contrast, urban populations typically have closer proximity to substance use treatment facilities but, even so, many individuals continue to prioritize emergency department (ED) use for treatment as a result of their insurance status or the perceived cost at specialized substance use treatment facilities (Wani et al., 2019). Multiple studies have found higher rates of general ED visits among SM; yet, when experiencing a behavioral emergency (e.g., mental health or SUD); SMs were less likely to seek ED care when compared to heterosexual peers (Frimpong et al., 2020; Sánchez et al., 2007).

Research highlighting rural and urban differences pertaining to substance use-related primary care and ED visits is lacking among SM populations. This study aims to develop a better understanding of rural and urban differences in sexual orientation disparities in the use of primary care and emergency departments for the treatment of substance use-related problems. To achieve this goal, we utilized five years of data (2015-2019) from the National Survey on Drug Use and Health to examine rural and urban differences between sexual minorities and heterosexuals across four key areas: 1) overall self-reported health; 2) use of primary care services for substance use treatment; 3) emergency department use for substance use treatment; and 4) total number of emergency department visits.

Methods

Study population

Data were from the National Survey on Drug Use and Health (NSDUH) datasets, a publicly available, nationally representative cross-sectional survey of private, non-institutionalized U.S. citizens across all 50 states and the District of Columbia. This is an annual survey that provides information on a host of health-related factors, including health conditions, tobacco, alcohol, and drug use, and substance use treatment utilization. Survey sample weights are provided by NSDUH as part of the publicly available data and are used to obtain nonbiased estimates for survey outcomes. All data are deidentified and publicly available; thus studies using this data are exempt from Institutional Review Board review.

Data from 2015 through 2019 was utilized as 2015 was the first year in which NSDUH assessed self-reported participant sexual identity and 2019 was the final year of data available pre-COVID pandemic. Although 2020 data were available, several changes were made to the survey that limited consistency across a host of measures and thus were not included here. The analytic sample was also limited to adults, as those under the age of 18 participating in the survey did not self-report questions related to sexual identity. Finally, data were aggregated across all study years 2015-2019 (N = 210,392) and models were adjusted for year of data collection to ensure consideration of year-specific differences.

Demographic measures

Survey participants self-reported demographic information including age, sex, sexual identity, race and ethnicity, yearly income, and current health insurance status. To prevent potential identification of participants, age was provided only as a categorical variable in the publicly available NSDUH dataset: 18-25, 26-34, and 35 years or older. As discussed elsewhere (Morgan et al., 2021), NSDUH includes a single item representing sex/gender, with two options: “male” and “female.” In NSDUH codebooks, there is a lack of clarity regarding whether sex and/or gender are assessed via this item and how this item is administered (self or interviewer reported). This item also does not capture transgender and non-binary identities. We will use the term sex to refer to the construct this item assesses and the options as female and male. Race and ethnicity are combined as a single variable: non-Hispanic white, non-Hispanic Black, non-Hispanic Native American/Alaska Native, non-Hispanic Native Hawaiian/Pacific Islander, non-Hispanic Asian, and non-Hispanic Multiracial. Participants reporting a Hispanic ethnicity were coded as such, regardless of their racial identity. Sexual identity was asked as follows, “Which of the following do you consider yourself to be?” and coded as heterosexual, lesbian/gay, or bisexual. Income was similarly self-reported at the time of interview and coded as yearly income of: <$20,000, $20,000 to $49,999, $50,000 to $74,999, and $75,000 or higher. Lastly, possession of any health insurance at time of interview (e.g., private insurance, Medicare, Medicaid, Tricare, etc.) was operationalized as a dichotomous variable.

Region of residence

In keeping with the few past studies that have assessed rural versus urban differences using NSDUH data (Monnat & Rigg, 2016), we utilized the trichotomous variable of residence in a Core Based Statistical Area (CBSA). Options for this variable were coded as large CBSA (>1 million persons), small CBSA (<1 million persons), or participant does not reside in a CBSA (rural). Suburban areas are included in both large and small CBSAs depending on the population size. In this analysis, we operationalized this as a dichotomous variable, urban (resides in a CBSA) or rural (does not reside in a CBSA).

Self-reported health and health services

Self-reported overall health was assessed by asking participants, “This question is about your overall health. Would you say your health in general is …” The variable was reverse coded and operationalized as a four-level categorical variable: fair/poor, good, very good, or excellent.

Three variables were used to assess treatment receipt over the past year. The first two assessed receipt of care for substance use: 1) at the doctor’s office (“During the past 12 months, have you received treatment for your [alcohol/substance] in a private doctor’s office?”); or 2) in the emergency department (“During the past 12 months, have you received treatment for your [alcohol/substance] in an emergency room?”). These variables were not mutually exclusive. Each were operationalized as binary variables (0 = no treatment sought, 1 = treatment sought). The third variable included assessed the number times treated in the emergency room over the past twelve months, operationalized as a continuous variable.

Statistical analyses

Participant characteristics were described using means, standard deviations, and proportions, as appropriate. Multivariable logistic regression models were utilized to assess the association between sexual identity and each of the healthcare treatment variables, adjusting for demographic characteristics, insurance status, and year of data collection. Multivariable linear regression models were similarly used to assess the association between sexual identity and self-reported health or number of times treated in the emergency room, again adjusting for demographic variables. All models were stratified by rural versus urban status. Finally, all models were weighted to account for NSDUH’s stratified cluster sampling design. Statistical significance was established at alpha <0.05. All analyses were performed in Stata 17.0.

Results

In total, 210,392 participants were in the analytic sample across the five-year study period (Table 1) with 16,649 (7.9%) residing in non-CBSA, rural environments, while 193,743 (92.1%) resided in a CBSA or an urban environment. Among all participants, 195,385 (92.9%) identified as heterosexual, 4,640 (2.2%) as gay or lesbian, and 10,367 (4.9%) as bisexual. Meanwhile, 5.4% of rural residents reported being a sexual minority while 7.3% of urban residents reported the same. A plurality of the sample was 35 or older (n = 98,607, 46.9%), identified as non-Hispanic white (n = 127,556, 60.6%), reported income of ≥$75,000 (n = 70,244, 33.4%), and had health insurance at the time of survey completion (n = 184,996, 88.5%). Rural residents reported a mean of 0.55 (standard deviation [SD] = 1.5) times treated in emergency departments while urban residents reported a mean of 0.64 (SD = 1.7) times treated. Meanwhile, 521 (0.3%) urban residents received ED treatment for substance use in the past year while 20 (0.2%) of rural residents received the same. And 864 (0.5%) urban residents sought treatment from their primary care provider compared to 81 (0.5%) of rural residents.

Table 1. Demographic characteristics of participants in the analytic sample, stratified by urbanicity of residence, NSDUH, 2015–2019.

	Total		Rural		Urban
	(N = 210,392)		(n = 16,649)		(n = 193,743)
Variable	n	%	n	%	n	%	p-value
Age^1							<0.001
18–25 years	68,684	32.7	4,810	28.9	63,874	33.0
26–34 years	43,101	20.5	3,134	18.8	39,967	20.6
35 or older	98,607	46.9	8,705	52.3	89,902	46.4
Sexual Orientation							<0.001
Heterosexual	195,385	92.9	15,749	94.6	179,636	92.7
Lesbian/Gay	4,640	2.2	226	1.4	4,414	2.3
Bisexual	10,367	4.9	674	4.1	9,693	5.0
Sex							0.749
Female	112,188	53.3	8,858	53.2	103,330	53.3
Male	98,204	46.7	7,791	46.8	90,413	46.7
Race							<0.001
NH White	127,556	60.6	13,191	79.2	114,365	59.0
NH Black	26,477	12.6	1,068	6.4	25,409	13.1
NH Native American/Alaska Native	2,991	1.4	1,025	6.2	1,966	1.0
NH Native Hawaiian/Other Pacific Islander	1,028	0.5	20	0.1	1,008	0.5
NH Asian	9,636	4.6	64	0.4	9,572	4.9
NH Multiracial	6,937	3.3	457	2.7	6,480	3.3
Hispanic	35,767	17.0	824	5.0	34,943	18.0
Income							<0.001
<$20,000	41,662	19.8	3,942	23.7	37,720	19.5
$20,000 - $49,999	65,467	31.1	5,952	35.8	59,515	30.7
$50,000 - $74,999	33,019	15.7	2,761	16.6	30,258	15.6
≥$75,000	70,244	33.4	3,994	24.0	66,250	34.2
Health Insurance, any	184,996	88.5	14,454	87.3	170,542	88.6	<0.001
Doctors Office Trtmt., past year							0.453
Alcohol or Illicit Substance Treatment	945	0.5	81	0.5	864	0.5
Emergency Room Trtmt., past year
Number of Times Treated, mean (SD)	0.55 (1.51)	–	0.55 (1.5)	–	0.64 (1.7)	–	<0.001
Alcohol or Illicit Substance Treatment	561	0.3	40	0.2	521	0.3	0.491
Year of survey							<0.001
2015	42,802	20.3	3,223	19.4	39,579	20.4
2016	41,859	19.9	3,187	19.1	38,672	20.0
2017	41,802	19.9	3,442	20.7	38,360	19.8
2018	42,097	20.0	3,314	19.9	38,783	20.0
2019	41,832	19.9	3,483	20.9	38,349	19.8

Abbreviations: NSDUH = National Survey on Drug Use and Health; STI = sexually transmitted infection; NH = Non-Hispanic; Am/Ak = American Indian/Alaska Native.

¹Analytic sample includes only those individuals over the age of 18 as minors did not receive sexual orientation-related questions.

Table 2 presents weighted linear regression models assessing the relationship between sexual identity and either self-reported health or the number of times treated in the emergency department in the past year. Here, both rural (β = −0.20; 95% CI: −0.29, −0.10) and urban (β = −0.13; 95% CI: −0.16, −0.11) sexual minorities reported worse overall health relative to their heterosexual counterparts. Both rural (β = 0.48; 95% CI: 0.24, 0.72) and urban (β = 0.23; 95% CI: 0.19, 0.28) sexual minorities reported receiving care in an emergency department more often in the past 12 months compared to rural and urban heterosexuals, respectively.

Table 2. Multivariable survey-weighted linear regression analyses examining the association between sexual minority status and measures of health, NSDUH, 2015–2019.

	Rural^1		Urban^1
	(n = 16,468)		(n = 192.494)
	aOR	95% CI	aOR	95% CI
Overall Self-Rated Health
Heterosexual	Ref	–	Ref	–
Sexual Minority^2	−0.20***	−0.29, −0.10	−0.13***	−0.16, −0.11
Number of Emergency Treatments
Heterosexual	Ref	–	Ref	–
Sexual Minority^2	0.48***	0.24, 0.72	0.23***	0.19, 0.28

***p < 0.001; **p < 0.01; *p < 0.05.

¹Models adjusted for age, sex, race, income, health insurance status, and year or survey.

²Sexual minorities combined into a single category due to limited power.

Table 3 presents weighted logistic regression models examining the relationship between sexual identity and each of the treatment variables, stratified by rural versus urban residence. Regarding receipt of primary care for substance use-related issues, urban sexual minorities were more likely to receive care relative to urban heterosexuals (aOR = 2.80; 95% CI: 2.13, 3.68); results were non-significant among rural sexual minorities. Meanwhile, regarding ED treatment, compared to urban heterosexuals, urban sexual minorities had significantly greater odds of receiving care for substance use (aOR = 3.02; 95% CI: 2.12, 4.30). Similar results were observed among rural sexual minorities (aOR = 2.99; 95% CI: 1.01, 8.87) relative to their heterosexual counterparts.

Table 3. Multivariable survey-weighted logistic regression analyses examining the association between sexual minority status and past year primary care or emergency room treatment, NSDUH, 2015–2019.

	Rural^1		Urban^1
	(n = 16,468)		(n = 192.494)
	aOR	95% CI	aOR	95% CI
Primary Care Treatment for Alcohol or Illicit Substances
Heterosexual	Ref	–	Ref	–
Sexual Minority^2	1.94	0.72, 5.22	2.80***	2.13, 3.68
ED Treatment for Alcohol or Illicit Substances
Heterosexual	Ref	–	Ref	–
Sexual Minority^2	2.99*	1.01, 8.87	3.02***	2.12, 4.30

***p < 0.001; **p < 0.01; *p < 0.05.

¹Models adjusted for age, sex, race, income, health insurance status, and year or survey.

²Sexual minorities combined into a single category due to limited power.

Discussion

This study utilized a nationally representative dataset to assess rural and urban differences in sexual orientation disparities in overall health and primary and emergency department care for substance use treatment. Both rural and urban sexual minorities self-rated their overall health worse than their heterosexual counterparts. Urban, but not rural, SMs were more likely than their heterosexual peers to seek substance use treatment in primary care settings. Meanwhile, both rural and urban SMs were more likely than heterosexuals to seek ED treatment for substance use. Finally, SMs in either locale utilized the ED for emergency treatment at higher rates than did heterosexuals. These findings suggest not only stark differences between SMs and heterosexuals in their use of primary care or ED for substance use treatment but that there are key differences based on residence in either rural or urban environments.

Trends toward increased ED visits suggest a deterioration in the already compromised primary care infrastructure. In fact, fewer than a tenth of primary care providers serve rural populations, often leaving emergency care as the only option for medical treatment (Gudbranson et al., 2017). The already inadequate substance treatment infrastructure has become more scarce and may result in worse downstream health disparities among rural communities (Greenwood-Ericksen & Kocher, 2019), particularly among sexual minorities of color who have lower rates of health care insurance (Hsieh & Ruther, 2017). The increased use of ED for substance use treatment among rural and urban SM as compared to their heterosexual peers highlights the ED’s importance as a critical access point, specifically for SM. EDs have long been a point of critical access for care in the United States but are not well equipped to handle the complexities of substance use treatment and the unique health needs of SM (Ahmad et al., 2022). This research highlights the importance of EDs as both a point of access to care but also emphasizes the importance of adequate and culturally relevant substance treatment resources embedded within EDs. Establishing a streamlined substance treatment referral pipeline is essential to offloading the burden of our ED healthcare team. Future research should explore the health outcomes of SMs seeking substance treatment in the ED as well as conceptualize interventions to support EDs to best provide alcohol and substance treatment. Variables such as embedded substance treatment professionals, care coordinators or establishing relationships with existing substance treatment infrastructure are potentially key areas for future research.

Current research has found higher rates of substance use and substance use treatment in SM populations in comparison to their heterosexual peers (Kecojevic et al., 2012; McCabe et al., 2009). Our findings here highlight important differences between rural and urban SMs that suggest more nuanced research is needed to better understand differences in substance use and treatment among SMs. Urban SM were more likely to seek primary care treatment for substance use than urban heterosexuals. And while this may be a direct consequence of fewer primary care physicians in rural environments (Nielsen et al., 2017), both rural and urban SMs self-reported worse overall health and a greater number of ED treatment visits compared to rural and urban heterosexuals, respectively. These results suggest that there are unobserved issues in care among sexual minorities that may be contributing to extant health disparities. Additionally, treatment of substance use in EDs for SM populations, both urban and rural, may be the consequence of inadequate SM affirming primary care and non-ED substance treatment facilities. Future research among SMs should prioritize investigating the mechanisms that drive the use of EDs for treatment, in particular its use for treatment of substance use among this population.

This study is not without limitations. The NSDUH dataset is cross sectional therefore limiting the authors ability to conclude the directionality or temporality of associations. Additionally, the NSDUH survey includes a single item representing sex/gender, with two options: “male” and “female.” This prevents the examination of disparities affecting transgender and gender diverse populations. Given the lack of power, the authors combined all SMs into a single group. Although this provided adequate power, it may obscure potential differences in substance use treatment utilization for subgroups of SMs. Additionally, all substance use and related disorders are self-reported in NSDUH thus may be subject to bias in reporting by the participants, limiting the number of reported substance users. Finally, although the ED and primary care treatment questions were framed as receiving treatment in these environments, it is also possible that these were bridges to external substance use treatment services. Future research should aim to develop a more nuanced understanding of exactly where treatment was received rather than where it was initially sought alone.

Notwithstanding its limitations, this study provides a novel set of findings among a nationally representative sample. First, we observed worse self-reported health among SMs broadly compared to heterosexuals. Second, we noted that rural sexual minorities were more likely to seek treatment in an ED, but not in primary care, for substance use compared to their heterosexual peers. Future research should explore ways to increase primary care treatment of substance use as this may reduce the burden on EDs and be a key point of intervention to reduce substance use disparities. This work highlights the potential negative care experiences in primary care and substance treatment facilities (e.g., stigma & discrimination) and how it may be influencing SM to delay care. Ultimately this delay in care requires SM to seek substance use treatment in ED settings, suggesting the ongoing need to examine healthcare experiences of SMs.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by a grant from the National Institute on Drug Abuse at the National Institutes of Health (R03DA052651, PI: Morgan). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute on Drug Abuse or the National Institutes of Health. The sponsor had no involvement in the conduct of the research or the preparation of the article.