The State of the Research on Opioid Outcomes Among Lesbian, Gay, Bisexual, Transgender, Queer, and Other Sexuality- and Gender-Diverse Populations: A Scoping Review

Abstract

Purpose:

Research on opioid misuse, opioid use disorder (OUD), and overdose (i.e., opioid outcomes) among lesbian, gay, bisexual, transgender, queer, and other populations within the LGBTQ umbrella (LGBTQ+) remains sparse. The purpose of this scoping review was to characterize the state of the research on opioid outcomes among LGBTQ+ populations, and identify gaps in the extant literature and areas for future research.

Methods:

We conducted a scoping review of peer-reviewed, English language articles published between 2011 and 2020 that examined opioid outcomes among LGBTQ+ populations in the CINAHL, Embase, PubMed, and PsycINFO databases. We extracted data from articles that focused on opioid outcomes within their specific aims or purpose. We include a general summary for articles that secondarily described opioid outcomes among LGBTQ+ populations.

Results:

Of 113 published studies that examined opioid outcomes among LGBTQ+ populations, 10% (n = 11) were specifically designed to focus on this topic. Across studies, bisexual populations, particularly women, were at highest risk for opioid misuse and OUD. Few studies examined opioid outcomes by more than one dimension of sexual orientation (n = 3, 27%), race and/or ethnicity (n = 3, 27%), or age (n = 5, 45%). Only two included transgender or gender diverse samples; only one explicitly measured gender identity.

Conclusions:

Future research is needed to understand the impact of the opioid epidemic on LGBTQ+ people, particularly transgender and other gender diverse individuals, and the intersectional role of race, ethnicity, and age in opioid disparities among LGBTQ+ individuals. Additional research could contribute to the development of much-needed affirming OUD treatment and other services for LGBTQ+ people.

Introduction

As of 2019, 1.6 million people in the United States aged 12 or older had a current opioid use disorder (OUD) diagnosis based on Diagnostic and Statistical Manual of Mental Disorders, Version 5 (DSM-5) criteria.^1,2 More than 70% of drug overdose deaths in 2019 involved opioids.³ Americans now have a higher lifetime risk of death by opioid overdose than car accidents and gun violence.⁴ Extensive research and news media have focused on the opioid crisis,⁵ which has disproportionately impacted historically marginalized populations.^6,7

Experiences of marginalization and health disparities, including high rates of substance use and disorders, are well-documented among lesbian, gay, bisexual, transgender, queer, and other populations within the LGBTQ umbrella (LGBTQ+; e.g., nonbinary individuals)^8–13; however, research on opioid outcomes among LGBTQ+ populations remains sparse.¹⁰ Understanding the scope of opioid misuse (e.g., nonprescribed or not-as-prescribed use), OUD, and overdose among LGBTQ+ groups, including how rates compare with heterosexual and cisgender (i.e., gender identity aligns with sex assigned at birth) populations, is crucial for targeting public health resources and for providing appropriate, affirming OUD treatment to LGBTQ+ people who have historically faced extensive barriers and discrimination in health care settings.^13–18

LGBTQ+ populations may also experience specific vulnerabilities to opioid-related problems.¹⁹ First, LGBTQ+ people may be at increased risk of opioid overdose given other substance use disparities (e.g., alcohol, methamphetamine) and the link between poly-substance use disorders and opioid overdose.²⁰ Moreover, transgender individuals (i.e., whose gender identity does not align with sex assigned at birth) may experience chronic pain at higher rates than cisgender individuals and may use prescription opioids as pain management.²¹ Rates of gender-affirming surgery are increasing among transgender individuals; opioids are often prescribed for postoperative management.¹⁹ In addition, HIV prevalence is much higher among LGBTQ+ populations than the general population; older individuals living with HIV also tend to experience high rates of chronic pain.²² These factors may put LGBTQ+ populations at increased risk of prescription opioid misuse, which can be a precursor to heroin use and OUD.^23,24

Existing research has also identified notable gender differences in opioid outcomes in the general population. Although men have historically experienced higher rates of prescription opioid misuse compared with women, that gap has steadily narrowed in more recent years.²⁵ Women, particularly lesbian women,²⁶ report higher rates of chronic pain and are more likely to experience comorbid mental health disorders and OUD than men.^27–31 Men who have sex with men (MSM) may be exposed to opioids more readily than lesbian, gay, and bisexual (LGB) women, given elevated levels of methamphetamine use and the proliferation of fentanyl-laced methamphetamine,^31,32 and co-injection of methamphetamine and opioids.³² Research on gender differences in opioid outcomes among sexual minority populations remains limited.

Purpose, rationale, and objectives

Considering the dearth of available research and the unique risk factors, the purpose of this scoping review was to characterize the state of the research on opioid misuse, OUD, and overdose (i.e., opioid outcomes) among LGBTQ+ populations, and identify gaps in the extant literature and areas for future research. Unlike systematic reviews, which are appropriate for answering inquiries about specific treatments or clinical practices, scoping reviews focus on identifying and characterizing the extant literature on a particular topic, and elucidating gaps in emerging research topics^33–35 like opioid-related disparities among LGBTQ+ populations.

Based on the population/concept/context framework,^36–38 we developed the following specific objectives for this scoping review: (1) identify how opioid outcomes vary between LGBTQ+ populations versus heterosexual and cisgender populations; and (2) understand how opioid outcomes differ among LGBTQ+ subgroups (e.g., lesbian vs. bisexual women; gay/bisexual men vs. lesbian/bisexual women; transgender vs. cisgender sexual minorities) and relative to heterosexual and cisgender populations.

Methods

Study design and procedures

For this scoping review, we conducted our search in four databases: CINAHL, Embase, PubMed, and PsycINFO. The landmark Institute of Medicine (IOM) report highlighted both research and funding gaps related to studying and improving the health of LGBTQ+ populations.⁸ The report also provided a comprehensive summary of the state of LGBTQ+ health research as of 2011. Of note, the report did not cite any research regarding the opioid epidemic among LGBTQ+ populations,⁸ despite opioid-related deaths rising precipitously since 1999³⁹ and documented evidence of substance use-related health disparities among LGBTQ+ people for other classes of drugs.⁸ For these reasons, our search was limited to articles published between January 1, 2011 and December 31, 2020.

First, we created a list of search terms and keywords for each of the four databases, modified from a previous scoping review by the first and second authors.⁴⁰ We divided search terms into: (1) sexual orientation and gender identity (SOGI) terms; and (2) opioid terms. We combined these groups using the Boolean operator AND to find articles with a minimum of one search term or keyword in each group (i.e., SOGI related, AND opioid related; see Supplementary Appendix SA1 for search syntax). After conducting our search in each database, we exported citations to Endnote⁴¹ and eliminated duplicate citations.

We reviewed each article to determine eligibility for inclusion in the final review, using a set of hierarchical exclusion categories (Table 1) to restrict our review to articles that reported on opioid outcomes by sexual orientation and/or gender identity. We reviewed titles, followed by abstracts, and then full texts for the remaining articles. Each co-author participated in eligibility review; the lead author monitored all review and data abstraction to ensure consistency. We resolved discrepancies in article inclusion decisions through virtual meetings and email discussions. To reduce selection bias, we also examined the reference lists of previously excluded review articles (e.g., systematic reviews, meta-analyses, literature reviews) and included any cited articles that met our inclusion criteria.

Table 1.

Hierarchical Exclusion Criteria Used for Article Review

Exclusion criteria	Description (if applicable)
Not English	N/A
Not human	Research with nonhuman animals
Not peer-reviewed	Abstracts, dissertations, news alerts, opinion pieces, etc.
Drugs but not opioids	Articles that focus exclusively on nonopioid substance use
HIV/STI focus (including hepatitis), no opioids included	Articles that focus on HIV, hepatitis, or other STIs but did not also include opioid outcomes
Not LGBTQ+	Articles that either excluded LGBTQ+ individuals or did not explicitly measure SOGI
Not relevant	Articles on unrelated topics (e.g., nonopioid medications used in recovery from surgery; prevalence of tuberculosis among the sample) or that focused on opioid prescribing for an FDA-approved indication (e.g., pain management)
Review article	Articles that did not present original research (e.g., systematic reviews, meta-analyses, literature reviews)
No opioid outcomes by SOGI	Articles that included opioid outcomes and SOGI variables separately, but did not report on opioid outcomes by SOGI

FDA, Food and Drug Administration; LGBTQ+, lesbian, gay, bisexual, transgender, queer, and other populations within the LGBTQ umbrella; N/A, not applicable; SOGI, sexual orientation and gender identity; STI, sexually transmitted infections.

Measures

After eligibility review, we examined each included article to collect the following information: study location (U.S. vs. non-U.S.); study population (e.g., national population-based survey); sexual orientation (i.e., sexual identity, attraction, and/or behavior) and gender identity measures; whether the study included LGBTQ+ subgroups such as sexual minority women and transgender individuals, and a heterosexual and/or cisgender comparison group; and whether studies examined differences by age or race and ethnicity. We also indicated each study's opioid outcome measures and key opioid-related findings.

Finally, because we were interested in reviewing studies that were specifically designed to look at opioid outcomes among LGBTQ+ people, we focused our in-depth review on eligible articles that specifically incorporated opioid outcomes into their study aims and/or design. A detailed review of these articles is reported in the Results section. A general summary is included for the remaining articles, which secondarily described opioid outcomes among LGBTQ+ populations (e.g., a study primarily focused on HIV risk behaviors that also reported rates of injection opioid use).

Results

Our search strategy and eligibility review yielded 113 articles that reported on opioid outcomes by SOGI variables (Fig. 1). Of these, 102 (90%) reported on opioid outcomes by SOGI variables but did not specifically include this topic in their primary aims. The majority of these articles (n = 50) focused on substance use more broadly, including opioids, among exclusively LGBTQ+ populations. Several of these studies described prescription drug misuse outcomes in general, including prescription opioids. An additional 18 articles included sexual orientation and/or gender identity as one among many covariates for opioid and other substance use.

FIG. 1.

PRISMA flow diagram of the scoping review. STI, sexually transmitted infections; LGBTQ+, lesbian, gay, bisexual, transgender, queer, and other populations within the LGBTQ umbrella; SOGI, sexual orientation and gender identity.

The remaining 34 articles described studies focused on HIV and other sexually transmitted infections (STI) and sexual risk behaviors. In these articles, opioid misuse was sometimes included as a covariate of HIV and/or STI-related behavior among exclusively LGBTQ+ samples, or sexual orientation and/or gender identity were included as covariates in studies focused on HIV/STIs and opioid misuse. Eleven of the 113 articles (10%) focused specifically on opioid outcomes in their primary aims, which we describe in detail hereunder.^42–52

Study characteristics

Table 2 provides characteristics of the 11 LGBTQ+ opioid-focused articles. All 11 studies were conducted in the United States.^42–52 All but one (published in 2014)⁴³ were published in 2019^44-46,51 or 2020.^{42,47–50,52} Over half (n = 6, 55%) reported on data from national population-based surveys: five used the National Survey on Drug Use and Health (NSDUH)^{42,44,46,47,51} and one drew from the Youth Risk Behavior Survey.⁵² Nine studies (81%) categorized participants based on sexual identity^{42,44–49,51,52}; three (27%) included more than one dimension of sexual orientation.^45,46,52 Eight (73%) included a heterosexual comparison group.^{42,44,46–49,51,52} Several studies reported on opioid outcomes among sexual minority subgroups: seven (64%) included data on sexual minority (e.g., lesbian/bisexual) women^{42,44,46–49,51} and eight (73%) on sexual minority men (e.g., gay/bisexual).^42–48,51 Seven (64%) also reported on subgroup differences by sexual orientation (e.g., lesbian vs. bisexual people).^{42,46–49,51,52}

Table 2.

Characteristics of Opioid and Sexual Orientation and Gender Identity-Focused Studies (n = 11)

	n	%
Study location
United States	11	100.00
National population-based survey	6	54.55
Sexual orientation measurement
Identity	9	81.82
Behavior	3	27.27
Attraction	1	9.09
More than one indicator	3	27.27
Not measured	1	9.09
Heterosexual comparison group	8	72.73
Sample by sexual orientation
Sexual minority women as a distinct group	7	63.64
Sexual minority men as a distinct group	8	72.73
Examination of subgroup differences in opioid use by sexual orientation (e.g., lesbian vs. bisexual people)	7	63.64
Gender identity measured	1	9.09
Transgender people combined into 1 sample	0	0.00
Transfeminine spectrum sample	1	9.09
Transmasculine spectrum sample	0	0.00
Examination of subgroup differences in opioid use by gender identity (e.g., transgender women vs. transgender men)	0	0.00
Cisgender comparison group	0	0.00
Examined racial/ethnic differences in opioid use	3	27.27
Examined age differences in opioid use	5	45.45
Opioids measured
Heroin	4	36.36
Prescription drug misuse	11	100.00

Although two studies included transgender or gender diverse samples, only 1 of the 11 studies explicitly measured gender identity; most used self-reported categorizations of “female/male” or “woman/man” as proxies for gender identity. For this reason, we refer to “sex differences” when describing any study that did not directly assess gender identity. The one study that reported on gender identity was comprised entirely of transgender women and girls.⁴⁹ A second study, which compared opioid use and prescription practices among individuals who received gender-affirming top surgery with those who received other types of breast reduction surgeries, presumably included transgender and nonbinary (TNB) people; however, gender identity was not reported.⁵⁰ Few studies examined opioid outcomes by race or ethnicity (n = 3, 27%),^43,48,49 or by age (n = 5, 45%).^{43–45,48,49}

Opioid outcomes measures and findings by sexual orientation

Table 3 provides detailed findings from the 11 studies in which opioid outcomes among LGBTQ+ populations were the primary aim. Unless otherwise noted, the studies did not report on whether participants were cisgender or transgender. All 11 reported on prescription opioid misuse, which was variably measured in terms of lifetime use; past 12-, 6-, and 3-month use; past 30-day use; DSM-IV opioid dependence criteria; risk perception; and pill consumption.^42–52 One of these studies combined heroin use with prescription opioid misuse into a single dependent variable of “illicit opioid use.”⁴⁴ Seven of the eight studies that included a heterosexual comparison group reported higher prevalence or odds of prescription opioid misuse among the LGB^{42,46–48,51,52} or “other” sexual identity group than the heterosexual group.⁴⁹ See Table 3 for additional details.

Table 3.

Detailed Opioid Use Measurement and Outcomes Among LGBTQ+ Opioid-Focused Studies (n = 11)

Authors	Year	Study aim, purpose, or goal	Opioid use assesseda (NA = not assessed)		Heterosexual and/or cisgender comparison group?	Key opioid-related findings
			Heroin use	Prescription opioid misuse
Anderson-Carpenter et al.42	2020	“The present study examines associations between sociodemographic factors and both lifetime and past 12-month pain reliever misuse among military veterans [heterosexual vs. LGB].”	NA	Lifetime Past 12 months	Yes	Higher odds of lifetime misuse among bisexual vs. straight veterans (aOR = 3.04, 95% CI = 1.72–5.38). Higher odds of lifetime misuse among bisexual vs. straight men (aOR = 2.68, 95% CI = 1.30–5.53). Higher odds of lifetime misuse among bisexual vs. straight women. (aOR: 4.14, 95% CI: 1.65–10.37). Only bisexual vs. straight women had higher odds of past 12-month use (aOR = 3.47, 95% CI = 1.28–9.41). No differences in misuse among gay/lesbian vs. straight veterans
Buttram et al.43	2014	“We sought to examine predictors of prescription opioid misuse among a sample of high-risk substance-using MSM in South Florida.”	Past 3 months	Past 3 months	No	Past 3-month prescription opioid misuse prevalence: 25%; heroin use prevalence: 2.9% No significant differences in past 3-month use by age or race/ethnicity. Use associated with higher odds of past 3-month prescription opioid misuse: recent binge drinking; cocaine use; drug injection; substance dependence; lifetime history of arrest. Use associated with lower odds of past 3-month prescription opioid misuse: HIV positive status. No association between demographic characteristics and prescription opioid misuse
Capistrant and Nakash44	2019	“To address this gap in evidence of prevalence of illicit opioid use among adults by sexual identity, gender, and age, we used data from the NSDUH. We pooled 2015–2017 data among adults aged 18+ to estimate age- and gender-stratified prevalence differences between LGB and heterosexual adults.”	Length of time since last use (past 30 days, >30 but <12 months, >12 months ago). Combined with prescription misuse into “illicit opioid use”	Past 12 months. Combined with heroin use into “illicit opioid use”	Yes	Significantly higher past 12-month prevalence among LGB vs. heterosexual women in the 18–25 (14.4% vs. 5.9%, PD: 8.5, 95% CI = 6.8–10.2), 26–34 (12.4% vs. 5.2%, PD = 7.2, 95% CI = 5.0–9.4), and 35–49 (9.5% vs. 4.1%, PD = 5.4, 95% CI = 2.6–8.1) age groups, but not in the 50+ age group (3.7% vs. 2.1%, PD = 1.6, 95% CI = −0.5 to 3.8). Significantly higher past 12-month prevalence among gay/bisexual vs. heterosexual men in the 18–25 (11.4% vs. 8.2%, PD: 3.2, 95% CI = 0.5–5.9) and 50+ age groups (6.0% vs. 2.8%, PD = 3.2, 95% CI = 0.3–6.0)
Chen et al.45	2019	“The purpose of this study is to…[describe] the use of any prescription opioids that has been linked to the progressive use of injection drugs and polysubstance use regardless of whether opioids are used medically or non-medically. We examine the prevalence of prescription opioid use among YBMSM as well as the individual and network correlates of prescription opioid use among YBMSM longitudinally.”	NA	Past 12 months	No	Past 12-month use prevalence: 4.7% (4.2% weighted) Cumulative incidence: 3.6% (weighted: 4.1%). No significant differences in past 12-month use by age. Prescription opioid misuse associated with higher odds of: economic hardship, criminal system involvement, being a victim of violence, illicit drug use other than marijuana, condomless anal sex. Prescription opioid misuse use associated with lower odds of having a mother figure
Duncan et al.46	2019	“The purpose of the current study is to examine sexual orientation (including both sexual identity and sexual attraction) differences in prescription opioid misuse and prescription opioid use disorder among a nationally representative sample of adults in the U.S.”	NA	Past 30 days Past 12 months DSM-IV criteria for opioid dependence	Yes	Prescription opioid misuse prevalence: Past-year: Heterosexual 4.5%; Gay/lesbian 8.6%; Bisexual —12.0%; Attracted only to opposite sex—4.4%; Mostly opposite sex—9.0%, equal to both sexes—7.5%; mostly same sex—7.8%; only same sex—6.9%. Past month: Heterosexual—1.3%; Gay/lesbian—1.7%; bisexual—4.4%; attracted only to opposite sex—1.3%; mostly opposite sex—2.6%, equal to both sexes—2.6%; mostly same sex—1.8%; only same sex—1.4%. Prescription opioid use disorder prevalence: heterosexual—0.7%; gay/lesbian—.8%; bisexual – 2.2%; attracted only to opposite sex—0.7%; mostly opposite sex—1.8%, equal to both sexes—1.0%; mostly same sex—1.0%; only same sex—1.5%. Bisexual individuals: higher odds of past-year (aOR = 1.53, 95% CI = 1.20–1.97) and past-month (aOR = 1.66, 95% CI = 1.14–2.42) prescription opioid misuse vs. heterosexual individuals. Those attracted mostly to the opposite sex (aOR = 2.15, 95% CI = 1.77–2.63) or equally attracted to both sexes (aOR = 1.78, 95% CI = 1.38–2.30): higher odds of past-year prescription opioid misuse vs. those only attracted to the opposite sex. In sex-stratified analyses, associations held only for female individuals
Morgan et al.47	2020	“We aimed to assess potential mechanisms [depression and suicidal ideation] underlying disparities in past-year prescription opioid misuse affecting SM adults in the NSDUH from 2015 to 2018.”	NA	Past 12 months	Yes	5.5% of the whole sample (N = 169,759; sexual minority n = 11,268) reported past 12-month prescription opioid misuse between 2015 and 2018. Prevalence of opioid misuse declined for all groups from 2015 to 2018, but remained higher for sexual minority vs. heterosexual groups in 2018 (lesbian/gay: 7.9%, bisexual: 9.8%; heterosexual: 4.3%). Higher odds of past-year prescription opioid misuse among lesbian (aOR = 1.89, 95% CI = 1.50–2.38) and bisexual (aOR = 2.93, 95% CI = 2.59–3.31) vs. heterosexual women, and gay (aOR = 1.62, 95% CI = 1.28–2.05) and bisexual (aOR = 1.62, 95% CI = 1.26–2.09) vs. heterosexual men (adjusting for demographics only). After adjusting for either major depression or suicidal ideation, the strength of the association decreased but remained significantly higher for most SM vs. heterosexual groups, except bisexual men in the model adjusted for suicidal ideation. Counterfactual analyses: rates of opioid misuse among SM groups would be lower if rates of depression and suicidal ideation were decreased to levels among heterosexual people
Pitzer et al.48	2020	“The current study examines the role of opioid-related attitudes—specifically, acceptance of misuse and risk perceptions—in relation to ever and past 6 months prescription opioid misuse by sexual orientation among a national sample of youth and young adults in the U.S.”	NA	Lifetime Past 6 months. Acceptance of misuse. Risk perception	Yes	Prevalence of lifetime (heterosexual: 19.7%, lesbian or gay: 29.2%; bisexual: 32.6%, p < 0.001) and past 6-month (heterosexual: 7.2%, lesbian or gay: 12.5%; bisexual: 9.2%, p = 0.013) prescription opioid misuse significantly higher among sexual minority vs. heterosexual groups. Compared with heterosexual young people, lesbian/gay and bisexual young people had higher average levels of acceptance and lower average levels of perceived risk. Tobacco (aOR = 2.26, 95% CI = 1.15–4.46) and marijuana use (aOR = 2.05, 95% CI = 1.21–3.47) associated with higher odds of ever misuse among bisexual participants. Having college educated parents (aOR = 3.36, 95% CI = 1.07–10.53) and not meeting basic expenses (aOR = 3.93, 95% CI = 1.25–12.36) associated with greater odds of ever misuse among lesbian/gay participants. Greater odds of ever misuse among older vs. younger bisexual participants (age 15–17 aOR = 0.33, 95% CI = 0.13–0.84 vs. age 25–34, ref). Greater odds of past 6-month use among younger gay/lesbian (age 15–17 aOR = 6.45, 95% CI = 1.30–32.07; age 18–21, aOR = 4.82, 95% CI = 1.10–21.04); and younger bisexual (age 18–21 aOR = 2.75, 95% CI = 1.09–6.92) vs. age 25–34 (ref). No significant differences in ever or past 6-month misuse among female vs. male lesbian/gay or female vs. male bisexual respondents. No significant differences in race or ethnicity among sexual minority participants, but higher odds of ever misuse (aOR = 1.31, 95% CI = 1.04–1.67) and past 6-month misuse (aOR = 2.41, 95% CI = 1.69–3.44). Higher acceptance associated with higher odds of ever misuse among bisexual participants (aOR = 3.66, 95% CI = 1.69–7.89) and past 6-month use among lesbian/gay (aOR = 2.97, 95% CI = 1.47–6.01) and bisexual (aOR = 3.47, 95% CI = 1.98–6.08) participants. Greater risk perception associated with misuse only among heterosexual participants (aOR past 6-month use: 0.71, 95% CI = 0.60–0.85)
Restar et al.49	2020	“We estimated the prevalence of nonmedical prescription opioid use among a convenience sample of young transgender women in 2 urban centers heavily affected by opioid addiction and misuse and identify factors associated with this use.”	NA	Lifetime use	Yes (compared opioid use prevalence/risk factors among transgender girls and young women by sexual identity)	Prevalence of lifetime prescription opioid misuse: 11.8% of transgender adolescent girls and young women (vs. 12.5% national lifetime prevalence among cisgender adolescent girls/young women). In unadjusted logistic regression models, increasing age, “other” sexual identity, history of arrest or incarceration, alcohol use disorder or other substance use disorder symptoms, forgoing mental health services because of discrimination, and cigarette smoking were all associated with greater odds of prescription opioid misuse. Black participants had lower odds of misuse vs. participants of other races. In adjusted models, cigarette smoking predicted higher odds of prescription opioid misuse (smoking month or less: aOR = 3.92; 95% CI = 1.10–13.89; smoking daily: aOR = 5.69; 95% CI = 1.87–17.33, vs. those who didn't smoke). In adjusted models, “Other” sexual identity had significantly higher odds of lifetime prescription opioid misuse vs. heterosexual (aOR = 3.69; 95% CI = 1.07–12.72)
Robinson et al.50	2020	“In this study, we compare patient characteristics and [opioid] prescribing practices in three surgeries where patients underwent removal of large portions or all of their breast tissue [gender-affirming mastectomy/masculinizing chest reconstruction; oncologic mastectomy; mammoplasty reduction].”	NA	Proportion of pills consumed out of pills prescribed	Unclear (gender identity of breast reduction and breast cancer mastectomy patients not reported)	Patients who received gender-affirming top surgery (49%), as well as those who underwent breast reduction surgery (54%), consumed a higher proportion of pills prescribed vs. those with oncologic mastectomy (<20% of pills prescribed). Breast cancer mastectomy patients prescribed significantly less morphine vs. gender-affirming and breast reduction patients. Gender-affirming patients prescribed significantly more morphine than breast reduction patients but consumed the same amount
Schuler et al.51	2019	“We examine LGB opioid-related disparities, relative to heterosexuals, in a national sample and characterize variation among LGB adults with respect to sexual identity and gender. We use 2015–2017 NSDUH data to examine LGB disparities in lifetime prescription pain reliever misuse, heroin use and injection heroin use; past-year opioid misuse, and opioid use disorder; and perceived heroin risk and heroin access.”	Lifetime general use Lifetime injection use Past 12-month Opioid use disorder DSM-IV criteria Ease of accessing heroin Perceived risk of use	Lifetime. Past 12-month Opioid use disorder DSM-IV criteria	Yes	Lifetime prevalence of prescription opioid misuse: gay men—19%; bisexual men—17%; heterosexual men—12%. Gay/lesbian women—17%; bisexual women—25%; heterosexual women—8.7%. Lifetime heroin use: gay men—2.8%; bisexual men—5.7%; heterosexual men—2.8%. Gay/lesbian women—2.6%; bisexual women—5.1%; heterosexual women—1.1%. Lifetime heroin use significantly higher among bisexual vs. heterosexual men (aOR = 1.8, 95% CI = 1.1–3.0). Lifetime injection heroin use significantly higher among bisexual vs. heterosexual women (aOR = 4.0, 95% CI = 2.7–6.1). Past 12-month prescription opioid misuse significantly higher among bisexual vs. heterosexual women (aOR = 2.4, 95% CI = 2.0–2.8); lesbian/gay vs. heterosexual women (aOR = 1.6, 95% CI = 1.1–1.2); gay vs. heterosexual men (aOR = 1.4, 95% CI = 1.0–2.0). Past 12-month OUD significantly higher among bisexual vs. heterosexual women (aOR = 2.5, 95% CI = 1.7–3.5). LGB women had lower perceived risk and more perceived heroin access vs. heterosexual women
Wilson et al.52	2020	“We examine adolescent disparities in lifetime odds of heroin and prescription opioid misuse by sexual minority status using YRBS data.”	Lifetime	Lifetime. Past 12 months	Yes	Lifetime prescription opioid misuse prevalence among youth: gay/lesbian (28.5%); bisexual (25.1%); heterosexual (12.5%). Prevalence of lifetime heroin use: gay/lesbian (10%); bisexual (4.1%), heterosexual (1.1%). Gay/lesbian (aOR = 1.96; 95% CI = 1.29–2.96), bisexual (aOR = 2.27; 95% CI = 1.79–2.88), and unsure youth (aOR = 1.44, 95% CI = 1.06–1.94) had higher odds of prescription opioid misuse vs. heterosexual youth. Having had both-sex sexual partners predicted significantly higher odds of prescription opioid misuse (aOR = 2.62; 95% CI = 2.10–3.25) vs. only different-sex sexual contact. Gay/lesbian (aOR = 4.84; 95% CI = 2.42–9.67), bisexual (aOR = 4.27, 95% CI = 2.39–7.63), and unsure youth (aOR = 8.20, 95% CI = 4.47–15.02) had higher odds of lifetime heroin use vs. heterosexual youth. Same-sex sexual partners and both-sex sexual partners predicted higher odds of lifetime heroin use vs. only different-sex sexual partners (aOR = 3.77; 95% CI = 1.68–8.44 and aOR = 7.44; 95% CI = 4.59–12.06, respectively)

^a Aside from heroin and prescription opioids, no other opioids were assessed among the focused studies included in this review.

aOR, adjusted odds ratio; CI, confidence interval; DSM-IV, Diagnostic and Statistical Manual of Mental Disorders, 4th Edition; LGB, lesbian, gay, bisexual; MSM, men who have sex with men; NA, not assessed; NSDUH, National Survey on Drug Use and Health; OR, odds ratio; OUD, opioid use disorder; PD, prevalence difference; SM, sexual minority; YBMSM, young Black men who have sex with men; YRBS, Youth Risk Behavior Survey.

Only four studies reported on heroin use,^43,44,51,52 which was variably measured as lifetime use (in general and injection); past 12- and 3-month use; time since last use; DSM-IV opioid dependence; ease of accessing heroin; and perceived risk of use. Only two studies with a non-LGBTQ+ comparison group reported on heroin use; both found that heroin use prevalence or odds were higher among LGB survey respondents than heterosexual respondents (neither included a TNB sample; Table 3).^51,52

Opioid outcome disparities among bisexual populations

Several studies noted particularly elevated rates of prescription opioid misuse and heroin use among bisexual and other nonmonosexual populations (e.g., those attracted to more than one sex or gender) versus heterosexual participants.^{42,46–48,51} These studies found that although rates of prescription opioid misuse declined among the general population from 2015 to 2018, disparities have persisted among LGB people⁴⁷ and are particularly pronounced for both adult^42–47 and youth^48,52 populations who are bisexually identified^{42,46–48,51} and/or bisexually attracted.^46,52

For example, Anderson-Carpenter et al. found that bisexual veterans, regardless of sex, reported significantly higher odds of lifetime prescription opioid misuse compared with heterosexual veterans (adjusted odds ratio [aOR] = 3.04, 95% confidence interval [CI] = 1.72–5.38) after adjusting for demographic characteristics.⁴² There were no significant differences in prescription opioid misuse among gay and lesbian versus heterosexual veterans.⁴²

Similarly, Duncan et al. found that bisexual individuals experienced higher odds of past 12-month (aOR = 1.53, 95% CI = 1.20–1.97) and past 30-day prescription opioid misuse (aOR = 1.66, 95% CI = 1.14–2.42) versus heterosexual individuals. Compared with participants only attracted to the “opposite sex,” those who were “mostly attracted to the opposite sex” (aOR = 2.15, 95% CI = 1.77–2.63), or equally attracted to “both sexes” (aOR = 1.78, 95% CI = 1.38–2.30) had higher odds of past 12-month misuse. Like Anderson-Carpenter et al., no significant differences emerged between gay/lesbian and heterosexual individuals in adjusted models (Table 3).⁴⁶

Opioid outcome differences by sex

Five of the 11 studies reported sex-stratified findings;^{42,44,46,47,51} only one compared outcomes by sex within sexual orientation subgroups (e.g., bisexual women vs. bisexual men).⁴⁸ None of these studies directly assessed gender identity. Like overall findings among bisexual versus heterosexual individuals, several of the studies that stratified by sex found opioid disparities for bisexual women and men versus heterosexual women and men. For example, Anderson-Carpenter et al. reported significantly higher odds of lifetime prescription opioid misuse among bisexual men versus heterosexual men (aOR = 2.68, 95% CI = 1.30–5.53) and bisexual women versus heterosexual women (aOR = 4.14, 95% CI = 1.65–10.37).⁴² Schuler et al. noted higher prevalence of heroin use among bisexual versus heterosexual men (aOR = 1.8, 95% CI = 1.1–3.0).⁵¹

Opioid disparities appear to be particularly pronounced for bisexual versus heterosexual women.^42,46,51 For example, in the Anderson-Carpenter et al. study, only bisexual women had higher odds of past 12-month use than heterosexual women (aOR = 3.47, 95% CI = 1.28–9.41); there were no significant differences in past 12-month use among subgroups of sexual minority men. Similarly, Duncan et al. found that after stratifying by sex, only bisexual versus heterosexual women and those who were attracted to “both sexes” versus only the “opposite sex” continued to have higher odds of past 30-day or past 12-month prescription opioid misuse, whereas there was no longer a significant difference between bisexual and heterosexual men.⁴⁶ In addition, Schuler et al. noted that bisexual women had significantly higher odds of lifetime injection heroin use (aOR = 4.0, 95% CI = 2.7–6.1) and past 12-month OUD versus heterosexual women (aOR = 2.5, 95% CI = 1.7–3.5).⁵¹

Although bisexual women appear to face higher odds of prescription opioid misuse than their heterosexual counterparts, some studies found that gay men faced the greatest disparities compared with heterosexual men.^47,51 For example, Morgan et al. found higher odds of past 12-month prescription opioid misuse among gay (aOR = 1.62, 95% CI = 1.28–2.05) and bisexual (aOR = 1.62, 95% CI = 1.26–2.09) versus heterosexual men in models that adjusted for demographic characteristics; however, after adjusting for suicidal ideation, the association only remained significant for gay versus heterosexual men.⁴⁷ Likewise, Schuler et al. observed higher odds of prescription opioid misuse among gay men (aOR = 1.4, 95% CI = 1.0–2.0) but not bisexual men, compared with heterosexual men.⁵¹

In the single study comparing prescription opioid misuse by sex within sexual orientation subgroups, no significant differences in prescription opioid misuse were observed between lesbian women versus gay men or bisexual women versus bisexual men. See Table 3 for additional details.⁴⁸

Opioid outcomes among TNB populations

Both the studies that included TNB samples focused on prescription opioid misuse.^49,50 In the study comprising transgender women and girls,⁴⁹ lifetime prevalence of prescription opioid misuse was lower among the study's convenience sample of TNB women/girls compared with a national sample of cisgender adolescent girls and young women (11.8% vs. 12.5%). In unadjusted logistic regression models, increasing age, “other” sexual identity, history of arrest or incarceration, alcohol use disorder or other substance use disorder symptoms, forgoing mental health services because of discrimination, and cigarette smoking were also associated with greater odds of prescription opioid misuse. After adjusting for these factors, transgender women and girls who identified their sexual identity as “other” had significantly higher odds of lifetime misuse compared with those who identified as heterosexual (aOR = 3.69; 95% CI = 1.07–12.72; Table 3).⁴⁹

The study that reported on differences by type of mastectomy/top surgery found that patients who received gender-affirming top surgery and those who underwent breast reduction surgery consumed a higher proportion of their postoperative opioid prescriptions than individuals who underwent oncologic mastectomy (respectively, 49%, 54%, and <20% of pills prescribed). In addition, patients who received gender-affirming top surgery were prescribed significantly higher doses of opioids than the other surgery groups (Table 3).⁵⁰

Opioid outcomes among MSM

Two studies focused on factors associated with prescription opioid misuse among MSM, without incorporating a non-LGBTQ+ comparison group.^43,45 Both studies observed an association between prescription opioid use and other types of substance use (recent binge drinking, cocaine use, drug injection, and substance dependence),^43,45 as well as adverse psychosocial factors (criminal justice involvement,^43,45 poverty, violence victimization, and condomless anal sex).⁴⁵ See Table 3 for additional details.⁴⁵

Age, race, and ethnicity comparisons

Less than half of the 11 studies looked at opioid outcomes among LGBTQ+ populations by age, race, or ethnicity. Of the five studies that reported on findings by age,^{43–45,48,49} two did not find any significant age differences in prescription opioid misuse, but both of these focused only on MSM.^43,45 In the study of prescription opioid misuse among transgender women and girls, odds of misuse increased with age in unadjusted logistic regression models.⁴⁹ A fourth study also reported greater odds of lifetime prescription opioid misuse among older versus younger bisexual participants (age 15–17, aOR = 0.33, 95% CI = 0.13–0.84 vs. age 25–34, reference), but greater odds of past 6-month misuse among younger gay/lesbian (age 15–17, aOR = 6.45, 95% CI = 1.30–32.07; age 18–21, aOR = 4.82, 95% CI = 1.10–21.04) and younger bisexual (age 18–21, aOR = 2.75, 95% CI = 1.09–6.92) participants versus those aged 25–34 years.⁴⁸

The fifth study stratified the sample by age to determine if sexual orientation differences in “illicit opioid use” (combined heroin use and prescription opioid misuse) were present across age groups.⁴⁴ This study found significantly higher past 12-month prevalence among LGB versus heterosexual women in the three younger age groups (age 18–25: prevalence difference [PD] = 8.5, 95% CI = 6.8–10.2; age 26–34: PD = 7.2, 95% CI = 5.0–9.4; age 35–49: PD = 5.4, 95% CI = 2.6–8.1), but not in the 50+ age group. Among men, higher prevalence was only found among gay and bisexual compared with heterosexual men in the 18–25 (PD = 3.2, 95% CI = 0.5–5.9) and 50+ age groups (PD = 3.2, 95% CI = 0.3–6.0).⁴⁴

Although one study considered race by focusing on opioid outcomes among young Black MSM,⁴⁵ only three studies examined racial and/or ethnic differences in opioid outcomes by SOGI.^43,48,49 Among these, two did not find any significant racial or ethnic differences among sexual minority participants.^43,48 The third study found that Black transgender women and girls had significantly lower odds of prescription opioid misuse than their white counterparts, but only before adjusting for other demographic characteristics, other substance use disorders, cigarette smoking, experiences of incarceration, and forgoing mental health services as a result of discrimination.⁴⁹

Discussion

This scoping review demonstrates a dearth of research on opioid outcomes among LGBTQ+ populations, with only 11 articles focused on this topic. All studies were conducted in the United States; six reported on data from national population-based surveys; eight included a heterosexual comparison group. All but one were published between 2019 and 2020. Although all 11 studies reported on prescription opioid misuse, only 4 included heroin use outcomes and none reported on opioid overdose outcomes. The extant literature reveals stark disparities among LGBTQ+ populations; unique disparities among LGBTQ+ subgroups such as bisexual people; few comparisons of sexual minority populations by sex, gender, race, or ethnicity; and limited research among TNB populations. These findings underscore the need for more research to understand nuances in opioid outcomes among LGBTQ+ people and to identify opportunities for intervention.

Opioid outcomes among bisexual and other nonmonosexual groups

Bisexual and other nonmonosexual populations, and particularly nonmonosexual women, appear to be at highest risk for opioid misuse and OUD compared with lesbian, gay, or heterosexual populations. Elevated substance use and substance use disorders among nonmonosexual individuals, especially women, may be related to the stigma that nonmonosexual people face from both heterosexual and lesbian/gay individuals.⁵³ As noted in the studies included in the current review,^42,46,51 such stigma may magnify the minority stress that nonmonosexual individuals already experience, potentially leading to higher rates of substance use as a coping mechanism.^54–56

This research and our findings suggest that OUD treatment for LGBTQ+ populations should be sensitive to the unique experiences of nonmonosexual individuals, particularly women, ideally with staff who are trained to recognize how stigma and sexism may play a role in OUD. Moreover, a previous review found that there is a scarcity of substance use interventions for sexually diverse women,⁴⁰ suggesting a need for more work in this area.

Opioid outcome differences by sex

Although most of the studies that stratified opioid outcomes by sex observed the highest disparities among bisexual women relative to lesbian and heterosexual women, two studies also suggest that gay men may have particular opioid vulnerabilities relative to heterosexual men. Of note, only one study in this review examined opioid outcomes by sex within sexual orientation subgroups (e.g., bisexual women vs. bisexual men).

General population studies suggest that women and men experience different psychosocial factors that may influence progression to OUD as well as OUD treatment outcomes, such as more caregiving responsibilities, greater likelihood of intimate partner violence, and lower levels of social support among women.^57–60 In studies of cisgender heterosexual individuals, sex differences in the prevalence of substance use disorders have also been attributed to differences in coping styles, with women being more likely to report internalized coping strategies such as rumination and men being more likely to engage in externalizing coping behaviors such as drug use.⁶¹ It is not known whether such findings apply to LGBTQ+ populations.

Most studies in this review used imprecise and often conflated methods for assessing sex and gender (e.g., comparing “females” and “males” to examine gender differences without directly measuring gender identity). Future research should use consistent measures for sex and gender to better inform preventive interventions and improve OUD treatment for LGB as well as TNB populations.

Opioid outcomes among TNB groups

Research on opioid outcomes among TNB populations is even more limited than among (presumably) cisgender LGB populations. Although two studies presumably included TNB populations, only one study in our review examined opioid outcomes among an explicitly transgender sample, reporting lower prevalence of prescription opioid misuse among transgender women compared with the general population. A recent study (published after the timeframe for inclusion in the current scoping review), however, found that transgender individuals had higher prevalence of OUD diagnoses compared with cisgender individuals.⁶² TNB individuals may also be more likely to be exposed to prescription opioids as a result of gender-affirming surgery, and opioid exposure is linked to higher rates of OUD.¹⁹ In fact, one of the reviewed articles demonstrated that those who had gender-affirming top surgery were prescribed and consumed higher dosages of opioid pain medications than individuals who underwent oncologic mastectomy.⁵⁰ Thus, future studies and tailored treatments are crucial to better understand and address these disparities.

Importance of multidimensional sexual orientation measurement

In addition to the gaps in the opioid and SOGI-focused literature, most of the reviewed studies only included one measure of sexual orientation—typically sexual identity. Multidimensional indicators of sexual orientation (e.g., sexual identity, sexual behavior, and sexual attraction) may be particularly important for understanding the scope of SOGI-related opioid outcomes, but such measures are seldom included in nationally representative surveys.⁶³ Individuals whose sexual identities do not align with societal expectations of sexual behavior (e.g., women who identify as heterosexual and who have same-sex partners) may experience particularly elevated rates of substance use, misuse, and disorders.^63,64 Sexual attraction may also be a more developmentally appropriate indicator of sexual orientation for examining opioid and SOGI-related outcomes among sexual minority youth, whose identities may be rapidly evolving.^46,65

Opioid outcomes among LGBTQ+ populations by race, ethnicity, and age

Few studies in this review examined outcomes by race, ethnicity, or age. LGBTQ+ people who are also people of color experience stigma, oppression, and discrimination motivated by the intersections of racism, homophobia, and transphobia, all of which can increase the potential for opioid misuse and OUD.⁶⁶ Thus, future research should examine differences in opioid outcomes among LGBTQ+ populations by exposure to racialized discrimination, and should include racially and ethnically diverse samples that make such analyses possible.⁴⁰ Such nuanced research is needed to inform services for LGBTQ+ people who experience multiple axes of discrimination and stigma.

Age is also an important factor for understanding the specific opioid-related risk factors among LGBTQ+ populations,⁶⁷ but is often overlooked. In the general population, rates of opioid misuse appear to be the most elevated among young adults, whereas overdose mortality is more prevalent among middle-aged adults.⁶⁸ Additional research suggests differences in patterns of opioid initiation, with younger adults more likely to initiate with prescription opioids and older adults more likely to do so with heroin.^67,69 Younger individuals are more likely than their older counterparts to obtain prescription opioids through nonmedical avenues such as peers, whereas older adults are more likely to misuse prescription opioids that they have been prescribed by a medical professional.⁷⁰ Research examining age differences in opioid outcomes among LGBTQ+ populations remains scarce,⁷¹ thus limiting opportunities to develop tailored prevention and treatment interventions.

Limitations

This scoping review presents a synthesis of the available literature on opioid outcomes among LGBTQ+ populations, a vastly understudied area. Still, this review has several limitations. While the four databases used in this review represent a variety of academic disciplines (i.e., medicine, public health, nursing, psychology), we are not able to comment on publications that are only indexed elsewhere. Focusing only on peer-reviewed literature may also present publication bias. Although we intentionally restricted our review to studies that included opioid outcomes by SOGI variables in their study aims and/or design, we did not conduct an in-depth review of articles that secondarily examined SOGI-related opioid outcomes (e.g., as part of overall substance use or in the context of HIV/STI prevention). Finally, we intentionally chose to review articles published in 2011 or later because the IOM LGBTQ+ report had already reviewed literature published before 2011 and this timeframe coincided with the second wave of the current opioid epidemic; therefore, we are not able to comment on studies published outside this timeframe.

Conclusions

This scoping review highlights the need for increased research focused on opioid outcomes among LGBTQ+ individuals. Findings also reveal particular opioid-related disparities among bisexual and other nonmonosexual people, especially women. At present, no nationally representative surveys that focus on drug use include measures to identify TNB respondents. The consequence of such omission is reflected in the paucity of TNB-focused studies in this review.

Given the gaps in the extant literature on opioid outcomes among LGBTQ+ populations, we suggest the following specific areas for future research on this topic: (1) explore why bisexual and other nonmonosexual populations, particularly nonmonosexual women, experience especially high rates of opioid misuse and OUD; (2) examine differences in opioid outcomes among subpopulations of cisgender sexual minorities (e.g., cisgender gay men vs. cisgender lesbian women); (3) investigate how different dimensions of sexual orientation (e.g., sexual identity, behavior, and attraction) relate to opioid outcomes; (4) include comprehensive measures of SOGI in nationally representative surveys such as the NSDUH to more accurately characterize LGBTQ+ opioid and other substance use disparities; and (5) investigate the intersectional role of race and ethnicity and other sociodemographic factors (e.g., age) in opioid disparities among LGBTQ+ populations. Additional research could ultimately contribute to the development of much needed affirming opioid treatment and other services for LGBTQ+ people.

Authors' Contributions

M.M.P.-W., J.D.K., and E.A.P. conceptualized the article, collaboratively developed article search and eligibility criteria, and agree to be accountable for all aspects of the work. M.M.P.-W. conducted article searches using search criteria, reviewed articles for study inclusion, extracted and analyzed data from included articles, managed the study data, and wrote the first draft of the article. J.D.K. and E.A.P. conducted article searches using search criteria, reviewed articles for study inclusion, extracted and analyzed data from included articles, revised the article, and gave final approval of the version of the article to be published.

Disclaimer

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

E.A.P. was supported by a training grant [T32 MH019139; PI: Theodorus Sandfort, PhD] from the National Institute of Mental Health at the HIV Center for Clinical and Behavioral Studies at the NY State Psychiatric Institute and Columbia University [P30-MH43520; Center Principal Investigator: Robert Remien, PhD]. J.D.K. was supported by the National Institute on Alcohol Abuse and Alcoholism [NIAAA K23-AA028296; PI: J.D.K.].

Supplementary Material

Supplementary Appendix SA1