Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Jan 27.
Published before final editing as: Am J Prev Med. 2025 Dec 1:108207. doi: 10.1016/j.amepre.2025.108207

Lifetime Non-Prescribed Opioid Use by Sex, Race/Ethnicity, and Sexual Identification Among U.S. High School Students

Sam D Gardner 1, Megan E Marziali 2, Noah Kreski 3, Alissa Davis 4, Silvia S Martins 5
PMCID: PMC12833745  NIHMSID: NIHMS2127254  PMID: 41338535

Abstract

Introduction:

The opioid crisis continues to be a public health concern. Little attention has been given to the relationship between the intersectional heterogeneity of racial/ethnic identification and sexual/gender minority (SGM) status on opioid use among U.S. high school students and how this risk varies by sex.

Methods:

Data were derived from the national 2023 Youth Risk Behavior Survey (n = 20,103) for U.S. high school students. This study estimated odds of lifetime non-prescribed opioid use (n=2,465) among exposures race/ethnicity, SGM status, and stratified by sex.

Results:

Compared to heterosexual White adolescents, all categories of SGM and race/ethnicity (White (aOR=2.23,95% CI=1.57, 3.14), Hispanic/Latino (aOR=3.29,95% CI=2.09, 5.19), Black (aOR=2.68, 95% CI=1.87, 3.85), and Other (aOR=2.84,95% CI=1.88, 4.31)) had elevated odds of lifetime non-prescribed opioid use, which varied by sex. Compared to heterosexual White girls, the strongest association was among SGM Hispanic/Latino girls (aOR=3.26,95% CI=1.98, 3.36), followed by SGM and Black (aOR=2.53,95% CI=1.70, 3.76). Compared to heterosexual White boys, the strongest association was among SGM racially Other (aOR=4.58,95% CI=2.38, 8.82), followed by SGM Hispanic/Latino (aOR=3.23,95% CI= 1.68, 6.19). Compared to heterosexual and White, heterosexual Black boys had elevated odds of opioid use, whereas no difference was found among heterosexual Black girls.

Conclusions:

This study highlights the heterogeneity of opioid use within intersectional adolescent groups and that sex, race/ethnicity, and sexual identity need to be examined as subsets of the population through an intersectional lens. Interventions tailored with an intersectional approach are needed to reduce structural stigma and promote targeted opioid use initiation reductions.

Keywords: opioid, heroin, U.S., youth, race/ethnicity, sexual/gender minority, sex

Introduction

The opioid crisis continues to be a public health concern. In 2023, approximately 5.7 million people ages 12 and older had a diagnosed opioid use disorder (OUD),1 and there were 79,358 fatal opioid-involved overdoses.2 There are differences in rates of OUD and overdose based on age, which is critical to highlight, as receipt of first-line, evidence-based medication for OUD (MOUD)—methadone and buprenorphine—is virtually lacking in adolescent populations.3 In 2023, among people ages 12–17, 316 thousand were diagnosed with OUD,1 and there were 13.5 per 100 thousand fatal drug overdoses for people ages 15–24.4 Among adolescents in the US who experienced an overdose, the most common substance reported was fentanyl (24.0%), followed closely by heroin (21.4%) and non-medical prescription opioids (NMPO; 12.3%).5 According to the 2023 Youth Risk Behavior Survey (YRBS), 12% of high school students engaged in NMPO; however, these results varied based on sex, race/ethnicity, and sexual identification.6 The aims of this study are to investigate the relationship between the intersectional heterogeneity of racial/ethnic identification and sexual/gender minority (SGM) status on NMPO and heroin use among U.S. high school students and how this risk varies by sex.

There is heterogeneity in NMPO and illegal opioid use (e.g., heroin, illegally-manufactured fentanyl, etc.) across racial and ethnic groups.7 Data from the 2015–2016 National Survey on Drug Use and Health (NSDUH) indicates that, among 12–17 year-olds, Black and non-Hispanic adolescents had higher odds of opioid use (20% and 50%, respectively) when compared to non-Hispanic White adolescents.8 According to the CDC (trends 2017–2021),9 NSDUH (2022)10, and YRBS surveys (trends 2013–2023),11 the prevalence of NMPO and illegal opioid use across racial and ethnic groups has been shifting, however groups with the largest magnitude of prevalence include those who identify as multi-racial, Black, and Hispanic, with the lowest prevalence found among those who identify as non-Hispanic White and Asian. Overdoses are a residual complication of NMPO and illegal opioid use, which also varies based on race/ethnicity. Between 2019–2020, fatal drug overdoses increased 94.03% among adolescents aged 14–18, with the largest increase among American Indian and Alaska Native adolescents, followed by Hispanic adolescents.12 Conversely, between 2022 and 2023, there was a slight reduction in adolescent drug fatalities from 721 to 708 (of which 76% involved fentanyl),13 yet rates of age-adjusted fatal overdoses increased among Black/African American and Native Hawaiìan/Other Pacific Islander populations.2

The heterogeneity of NMPO and illegal opioid use risk varies based on sex and sexual/gender identity. In general, boys have been shown to use substances more frequently than girls, and at higher levels, among 12th graders in the U.S. compared to those in earlier grades.7 Data from the 2015–2019 NSDUH suggest that the magnitude of differences in NMPO and OUD varied by age and sex; the highest odds of NMPO (1.22) and OUD (1.85) were among girls in the youngest cohort (12–17 years).14 Similarly, CDC data (2017–2023) indicates that adolescent girls have higher rates of NMPO in comparison to older cohorts and boys.6 Sexual minority individuals (e.g., lesbian, gay, etc.) are people whose sexual identity and/or behaviors are not exclusively heterosexual and gender minority individuals (e.g., transgender, non-binary, etc.) are people whose sex assigned at birth does not align with their identity.15 It is well-documented that people who identify as SGM have higher rates of drug use, including opioids, in adult and adolescent populations when compared to their heterosexual and cisgendered counterparts.16

Disparities in substance use rates can, in part, be attributed to minority stress,17 which posits that SGM individuals experience stressors above and beyond the stressors experienced by their heterosexual and cisgendered counterparts.15,17,18 In 2019, the prevalence of NMPO and heroin use was greater among lesbian, gay, and bisexual high school students, and those who reported their sexual identity as “unsure” or “questioning”, compared to heterosexual students.19 Similarly, in 2023, the YRBS reported that high school students who identified as SGM were nearly twice as likely to have engaged in NMPO (18%) and used illicit drugs (15%) compared to their heterosexual counterparts (8%, 8%, respectively).6

Recent research assessing the intersections of sex, race/ethnicity, and sexual identity in a representative U.S. adult sample indicate varied vulnerabilities associated with opioid-related outcomes, highlighting pronounced disparities and higher odds of substance use among women who are SGM, multiracial SGM, Black SGM, and Hispanic bisexual.20 Among men, odds of substance use were increased among those identifying as multi-racial.20 However, there has been limited research dedicated to the youth population in the U.S. on the relationship between these three intersecting minority identities (sex, race/ethnicity, SGM) and their impact on opioid-related outcomes.

Methods

Study Sample

Data were derived from the 2023 Youth Risk Behavior Survey (YRBS), which includes high school students in grades 9–12 (n = 20,103). The 2023 YRBS consists of a three-stage cluster sampling design that was used to produce a nationally representative sample. The YRBS is a national school-based survey, conducted biennially, that was established in 1991 to monitor priority health behaviors contributing to leading causes of mortality, morbidity, and social health among youth/adults.21 In 2023, the school response rate was 49.8% and the overall response rate was 35.4%.21 The 2023 YRBS sampling frame consisted of all regular public, charter, parochial, and other nonpublic schools with students in at least one grade (9–12) in the 50 U.S. States and the District of Columbia. Alternative schools, special education, and schools with an enrollment of ≤40 students, vocational schools serving only students who also attended another school, and schools operated by the U.S. Department of Defense or the Bureau of Indian Education were excluded.21

Measures

The outcome variable for this analysis is lifetime non-prescribed opioid use. Lifetime non-prescribed opioid use was constructed based on two questions from the survey: “During your life, how many times have you taken prescription pain medicine without a doctor’s prescription or differently than how a doctor told you to use it” and “During your life, how many times have you used heroin?”.22 Respondents who reported (1) at least one instance of heroin use and/or (2) prescription pain medicine misuse, were coded as having lifetime non-prescribed opioid use. Otherwise, respondents were coded as never having used opioids.

Race/ethnicity was treated as a multi-level independent categorical variable and SGM status was treated as a binary independent variable. Race/ethnicity was categorized into a four-level variable (Non-Hispanic White (herein referenced as White), Hispanic/Latino, non-Hispanic Black/African American, and Other (which includes American Indian/Alaska Native/Native Hawai’ian/Pacific Islander/Multiple non-Hispanic/Asian)). This study combined multiple racial and ethnic identities due to small cell counts. Collapsing categories prohibited comparisons of lifetime non-prescribed opioid use between certain racial and ethnic identities. This study’s conceptual framework builds upon the risk environment framework and minority stress theory, theorizing that added levels of discrimination and stigma based on social positioning will increase opioid use, but not uniformly, as certain communities may have developed protective factors;17,23 thus, the referent group for regressions is White.

Respondents identified their sexual identification as “heterosexual (straight)”, “gay or lesbian”, “bisexual”, “some other way”, or “not sure”. Sex was determined with the question “What is your sex (Male/Female)” and those who identified as transgender endorsed the question “Some people describe themselves as transgender when their sex does not match the way they think or feel about their gender. Are you transgender?” Students who endorsed identifying as heterosexual and not transgender were coded as “Heterosexual (Cisgendered)”. To capture as many respondents who identify as SGM, those who endorsed “not sure” (4.69%) were included as part of the SGM category. This response could indicate students are questioning their sexuality or are experiencing discomfort in assigning themselves a sexual identity.24 Due to small cell sizes after stratifying by race/ethnicity, the SGM variable collapsed gay, lesbian, bisexual, some other way, not sure, and transgender-identified respondents into one SGM category and was dichotomized as heterosexual (cisgendered) (referent group) versus SGM.

Age, grade (9th, 10th, 11th, 12th, and “ungraded or other grade”), sex (male, female), and housing stability in the last 30 days (stable, unstable) were identified a priori as potential confounders, consistent with prior research.24–26 Age and grade were collinear; grade was included as a confounder.

Statistical Analysis

Frequencies of the outcome and the distribution of demographic variables by race/ethnicity and sexual identification were first examined, separately, across the analytic sample. Then, separate logistic regressions (Proc surveylogistic)27,28 were used to examine associations between each independent variable (race/ethnicity, SGM) and outcome (lifetime non-prescribed opioid use), which were then adjusted for grade, sex, and housing stability. An 8-level categorical variable was created combining race/ethnicity and sexual identity. Given the potential heterogeneity of SGM and race/ethnicity dynamics by sex,6 the sample was further stratified based on respondent-identified sex.

Only weighted data are included in this analysis. Individuals with missing values for outcome lifetime non-prescribed opioid use (n=275, 1.36% of total sample population) were removed from the dataset. The study uses de-identified public-use data; no ethics approval was needed. Data is weighted based on student sex, race/ethnicity, and grade to be nationally representative of all students in grades 9–12 attending U.S. public/private schools.24 Statistical analyses were conducted using SAS (version 9.4; SAS Institute) software to account for the complex sampling design.29

Results

Table 1 provides the overall sample characteristics. The total sample population, prior to restrictions, was 20,103. The study sample consisted of 19,828 individuals, which amounted to 19,799 after weighting. Overall, 2,465 (11.89%) respondents reported lifetime non-prescribed opioid use; 4,340 individuals (23.01%) self-identified as SGM, of which 844 (18.75%) endorsed lifetime non-prescribed opioid use. 47.54% of study sample identified as White (of which 1,040 (9.98%) endorsed lifetime non-prescribed opioid use), 26.82% identified as Hispanic/Latino (of which 556 (13.84%) endorsed lifetime non-prescribed opioid use), 12.79% identified as Black (of which 220 (12.75%) endorsed lifetime non-prescribed opioid use, and 10.91% identified as one of the following race/ethnicities defined as “Other” race and ethnicity: American Indian, Alaska Native, Native Hawai’ian or other Pacific Islander, multiple non-Hispanic, or Asian (of which 582 (13.05%) endorsed lifetime non-prescribed opioid use). The study collapsed these racial/ethnic identities into one category, due to small cell sizes when stratified by the outcome and by sex. Descriptive characteristics with cell counts for each race/ethnicity and stratified by the outcome can be found in Appendix Table 1 and stratified by sex in Appendix Table 2.

Table 1.

Sample Characteristics of United States Adolescents by Lifetime Non-Prescribed Opioid Use, 2023 Youth Risk Behavior Survey (n=20,103).

Descriptive Characteristics
Unweighted No. (weighted %) Lifetime Non-Prescribed Opioid Use Unweighted No. (weighted %) Never Opioid Use-Unweighted No. (weighted %)
Study Sample 19,828 (100) 2,465 (11.89) 17,363 (88.11)
Missing from total sample 275 (1.36)
Sex
Male 9,899 (51.23) 1,042 (9.81) 8,857 (90.19)
Female 9,775 (47.96) 1,392 (13.87) 8,383 (86.13)
Missing 154 (0.81)
Housing Stability (past 30 days)
Stable Housing 14,014 (86.26) 1,655 (10.74) 12,359 (89.26)
Unstable Housing 489 (2.88) 198 (44.36) 291 (55.64)
Missing 5,325 (10.86)
Grade
9th Grade 5,600 (26.14) 722 (12.46) 4,878 (87.54)
10th Grade 5,333 (25.52) 696 (13.45) 4,637 (86.55)
11th Grade 4,756 (24.08) 564 (10.70) 4,192 (89.30)
12th Grade 3,905 (23.12) 430 (10.17) 3,475 (89.83)
Ungraded or Other Grade 46 (0.23) 26 (61.60) 20 (38.40)
Missing 188 (0.91)
Race/Ethnicity
Non-Hispanic White 9,612 (47.54) 1,040 (9.98) 8,572 (90.02)
Hispanic/ Latinoa 3,926 (26.82) 556 (13.84) 3,370 (86.16)
Black or African American 1,747 (12.79) 220 (12.75) 1,527 (87.25)
Otherb 4,182 (10.91) 582 (13.05) 3,600 (86.95)
Missing 361 (1.94)
Sexual/Gender Identity
Heterosexual (Cisgendered) 13,214 (69.41) 1,293 (8.94) 11,921 (91.06)
Sexual/Gender Minorityc 4,340 (23.01) 844 (18.75) 3,496 (81.25)
Missing 2,274 (7.58)
Sexual/Gender Identity and Race/Ethnicity
Heterosexual (Cisgendered)
Heterosexual and non-Hispanic White 6,593 (33.25) 562 (7.39) 6,031 (92.61)
Heterosexual and Hispanic/Latinoa 2,691 (19.09) 289 (10.56) 2,402 (89.44)
Heterosexual and Black/African American 1,193 (8.90) 129 (10.83) 1,064 (89.17)
Heterosexual and Otherb 2,601 (7.61) 295 (9.33) 2,306 (90.67)
Sexual/Gender Minority c
SGM and White 2,087 (11.22) 345 (15.77) 1,742 (84.23)
SGM and Hispanic/Latinoa 861 (5.69) 201 (22.55) 660 (77.45)
SGM and Black/African American 432 (3.23) 80 (18.80) 352 (81.20)
SGM and Otherb 890 (2.50) 191 (19.74) 699 (80.26)
Missing 2,480 (8.51)
Open in a new tab
a

Hispanic/Latino includes Hispanic/Latino and Multiple Hispanic/Latino

b

Other includes American Indian/Alaska Native, Asian, Native Hawaiian and other Pacific Islanders, multiple non-Hispanic

c

Sexual/Gender Minority includes gay, lesbian, bisexual, some other way, not sure, and transgender

SGM, sexual/gender minority

Table 2 provides the multivariable analyses that were adjusted for sex, grade, and housing stability. Compared to White adolescents, those who identified as Black and Hispanic/Latino had 1.31 (95% Confidence Interval [CI]= 1.06, 1.61) and 1.41 (95% CI= 1.12, 1.77) times the odds, respectively, of lifetime non-prescribed opioid use. Compared to heterosexuals, SGM had 2.17 (95% CI= 1.71, 2.75) times the odds of lifetime non-prescribed opioid use. Assessing intersectional sexual and racial/ethnic identities, the highest odds of lifetime non-prescribed opioid use among SGM were among adolescents that identified as Hispanic/Latino (aOR=3.29, 95% CI=2.09, 5.19) compared to Heterosexual White adolescents.

Table 2:

Lifetime Non-Prescribed Opioid Use on Racial/Ethnic and/or Sexual/Gender Minorities Among U.S. High School Students, 2023 YRBS (weighted n=19,828)

OR (95% CI) aOR (95% CI)d
Sexual/Gender Identity
Heterosexual (Cisgendered) (n=13,214) 1 1
Sexual/Gender Minoritya (n=4,340) 2.35 (1.86, 2.98) 2.17 (1.71, 2.75)
Racial/Ethnic Identity
Non-Hispanic White (n=9,612) 1 1
Hispanic/ Latinob (n=3,926) 1.45 (1.13, 1.86) 1.41 (1.12, 1.77)
Black/African American (n=1,747) 1.32 (1.07, 1.63) 1.31 (1.06, 1.61)
Otherc (n=4,182) 1.35 (1.02, 1.80) 1.35 (1.03, 1.78)
Sexual/Gender Identity and Race/Ethnicity
Heterosexual and non-Hispanic White (n=6,593) 1 1
Heterosexual and Hispanic/ Latinob (n=2,691) 1.48 (1.15, 1.91) 1.46 (1.14, 1.87)
Heterosexual and Black/African American (n=1,193) 1.52 (1.11, 2.09) 1.54 (1.12, 2.12)
Heterosexual and Otherc (n=2,601) 1.29 (0.90, 1.85) 1.31 (0.92, 1.87)
SGMa and non-Hispanic White (n=2,087) 2.34 (1.68, 3.27) 2.23 (1.57, 3.14)
SGMa and Hispanic/ Latinob (n=861) 3.66 (2.27, 5.86) 3.29 (2.09, 5.19)
SGMa and Black/African American (n=432) 2.90 (2.02, 4.16) 2.68 (1.87, 3.85)
SGM and Otherc (n=890) 3.08 (2.06, 4.60) 2.84 (1.88, 4.31)
Open in a new tab
a

Sexual/Gender Minority includes gay, lesbian, bisexual, some other way, not sure, and transgender

b

Hispanic/Latino includes Hispanic/Latino and Multiple Hispanic/Latino

c

Other includes American Indian/Alaska Native, Asian, Native Hawaiian and other Pacific Islanders, multiple non-Hispanic.

d

Multivariable models were adjusted for respondent grade, sex, and housing stability

Boldface indicates statistical significance at p≤0.05

AOR, adjusted odds ratio; CI, confidence interval; SGM, sexual/gender minority; OR, odds ratio

Table 3 provides the multivariable analyses exploring the relationship between racial/ethnic identification and SGM, separately, and a combined racial/ethnic and SGM status category, and how these associations vary by sex, adjusted for grade and housing stability. Among girls, those who identified as SGM had 2.09 (95% CI=1.62, 2.70) times the odds of lifetime non-prescribed opioid use compared to girls who identified as heterosexual. Among boys, those who identified as SGM had 2.26 (95% CI=1.57, 3.25) times the odds of lifetime non-prescribed opioid use, compared to heterosexual boys. Regarding race/ethnicity, the only significant association found among boys was among Hispanic/Latino boys (aOR=1.42, 95% CI=1.04, 1.95) and for girls was among those who identified as racially/ethnically Other (aOR=1.43, 95% CI=1.04, 1.98) compared to White girls.

Table 3:

Lifetime Non-Prescribed Opioid Use on Racial/Ethnic and/or Sexual/Gender Minorities Among U.S. High School Students, Stratified by Sex, 2023 YRBS (n=19,828)

Girls (n=9,775) Boys (n=9,899)
OR (95% CI) aOR (95% CI)d OR (95% CI) aOR (95% CI)d
Sexual/Gender Identity
Heterosexual (Cisgendered) 1 1 1 1
Sexual/ Gender Minoritya 2.09 (1.63, 2.68) 2.09 (1.62, 2.70) 2.27 (1.55, 3.33) 2.26 (1.57, 3.25)
Racial/Ethnic Identity
Non-Hispanic White 1 1 1 1
Hispanic/ Latinob 1.43 (1.01, 2.01) 1.38 (0.99, 1.92) 1.44 (1.04, 1.98) 1.42 (1.04, 1.95)
Black/African American 1.21 (0.89, 1.65) 1.21 (0.89, 1.65) 1.42 (0.99, 2.03) 1.41 (0.97, 2.05)
Otherc 1.45 (1.05, 2.00) 1.43 (1.04, 1.98) 1.31 (0.94, 1.83) 1.32 (0.94, 1.85)
Sexual/Gender Identity and Race/Ethnicity
Heterosexual and non-Hispanic White 1 1 1 1
Heterosexual and Hispanic/Latinob 1.49 (1.07, 2.08) 1.46 (1.06, 2.01) 1.43 (1.04, 1.98) 1.43 (1.03, 1.98)
Heterosexual and Black/African American 1.41 (0.95, 2.09) 1.46 (0.99, 2.16) 1.61 (1.06, 2.46) 1.62 (1.05, 2.50)
Heterosexual and Otherc 1.51 (0.94, 2.42) 1.53 (0.96, 2.45) 1.16 (0.81, 1.65) 1.16 (0.81, 1.66)
SGMa and non-Hispanic White 2.17 (1.56, 3.02) 2.22 (1.60, 3.09) 2.12 (1.27, 3.55) 2.16 (1.30, 3.59)
SGMa and Hispanic/ Latinob 3.35 (2.03, 5.53) 3.26 (1.98, 3.36) 3.33 (1.63, 6.81) 3.23 (1.68, 6.19)
SGMa and Black/African American 2.50 (1.69, 3.72) 2.53 (1.70, 3.76) 2.96 (1.53, 5.75) 2.94 (1.54, 5.61)
SGMa and Otherc 2.50 (1.62, 3.85) 2.49 (1.61, 3.86) 4.45 (2.32, 8.52) 4.58 (2.38, 8.82)
Open in a new tab
a

Sexual/Gender Minority includes gay, lesbian, bisexual, some other way, not sure, and transgender

b

Hispanic/Latino includes Hispanic/Latino and Multiple Hispanic/Latino

c

Other includes American Indian/Alaska Native, Asian, Native Hawaiian and other Pacific Islanders, multiple non-Hispanic

d

Multivariable models were adjusted for respondent grade and housing stability

Boldface indicates significance at p≤0.05

AOR, adjusted odds ratio; CI, confidence interval; SGM, sexual/gender minority; OR, odds ratio

Assessing the heterogeneity of intersectional sexual and racial/ethnic identities on lifetime non-prescribed opioid use, stratified by sex (Table 3), suggested differences in effect sizes. Compared to White heterosexual girls or boys, respectively, the strongest association among girls were those who identified as SGM and Hispanic/Latino (aOR=3.26, 95% CI=1.98, 3.36), followed by SGM and Black (aOR=2.53, 95% CI=1.70, 3.76); among SGM boys, the strongest associations were those who identified as SGM and racially/ethnically-identified Other (aOR= 4.58, 95% CI=2.38, 8.82), followed by SGM and Hispanic/Latino (aOR=3.23, 95% CI=1.68, 6.19).

Discussion

This study examined differences in lifetime non-prescribed opioid use associated with racial/ethnic and SGM status and how these associations varied by sex among a representative U.S. sample of high school students in 2023. Likelihood of lifetime non-prescribed opioid use is greater among participants with intersectional identities of SGM status and race/ethnicity, including Hispanic/Latino, Black, and Other, compared to the likelihood of lifetime non-prescribed opioid use when examining siloed identities of racial/ethnic identity or SGM status. Additionally, results indicated heterogeneity of lifetime non-prescribed opioid use between sexes, particularly among heterosexual and SGM Hispanic girls and boys, with the greatest effect size found among SGM and racially/ethnic Other boys (compared to heterosexual White).

This study’s findings are broadly consistent with previous studies assessing disparities of lifetime non-prescribed opioid use among racial/ethnic and SGM youth,30–32 with the highest odds found among girls who are SGM Hispanic/Latino and Black and among boys who are SGM Hispanic/Latino and racially Other. Similar to this study’s findings among adolescents, adult SGM women have the greatest risk of lifetime non-prescribed opioid use when compared to men that are SGM as well as to heterosexual men and women.14,31,33–35 However, this finding is not uniform across all racial/ethnic categories of SGM girls. For instance, SGM girls that are either White or Hispanic/Latino have elevated odds of opioid use, when compared to their heterosexual White counterparts, and have a larger effect size compared to their boy counterparts; SGM boys who are either Black or racially Other have greater odds of opioid use, compared to their heterosexual White counterparts, and have a larger effect size compared to their girl counterparts.

Previous research has examined intersections of identity and substance use in adult populations by sex, race/ethnicity, and sexual identity,20 but lacked precision regarding the type of substance used (i.e. collapsing substances into illicit drug use/substance use disorders).36,37 However, with regard to adolescent opioid use, studies have only reported opioid-related outcomes through siloed identities of sex38, race/ethnicity38 and sexual identity.38,39 Focusing on opioid-related outcomes through siloed identities of race/ethnicity, sexual identity, and sex, effectively creates homogenous groups that lack context of the micro- and macro-level environmental risk factors of opioid use.23 This study highlights the complexity of the likelihood of opioid use that differs not just on race/ethnicity, SGM status, and/or sex, but that each intersectional identity has a unique relationship to opioid use. There is a dynamic interplay between social positioning and protective or risk factors associated with lifetime non-prescribed opioid use, which the present study illuminates.

Few studies have focused on opioid-related outcomes among SGM by examining differences by race/ethnicity and how these differences may vary by sex.40 The present findings indicate that that an intersectional framework does not presume an additive effect on lifetime non-prescribed opioid use simply due to multiple minority identities.41 Similar to recent findings examining the intersection in substance use disorders by sex, race/ethnicity, and sexual identity,20,40 this study’s findings indicate that girls who are heterosexual and Black showed no significant differences in lifetime non-prescribed opioid use compared to White heterosexual girls, suggesting that heterosexual Black girls may have developed effective coping mechanisms (i.e. resiliency that helps this population navigate minority-related stressors)42 to navigate the combined opioid-related-stressors of racism and sexism. Conversely, heterosexual Hispanic/Latina girls have an elevated likelihood of lifetime non-prescribed opioid use compared to their White counterparts, with significant increases in likelihood of opioid use among those who identify as SGM. Recent literature examining the intersection of Hispanic/Latino SGM adolescents reveals a gap in the literature regarding opioid use (and the mechanisms underlying the risk) for this population, of which this study offers critical and timely results.44 Minority stress and intersectional risk environment theories allow for a nuanced approach to the study’s findings that are specific to each set of intersecting identities. These findings suggest a potential protective factor for opioid use among heterosexual Black girls that may not be accessible to those who are SGM or for those who are heterosexual Black boys. Additionally, the heterogeneity of opioid use among Hispanic/Latina girls (both heterosexual and SGM) reveal multi-levels of intersectional minority stress, which may add to the increased risk of lifetime non-prescribed opioid use. Incorporating the complexity and nuance in which intersectional risk contributes to the heterogeneity of opioid-related outcomes can contribute to a more inclusive approach to intervention development.23

Minority stress theory and the intersectional risk environment framework can work in tandem with each other and provide nuanced approaches for interpreting varied non-prescribed opioid and heroin use risk factors and stressors of intersectional minority groups.17,23 These stressors are unique, chronic, socially-constructed, and conceptualized to include political/social policies (e.g., presence or absence of anti-discrimination laws based on sexual/gender identity), as well as social/interpersonal stressors (e.g., stigma and discrimination).15,45 The intersectional risk environment framework emphasizes drug- and health-related outcomes that is a result of processes, objects, and places that interact within specific social, historical, and geographic contexts and produce varied health-related related outcomes based on social positionality (e.g., sex, gender, sexuality, race/ethnicity).23 While people who identify as SGM share many experiences, particularly regarding stigma and discrimination on the basis of SGM status,17 SGM is not a homogenous group, with important variations existing within sub-categories (e.g., Hispanic/Latina lesbian women, Asian gay men, etc.,).15

Limitations

This study has several limitations. Data are limited to youth who attended school and are, therefore, not representative of all persons in this age group. The YRBS limits response options for sex (female/male, with a separate question for transgender identity) and doesn’t include gender identities such as nonbinary or genderqueer. Response options in the YRBS do not represent the diversity of sex and gender identities, to which each population experiences unique aspects of social positionality that may impact perceived stigma, discrimination, and social stress.17,42 As such, prevalence estimates for these populations cannot be assessed. Additionally, there is potential misclassification of transgender-identified individuals when analyses stratify by sex, potentially impacting results inferences47 Similarly, the exposure variable of sexual identity included those who are “not sure”, which might have captured data that was not specific to SGM. Racial and ethnic categories collapsed into one variable prohibited comparison of certain racial/ethnic identities. This may have masked differences between racial/ethnic groups at high risk of drug use. Future studies should recruit adequate sample sizes of populations underrepresented in research, particularly among Indigenous youth.12 Due to the potential of obscured findings related to collapsed SGM and race/ethnicity variables, findings should be interpreted with caution. YRBS data analysis is cross-sectional and thus cannot establish temporality or causality. Lastly, the YRBS does not capture fentanyl use, thus intended/unintended fentanyl use may lead to misclassification of lifetime non-prescribed opioid use.

Conclusion

In the present study, higher rates of lifetime non-prescribed opioid use was found among all race/ethnicity groups that identified as SGM, and these results varied by sex. This study’s findings indicate the largest association of lifetime non-prescribed opioid use among SGM girls are among those who are Hispanic/Latina, followed by Black. Interestingly, heterosexual Black boys reported higher opioid use compared to heterosexual White boys, whereas heterosexual Black girls were not significantly different from heterosexual White girls. Conversely, this study revealed multiple levels of intersectional stressors among Hispanic/Latina girls (both heterosexual and SGM) of increased lifetime non-prescribed opioid use. These findings highlight the heterogeneity within intersectional groups, potential resiliency factors among intersecting identities, and that each sex, race/ethnicity, and sexual identity needs to be examined through an intersectional lens. There may be certain risk factors among heterosexual Black boys that are different from heterosexual Black girls versus different intersectional risk factors among Hispanic/Latina girls that could be the target of future interventions.

Supplementary Material

1

Acknowledgments

This study was funded by NIH-NIDA grant R01DA053745 (Martins and Philbin, MPIs) and NIH-NIDA grant R01DA059376 (PI Martins). MEM is supported by NIH-NIDA grant R36DA061635. The present study’s sponsors had no role in study design, analysis, or interpretation of the data.

Footnotes

Conflict of interest statement:

The authors of this paper report no financial or non-financial conflicts of interest.

Financial disclosure:

No financial disclosures were reported by the authors of this paper

CRediT author statement

Martins and Gardner: Conceptualization; Data curation: Gardner; Formal analysis: Gardner, Martins; Funding acquisition: Martins; Methodology: Gardner, Martins, Kreski; Project administration: Martins; Resources: Martins; Supervision: Martins; Roles/Writing-original draft: Gardner, Martins; Roles/Writing-review and editing: Gardner, Martins, Kreski, Marziali, Davis.

Declaration of competing interests:

All authors (Gardner, Martins, Kreski, Marziali, and Davis) have no competing interests to declare.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Contributor Information

Sam D. Gardner, Columbia University Mailman School of Public Health, School of Social Work, Columbia University.

Megan E. Marziali, Columbia University Mailman School of Public Health.

Noah Kreski, Columbia University Mailman School of Public Health.

Alissa Davis, School of Social Work, Columbia University.

Silvia S. Martins, Columbia University Mailman School of Public Health.

6. References

  • 1.Substance Abuse and Mental Health Services Administration (SAMHSA). 2023 National Survey on Drug Use and Health (NSDUH) Detailed Tables: Section 5 – Personal Substance Use. 2024. https://www.samhsa.gov/data/sites/default/files/reports/rpt47100/NSDUHDetailedTabs2023/NSDUHDetailedTabs2023/2023-nsduh-detailed-tables-sect5pe.htm [Google Scholar]
  • 2.Hedegaard H, Miniño AM, & Spencer MR,. Drug Overdose Deaths in the United States, 2003–2023. 2024. NCHS Data Brief No 522. https://www.cdc.gov/nchs/products/databriefs/db522.htm#section_2 [PubMed] [Google Scholar]
  • 3.Mauro PM, Gutkind S, Annunziato EM, Samples H. Use of Medication for Opioid Use Disorder Among US Adolescents and Adults With Need for Opioid Treatment, 2019. JAMA Netw Open. Mar 1 2022;5(3):e223821. doi: 10.1001/jamanetworkopen.2022.3821 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Centers for Disease Control and Prevention National Center for Health Statistic. Data from: National Vital Statistics System, Mortality 2018–2023 on CDC WONDER Online Database. 2024. Atlanta, GA. [Google Scholar]
  • 5.Jiang X, Guy GP, Schmit K, et al. Substance use patterns and characteristics using real world data from adolescents assessed for substance use and treatment planning-united states, 2017–2021. Substance Use & Misuse. 2024;doi: 10.1080/10826084.2024.2383609 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Centers for Disease Control and Prevention (CDC). Youth Risk Behavior Survey Data Summary & Trends Report: 2013–2023 2024. 2024. https://www.cdc.gov/yrbs/dstr/index.html [Google Scholar]
  • 7.Patrick ME, O’Malley PM. The Epidemiology of Substance Use Among Adolescents in the United States. In: Zucker RA, Brown SA, eds. The Oxford Handbook of Adolescent Substance Abuse. Oxford University Press; 2019:0. [Google Scholar]
  • 8.Groenewald CB, Patel KV, Rabbitts JA, Palermo TM. Correlates and motivations of prescription opioid use among adolescents 12 to 17 years of age in the United States. Pain. 2020;161(4):742–748. doi: 10.1097/j.pain.0000000000001775 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Centers for Disease Control and Prevention (CDC). Youth Risk Behavior Survey: Data Summary & Trends Report 2011–2021. 2023. https://www.cdc.gov/healthyyouth/data/yrbs/pdf/YRBS_Data-Summary-Trends_Report2023_508.pdf [Google Scholar]
  • 10.Substance Abuse and Mental Health Services Administration (SAMHSA). 2022 National Survey on Drug Use and Health: Race and Ethnicity Highlights. 2023. https://www.samhsa.gov/data/sites/default/files/reports/rpt42731/2022-nsduh-race-eth-highlights.pdf [Google Scholar]
  • 11.Centers for Disease Control and Prevention (CDC). Youth Risk Behavior Survey Data Summary & Trends Report: 2013–2023. 2024. https://www.cdc.gov/yrbs/dstr/index.html#cdc_publication_summary_guidance-key-findings [Google Scholar]
  • 12.Ball A, Hadland S, Rodean J, Hall M, Mendoza J, Ahrens K. Trends in Substance-Related Visits Among Youth to US Children’s Hospitals, 2016–2021: An Analysis of the Pediatric Health Information System Database. J Adolesc Health. Jul 2024;75(1):76–84. doi: 10.1016/j.jadohealth.2024.02.016 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Panchal NZS. Teens, Drugs, and Overdose: Contrasting Pre-Pandemic and Current Trends. 2024. Octobe 15, 2024. Accessed 08/04/2025. https://www.kff.org/mental-health/issue-brief/teens-drugs-and-overdose-contrasting-pre-pandemic-and-current-trends/ [Google Scholar]
  • 14.McHugh RK, Nguyen MD, Chartoff EH, Sugarman DE, Greenfield SF. Gender differences in the prevalence of heroin and opioid analgesic misuse in the United States, 2015–2019. Drug Alcohol Depend. Oct 1 2021;227:108978. doi: 10.1016/j.drugalcdep.2021.108978 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Parent MC, Arriaga AS, Gobble T, Wille L. Stress and substance use among sexual and gender minority individuals across the lifespan. Neurobiol Stress. Feb 2019;10:100146. doi: 10.1016/j.ynstr.2018.100146 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Fernandez J, Gonzalez R, Oves JC Jr., Rodriguez P, Castro G, Barengo NC. Illicit Substance Use Disparities Among Lesbian, Gay, and Bisexual High School Students in the U.S. in 2017. Journal of Adolescent Health. 2021;68(6):1170–1175. doi: 10.1016/j.jadohealth.2021.02.004 [DOI] [PubMed] [Google Scholar]
  • 17.Meye IH. Prejudice, social stress, and mental health in lesbian, gay, and bisexual populations: conceptual issues and research evidence. Psychol Bull. Sep 2003;129(5):674–697. doi: 10.1037/0033-2909.129.5.674 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Paschen-Wolff MM, DeSousa A, Paine EA, Hughes TL, Campbell ANC. Experiences of and recommendations for LGBTQ+-affirming substance use services: an exploratory qualitative descriptive study with LGBTQ+ people who use opioids and other drugs. Substance Abuse Treatment, Prevention, and Policy. 2024/01/03 2024;19(1):2. doi: 10.1186/s13011-023-00581-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Underwood JM, Brener N, Thornton J, et al. Overview and Methods for the Youth Risk Behavior Surveillance System - United States, 2019. MMWR Suppl. Aug 21 2020;69(1):1–88. doi: 10.15585/mmwr.su6901a1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Schilt-Solberg MA, Blair LM, Kurzer JAMJ. Intersectionality in substance use disorders: Examining gender, race/ethnicity, and sexual orientation in the 2021–2022 National Survey on Drug Use and Health. Addictive Behaviors Reports. 2025/06/01/ 2025;21:100587. doi: 10.1016/j.abrep.2025.100587 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Brener ND, Mpofu JJ, Krause KH, et al. Overview and Methods for the Youth Risk Behavior Surveillance System - United States, 2023. MMWR Suppl. Oct 10 2024;73(4):1–12. doi: 10.15585/mmwr.su7304a1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Centers for Disease Control and Prevention (CDC). 2023. National Youth Risk Behavior Survey Data User’s Guide. https://www.cdc.gov/yrbs/media/pdf/2023/2023_National_YRBS_Data_Users_Guide508.pdf [Google Scholar]
  • 23.Collins AB, Boyd J, Cooper HLF, McNeil R. The intersectional risk environment of people who use drugs. Social Science & Medicine. 2019/08/01/ 2019;234:112384. doi: 10.1016/j.socscimed.2019.112384 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Suarez NA, Trujillo L, McKinnon II, et al. Disparities in School Connectedness, Unstable Housing, Experiences of Violence, Mental Health, and Suicidal Thoughts and Behaviors Among Transgender and Cisgender High School Students - Youth Risk Behavior Survey, United States, 2023. MMWR Suppl. Oct 10 2024;73(4):50–58. doi: 10.15585/mmwr.su7304a6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Miller-Jacobs C, Operario D, Hughto JMW. State-Level Policies and Health Outcomes in U.S. Transgender Adolescents: Findings from the 2019 Youth Risk Behavior Survey. LGBT Health. Aug-Sep 2023;10(6):447–455. doi: 10.1089/lgbt.2022.0247 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Li DH, Turner BC, Mustanski B, Phillips GL 2nd. Sexual orientation disparities in prescription drug misuse among a nationally representative sample of adolescents: Prevalence and correlates. Addict Behav. Feb 2018;77:143–151. doi: 10.1016/j.addbeh.2017.09.021 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.SAS Institute Inc. SURVEYLOGISTIC Procedure. https://support.sas.com/documentation/onlinedoc/stat/142/surveylogistic.pdf [Google Scholar]
  • 28.Centers for Disease Control and Prevention. 2019. YRBS Analysis Software. https://www.cdc.gov/healthyyouth/data/yrbs/pdf/2019/2019_yrbs_analysis_software.pdf [Google Scholar]
  • 29.SAS Institute Inc. SAS/STAT® 14.2 User’s Guide. SAS Institute Inc.; 2016. [Google Scholar]
  • 30.Jones AA, Schneider KE, Brighthaupt SC, Johnson JK, Linton SL, Johnson RM. Heroin and nonmedical prescription opioid use among high school students in urban school districts. Drug Alcohol Depend. Dec 1 2019;205:107664. doi: 10.1016/j.drugalcdep.2019.107664 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Klare DL, McCabe SE, Ford JA, Schepis TS. Prescription drug misuse, other substance use, and sexual identity: The significance of educational status and psychological distress in US young adults. Subst Abus. 2021;42(3):377–387. doi: 10.1080/08897077.2020.1784358 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Demant D, Hides L, Kavanagh DJ, White KM, Winstock AR, Ferris J. Differences in substance use between sexual orientations in a multi-country sample: findings from the Global Drug Survey 2015. J Public Health (Oxf). Sep 1 2017;39(3):532–541. doi: 10.1093/pubmed/fdw069 [DOI] [PubMed] [Google Scholar]
  • 33.Capistrant BD, Nakash O. Lesbian, Gay, and Bisexual Adults have Higher Prevalence of Illicit Opioid Use than Heterosexual Adults: Evidence from the National Survey on Drug Use and Health, 2015–2017. LGBT Health. Aug/Sep 2019;6(6):326–330. doi: 10.1089/lgbt.2019.0060 [DOI] [PubMed] [Google Scholar]
  • 34.Ploderl M, Tremblay P. Mental health of sexual minorities. A systematic review. Int Rev Psychiatry. 2015;27(5):367–85. doi: 10.3109/09540261.2015.1083949 [DOI] [PubMed] [Google Scholar]
  • 35.Diaz M, Tenney C, Boyd KE, Ford JA. Sexual Identity and Motivations for Prescription Drug Misuse Among U.S. Adults. LGBT Health. Feb-Mar 2021;8(2):107–115. doi: 10.1089/lgbt.2020.0294 [DOI] [PubMed] [Google Scholar]
  • 36.Schuler MS, Rice CE, Evans-Polce RJ, Collins RL. Disparities in substance use behaviors and disorders among adult sexual minorities by age, gender, and sexual identity. Drug Alcohol Depend. Aug 1 2018;189:139–146. doi: 10.1016/j.drugalcdep.2018.05.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Rodriguez-Seijas C, Eaton NR, Pachankis JE. Prevalence of psychiatric disorders at the intersection of race and sexual orientation: Results from the National Epidemiologic Survey of Alcohol and Related Conditions-III. J Consult Clin Psychol. Apr 2019;87(4):321–331. doi: 10.1037/ccp0000377 [DOI] [PubMed] [Google Scholar]
  • 38.Hoots BE, Li J, Hertz MF, et al. Alcohol and Other Substance Use Before and During the COVID-19 Pandemic Among High School Students - Youth Risk Behavior Survey, United States, 2021. MMWR Suppl. Apr 28 2023;72(1):84–92. doi: 10.15585/mmwr.su7201a10 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Pitzer L, Bennett M, Simard B, Schillo BA, Vallone DM, Hair EC. Prescription Opioid Misuse: Examining the Role of Opioid-Related Attitudes among Youth and Young Adults by Sexual Orientation. Subst Use Misuse. 2020;55(10):1601–1609. doi: 10.1080/10826084.2020.1753774 [DOI] [PubMed] [Google Scholar]
  • 40.Paschen-Wolff MM, Kidd JD, Paine EA. The State of the Research on Opioid Outcomes Among Lesbian, Gay, Bisexual, Transgender, Queer, and Other Sexuality- and Gender-Diverse Populations: A Scoping Review. LGBT Health. Jan 2023;10(1):1–17. doi: 10.1089/lgbt.2022.0036 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Bowleg L The problem with the phrase women and minorities: intersectionality-an important theoretical framework for public health. Am J Public Health. Jul 2012;102(7):1267–73. doi: 10.2105/AJPH.2012.300750 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Meyer IH. Identity, Stress, and Resilience in Lesbians, Gay Men, and Bisexuals of Color. Couns Psychol. Apr 1 2010;38(3)doi: 10.1177/0011000009351601 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Substance Abuse and Mental Health Services Administration (SAMHSA). 2023 National Survey on Drug Use and Health (NSDUH) Detailed Tables - Section 1: Illicit Drug Use/Misuse Tables. 2024. Accessed 08/04/2025. https://www.samhsa.gov/data/report/2023-nsduh-detailed-tables [Google Scholar]
  • 44.Rosales R, Zelaya DG, Moreno O, et al. Latinx Sexual Minority Adolescent Substance Use: State of the Science and Call for Intersectional Minority Stressors and Protective Factors. Curr Addict Rep. Sep 2023;10(3):396–411. doi: 10.1007/s40429-023-00503-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Janssen T, Gamarel KE, Mereish EH, Colby SM, Haikalis M, Jackson KM. Associations Among Enacted Stigma, Perceived Chances for Success, Life Satisfaction, and Substance Use Among Sexual Minority and Heterosexual Youth. Subst Use Misuse. 2023;58(9):1121–1131. doi: 10.1080/10826084.2023.2212282 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Kumar R VO, Saadabadi A. Buprenorphine. StatPearls. StatPearls Publishing; 2024. [PubMed] [Google Scholar]
  • 47.Restar AJ, Lett E, Menezes NP, et al. Getting precise about gender and sex measurement: a primer for epidemiologists. Am J Epidemiol. Dec 2 2024;193(12):1861–1867. doi: 10.1093/aje/kwae144 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS2127254-supplement-1.pdf (122KB, pdf)

RESOURCES