Minority stressors and tobacco use among a US sample of sexual and gender minority young adults

Abstract

Introduction:

Young adults who are sexual and gender minorities (SGM) are at the highest risk for tobacco initiation in young adulthood. Minority stress theory suggests that sexual orientation and gender identity (SOGI)-based discrimination may contribute to nicotine and tobacco use disparities. Our study aimed to quantify the association between SOGI-based distal minority stressors and current tobacco use among SGM young adults living in the United States (US).

Methods:

Eligible participants—including young adults (aged 18-35 years old), who identified as SGM, and were currently residing in the US (N=1,116) —were recruited via Prolific into an online survey. We applied stepwise binary regressions with backward selection to model the association between average past 30-day distal minority stress and current tobacco use (i.e., combustible cigarettes or e-cigarettes), controlling for perceived stress and sociodemographic covariates. We also tested interactions between minority stress and SGM status. Exploratory analyses assessed associations between minority stress and current tobacco use among YA, stratified by SGM subgroup.

Results:

A 1-unit increase in experiencing minority stress in the past 30-days was associated with 1.02 greater odds of current tobacco use among SGM young adults. No difference between SGM subgroups in this association was found. Examining stratified SGM subgroups, a 1-unit increase in minority stress was associated with 1.11 greater odds of current tobacco among transgender adults only.

Conclusion:

Distal minority stress is differentially associated with current tobacco use for transgender young adults, which suggests that tobacco prevention and cessation interventions may need tailoring for subgroups.

1. Introduction

1.1. Prevalence of tobacco use by sexual orientation and gender identity

In the United States (US), tobacco use remains the leading cause of premature and preventable death (National Center for Chronic Disease Prevention and Health Promotion (US) Office on Smoking and Health, 2014) among adults. Over time, the average age of cigarette smoking initiation has shifted from adolescence to young adulthood, making interventions to reduce tobacco use targeted to this population especially salient (Barrington-Trimis et al., 2020). This is especially important for sexual minorities (i.e., lesbian, gay, bisexual, and other non-heterosexual people) who experience disparities in combustible and e-cigarette use as they transition into young adulthood (Krueger et al., 2020). Young adult sexual minority women (SMW) are more at risk, as they report higher rates of combustible cigarette (32.3-44.3%) and e-cigarette use (6.3-8.9%) than heterosexual young adult women (18.2% combustible and 3.3% e-cigarette). For young adult sexual minority men (SMM), differences in combustible cigarette smoking (27.2-35.8%) and e-cigarette use (6.5-8.7%) are less stark, but still elevated compared to heterosexual men (27.9% combustible and 7.2% e-cigarette) (Wheldon et al., 2018). These studies also suggests intragroup variance across subgroups of young adult SMW and SMM, such that bisexual women and men demonstrate the highest rates of combustible cigarette smoking, while e-cigarette use is elevated among bisexual women (Wheldon et al., 2018). Less is known about tobacco use prevalence among transgender and gender diverse (TGD) populations, as prior studies demonstrate inconsistent findings. A analysis of Population Assessment of Tobacco and Health (PATH) Wave 4 data found that TGD individuals were 2-3 times more likely than cisgender individuals to report current cigarette and e-cigarette use (Sawyer et al., 2022). Yet, a similar study using PATH Wave 2 data found that cigarette and e-cigarette prevalence rates were statistically equivalent between transgender and cisgender populations (Wheldon and Wiseman, 2019). For both studies, TGD individuals were combined into a single analytic group, such that examining differences in the prevalence of tobacco use between transgender and other gender diverse individuals (i.e., non-binary, agender, genderqueer) was not possible.

1.2. Factors that influence tobacco use among sexual and gender minorities

Several sociodemographic risk factors have been linked to higher tobacco use among sexual minorities and TGD young adults (sexual and gender minorities; SGM) including urbanicity and being of lower income, lower education, or a racialized minority (Bennett et al., 2015; Budenz et al., 2022; Sawyer et al., 2022). Additionally, individual-level psychosocial factors, including higher rates of depression, anxiety, and increased use of other substances associated with tobacco co-use (e.g., drinking alcohol) are risk factors for increased tobacco use among SGM individuals (Blosnich et al., 2013). Researchers have also posited that minority stress, or the excess stress experienced by SGM people due to experiencing sexual orientation and gender identity (SOGI)-based stigma and discrimination, is a risk factor for elevated tobacco use among SGM young adults.

1.3. Minority stress as a framework to understand tobacco use among sexual and gender minorities

Minority stress theory posits that SGM individuals experience unique minority stressors beyond the daily stressors experienced by the general population, and these minority stressors contribute to risky health behaviors, including tobacco use (Meyer and Frost, 2012). Minority stressors may be distal (i.e., experiences that occur externally, such as transphobic harassment) or proximal (i.e., internal processes, including internalized homophobia). Previous research has found that distal minority stressors, including explicit threats and harassment, are independently associated with higher odds of current cigarette smoking among SGM (Gordon et al., 2021). Among TGD populations, distal minority stressors including housing, employment, and healthcare discrimination are also associated with tobacco use (Wolford-Clevenger et al., 2022).

1.3.1. Gap in research

Although previous work has found an association between minority stress and use of tobacco among SGM generally, the strength of this association within subgroups of SGM people (e.g., at the intersection of sexual orientation and gender identity) is not well understood. Moreover, the existing literature aggregates transgender and nonbinary individuals into a single analytic group, masking any differences in the association of minority stress and tobacco use between these subpopulations. This is problematic as scholars have theorized that individuals who do not identify in gender binary terms (i.e., nonbinary people) experience unique minority stressors compared to cisgender sexual minority people and transgender people who identify within the gender binary (Lefevor et al., 2019). To our knowledge, no published studies have investigated the association between minority stress and tobacco use among subpopulations of nonbinary and transgender individuals. However, there is evidence that transgender and nonbinary adults experience differences in minority stressors and mental health outcomes (Nowaskie et al., 2023; Reisner and Hughto, 2019; Scandurra et al., 2019). It is, thus, possible that the association between minority stress and tobacco use, a behavior qualitatively described as a strategy to cope with general and minority stress by TGD people (Hinds et al., 2022; Tan et al., 2021), also differs between transgender and nonbinary young adults. The current study aimed at filling this gap in the literature by purposefully examining the association between distal minority stress and tobacco use among transgender and non-binary young adults separately.

1.3.2. Purpose of study

The aim of this study was to quantify the relationship between minority stress and current tobacco use among a sample of SGM young adults, by sexual orientation and gender identity.

2. Methods

This analysis was part of a parent study to identify tobacco public education messages that effectively communicate the risks of combustible cigarettes and e-cigarettes to SGM young adults. Participants were recruited for an online survey experiment via Prolific (https://www.prolific.co/), a platform for online subject recruitment for research. Prolific offers access to over 38,000 US nationals; approximately 1/3 of whom report ages < 35 years old, which makes Prolific an ideal pool for recruiting young adult survey participants. Prolific is also a viable method for recruiting hard-to-reach populations for survey research, including SGM people (Palan and Schitter, 2018). We purposively sampled participants per race and ethnicity estimates per the 2020 US Census (Wyatt and Drozd, 2021) and oversampled for SGM participants, people who smoke cigarettes, and people who use e-cigarettes.

Prescreening was conducted via Prolific. Eligible participants were young adults (aged 18-35 years old) who were currently residing in the US. Individuals who met eligibility criteria were directed to an online consent form. Those who consented (N=2,857) were directed to an online survey administered via Qualtrics for which they received $4.50 via Prolific for participating. For this secondary analysis, participants had to identify as SGM; thus, our final analytic sample was N = 1,116.

2.1. Measures

Current tobacco use

Our main outcome of interest was current tobacco use, which was defined as using either combustible cigarettes or e-cigarettes. Participants were asked if they had “ever smoked a combustible tobacco cigarette, even just one puff? (for e-cigarettes: ever used a nicotine vape or e-cigarette, even one time”). Those responding, “Yes” were asked: “Do you currently smoke cigarettes (for e-cigarettes: e-cigarettes or electronic nicotine vapes) everyday, some days, or not at all?” Participants reporting never use of combustible or e-cigarettes, and those reporting ever but not current use of either product, were coded 0 (not current tobacco use). Those reporting using at least one product every day or some days were coded 1 (current tobacco use).

Minority stress

Our main predictor of interest, minority stress, was adapted from the 8-item measure the Daily Sexual Minority Stressors in Lesbian Women Scale (all 8-items were used in the current study). The original measure was intended to capture daily minority stressors and had the following prompt “For each of the following statements rate how much this experience describes something that happened to you because you identify as a sexual minority woman.” We adapted it as follows, “For each of the following statements, rate how much this experience describes something that happened to you in the past 30 days because you identify as LGBTQ+.” The time period for self-reported minority stress experiences aligned with our general stress measure (self-reported past 30-day perceived stress). Items were coded 0 = Not at all to 7 = very much and included: 1) being verbally harassed, 2) being told you are overreacting regarding sexual minority issues, 3) having someone respond defensively when their heterosexist language is pointed out, 4) hearing others make fun of sexual minority people, 5) having someone laugh or make jokes due to your LGBTQ+ status, 6) hearing anti-LGBTQ+ talk, 7) explicitly being threatened with harm as a result of their LGBTQ+ status, and 8) perceiving a situation, individual, or environment to be unsafe because of their sexual minority identity (α = 0.86). An overall average minority stress score (Range: 0-56) was used in the main models.

Perceived stress

To control for the effect of general stress experienced by individuals (i.e., parsed out from specific SGM identity-based stress as measured via the minority stress scale), we included a 4-item measure of perceived stress in our models (Cohen et al., 1983). The prompt was as follows: “The questions in this scale ask you about your feelings and thoughts during the last month. In each case, please indicate how often you felt or thought a certain way.” Items included not being able to control important things in their life, confidence about handling personal problems, general feeling of things going their way, and feeling that difficulties were piling up so high that they could not overcome (alpha = 0.82).

Gender and sexual orientation

We applied a two-step method to assess current gender identity (Lagos and Compton, 2021). Participants were asked to report their sex assigned at birth and their current gender identity. Responses were recoded to create a categorical measure of gender identity, representing cisgender, transgender, and nonbinary groups.

Participants reported their sexual identity as heterosexual, gay/lesbian, bisexual, or another non-heterosexual orientation. Because we characterized our analytical groups at the intersection of sexual orientation and gender identity, the small subsample of sexual minority participants who endorsed a sexual identity other than gay/lesbian or bisexual were not included in our final analytical sample (for example, among cisgender men there were only 23 individuals who endorsed a non-heterosexual orientation other than gay or bisexual).

Measures of gender and sexual identity were used to define subpopulations of SGM young adults and create a six-level categorical variable as follows: cisgender gay men, cisgender bisexual men, cisgender lesbian women, cisgender bisexual women, transgender (any sexual orientation), and nonbinary (any sexual orientation).

Sociodemographic characteristics

Finally, we assessed sociodemographic variables associated with tobacco use including urbanicity, race, and education (Blosnich et al., 2013). For urbanicity, participants self-reported whether they lived in a large city, suburb near a large city, small city or town, or rural area. Due to small subgroup sample sizes of racial and ethnic minoritized participants, race/ethnicity was binary coded as non-Hispanic White or BIPOC (Black, Indigenous, and People of Color, including Hispanic/Latinx participants). Education was binary coded as having attained a 4-year college degree or higher, or not having attained a 4-year college degree. Urbanicity was binary coded as large city/suburb near a large city and small city/rural area (see Table 1 for all possible choices presented to participants for our sociodemographic variables).

Table 1.

Demographics by sexual orientation and gender identity

		Total sample (N= 1116)		Cisgender gay male (n = 119, 10.7%)		Cisgender bisexual male (n = 192 ,17.2 %)		Cisgender lesbian female (n=101, 9.0%)		Cisgender bisexual female (n=476, 42.6%)		Transgender (n = 77, 6.9 %)		Non-binary (n =151, 13.5%)
Variable		Mean / n	SD / %	Mean / n	SD / %	Mean / n	SD/ %	Mean / n	SD / %	Mean / n	SD/ %	Mean / n	SD / %	Mean / n	SD / %	F / χ2
Race
	White	709	63.5	74	62.2	99	51.6	70	69.3	312	65.5	49	63.6	105	69.5	52.23**
	Black	130	11.6	16	13.4	44	20.7	14	13.9	42	8.8	7	9.1	7	4.6
	Asian	81	7.3	10	8.4	17	8.9	5	5.0	37	7.8	3	3.9	9	6.0
	Hispanic	144	12.9	15	12.6	28	14.6	7	6.9	63	13.2	11	14.3	20	13.2
	Another racea	52	4.7	4	3.4	4	2.1	5	5.0	22	4.6	7	9.1	10	6.6
Education	Less than HS	22	2.0	2	1.7	1	0.5	0	0.0	10	2.1	4	5.2	5	3.3	42.79**
	HS or GED	224	20.1	16	13.4	50	26.0	14	13.9	101	21.2	18	23.4	25	16.6
	Technical college or some college	398	35.7	41	34.5	56	29.2	37	36.6	170	35.7	35	45.5	59	39.1
	Bachelors	386	34.6	45	37.8	71	37.0	47	46.5	158	33.2	18	23.4	47	31.1
	Doctoral or Professional degree	86	7.7	15	12.6	14	7.3	3	3.0	37	7.8	2	2.6	15	9.9
Urbanicity	Large city	388	31.4	41	34.5	66	34.4	43	42.6	133	27.9	20	26.0	54	35.8	21.85
	Suburb near large city	416	33.6	49	41.2	63	32.8	29	28.7	161	33.8	26	33.8	46	30.5
	Small town	317	25.6	23	19.3	50	26.0	21	20.8	128	26.9	23	29.9	40	26.5
	Rural area	115	9.3	6	5.0	13	6.8	8	7.9	54	11.3	8	10.4	11	7.3
Current use tobacco	Yes	510	45.7	57	47.9	92	47.9	43	42.6	233	48.9	30	39.0	55	36.4	9.68
	No	606	54.3	62	52.1	100	52.1	58	57.4	243	51.1	47	61.0	96	63.6
Minority stress b		12.52	12.07	11.95	12.74	11.87	12.16	12.21	12.08	10.06	10.52	21.77	12.64	16.83	12.53	18.33**
Perceived stress c		7.85	3.68	7.44	3.93	8.11	3.49	8.45	3.16	8.73	3.58	9.65	3.40	9.05	3.72	5.21**

^a.Other races included Middle Eastern, Native American/ Alaskan Native and Pacific Islander,

^**p < 0.01;

^b.Range 0-56;

^c.Range 0-16

2.2. Analyses

We calculated descriptive statistics for all covariates, by sexual orientation and gender identity. We examined differences among our demographic variables using chi-square tests for categorical variables and one-way ANOVAs for continuous variables. We also used post-hoc comparison tests with Bonferroni corrections to examine specific differences by SGM subgroup. We then conducted stepwise binary logistic regression models with backward selection to assess the association between minority stress and current tobacco use among SGM young adults, controlling for perceived stress and sociodemographic covariates (all models were adjusted). Minority stress, perceived stress, SGM subgroup, race, education, and urbanicity were entered in the first step, and interactions were entered in the second step. The following interactions were entered in the model: SGM subgroup x minority stress, SGM subgroup x perceived stress, SGM subgroup x race, SGM subgroup x education, and SGM subgroup x urbanicity. Using backward selection, we removed non-significant covariates and covariate interactions leaving only SGM subgroup x education in the final model. To interpret any significant interactions, we used the following equation: OR_int = OR_education * OR_{SGM subgroup} * OR_{education x SGM subgroup}. The referent groups are as follows: for SGM group the referent is gay males, for education the referent is individuals with a 4-year college degree. As exploratory analyses, we ran binary logistic regressions stratified by SGM identity to examine if minority stress was associated with tobacco use for subgroups of SGM young adults, controlling for perceived stress and sociodemographic variables. To control for Type 1 error, we applied a Bonferroni-corrected α = 0.008. To be parsimonious, we only report significant findings in our exploratory analyses. Hosmer and Lemeshow’s test indicated that all final models had good model fit (p >.05). For all final models, both omnibus chi-square and Nagelkerke R² are reported as the variance explained in current tobacco use.

3. Results

3.1. Sociodemographic characteristics and variables of interest

As shown in Table 1, most participants were white (63.5%), bisexual female (42.6%) or bisexual male (17.2%), and cisgender (79.6%). Most did not have a 4-year degree (57.8%) and lived in a large city or suburb near a large city (65.0%). With regards to race, statistically significant differences by SGM subgroup emerged, χ² = 52.23 (20), p < 0.01, with non-binary individuals (69.5%) and bisexual females (65.5%) more likely to be white compared to cisgender bisexual males (51.6%), while cisgender bisexual males were more likely to be Black (22.9%) compared to non-binary (4.6%) and bisexual females (8.8%). With regards to education, χ² = 42.79 (20), p < 0.01, fewer transgender individuals reported having a bachelor’s degree (23.4%) compared to cisgender lesbian females (46.5%). There were no statistically significant differences with regards to urbanicity or current use of tobacco. There was a significant difference in average minority stress scores across the SGM subgroups, F (5, 1110) = 18.33, p < 0.001. Transgender individuals (M = 21.77) reported the greatest minority stress experiences, followed by non-binary individuals (M = 16.83). Finally, there was a statistically significant difference in perceived stress among SGM subgroups, F (5, 1110) = 5.21, p < 0.001. Cisgender gay males reported lower perceived stress (M = 7.44) compared to cisgender bisexual females (M = 8.45) transgender individuals (M = 9.65), and non-binary individuals (M = 9.05). Bisexual males also reported lower perceived stress (M = 8.11) compared to transgender individuals.

3.2. Associations between minority stress and current tobacco use

As seen in Table 2, there was a statistically significant association between minority stress on current tobacco use among our sample. A 1-unit increase in past 30-day minority stress score was associated with a 1.02 increase in the odds of current tobacco use (aOR = 1.02, 95% CI: 1.01, 1.03). With regards to perceived stress, a 1-unit increase in perceived stress score was associated with a 1.04 increase in the odds of current tobacco use (aOR = 1.04; 95% CI: 1.01, 1.08). As seen in Table 2, the association between minority stress, perceived stress and current tobacco use held with and without (step 1) the interactions in the model.

Table 2.

Binary logistic regression of current tobacco use among SGM young adults living in the US (N = 1,116)

Variables in the model
	Step 1		95% CI for OR			Step 2		95% CI for OR
	B	SE	OR	LL	UL	B	SE	OR	LL	UL
Minority stress	0.02	0.01	1.02**	1.01	1.03	0.02	0.01	1.02**	1.01	1.03
Perceived stress	0.04	0.02	1.04*	1.01	1.08	0.04	0.02	1.04*	1.01	1.08
Education
4-year degree or higher	Ref					Ref
no 4-year degree	−0.25	0.13	0.78*	0.60	0.99	−0.16	0.37	0.85	0.41	1.77
Race
Non-BIPOC	Ref
BIPOC	−0.16	0.13	0.85	0.66	1.09
Urbanicity
Large city/suburb	Ref
Small city/rural	−0.06	0.13	0.94	0.73	1.22
SGM group
Gay male	Ref					Ref
Bi male	−0.02	0.24	0.98	0.61	1.57	0.08	0.34	1.09	0.56	2.12
Lesbian	−0.28	0.28	0.75	0.44	1.30	−0.42	0.39	0.66	0.31	1.42
Bi female	0.01	0.21	1.01	0.67	1.54	−0.57	0.30	0.56	0.31	1.02
Transgender	−0.75	0.31	0.47*	0.26	0.87	−0.01	0.53	0.99	0.35	2.78
Non-binary	−0.70	0.26	0.49**	0.30	0.82	−0.73	0.38	0.48	0.23	1.01
SGM x education
Gay male						Ref
Bi male						−0.178	0.48	0.84	0.33	2.13
Lesbian						0.25	0.55	1.29	0.44	3.81
Bi female						1.04	0.42	2.83*	1.24	6.46
Transgender						0.90	0.65	0.41	0.11	1.47
Non-binary						0.12	0.51	1.13	0.41	3.09
Omnibus	44.31**					65.39**
R2	0.05					0.08

^*p < 0.05,

^**p < 0.01

Although there was no significant interaction between minority stress and SGM subgroup, as seen in table 2, step 1 indicated that transgender individuals (aOR = 0.47, 95% CI: 0.26, 0.87) and non-binary individuals (aOR = 0.49, 95% CI: 0.30, 0.82) had lower odds of currently using tobacco compared to cisgender gay males. Additionally, there was a significant interaction between education and SGM subgroup (Wald = 33.34, p < 0.001). Cisgender bisexual females who had a 4-year education or higher had 1.35 higher relative odds of current tobacco use than cisgender gay males with a 4-year degree or higher.

3.4. Exploratory analyses to examine the association between past 30-day minority stress and tobacco use, stratified by SGM subgroup

After controlling for Type 1 error, the only SGM subgroup with statistically significant results was transgender individuals (p<.001). As seen in Table 3, exploratory analyses found that, among transgender young adults, a 1-unit increase in past 30-day minority stress score was associated with a 1.11 increase in the odds of reporting current tobacco use, controlling for perceived stress and sociodemographic variables (aOR = 1.11; 95% CI: 1.05, 1.17).

Table 3.

Binary logistic regression of current tobacco use among transgender individuals

			Transgender (n = 77)
Variables in the model			95% CI for OR
	B	SE	OR	LL	UL
Minority stress	0.10	0.03	1.11**	1.05	1.17
Perceived stress	−0.07	0.09	0.93	0.78	1.12
Race/Ethnicitya
NH White	Ref
BIPOC	−0.33	0.62	0.72	0.21	2.42
Education
4-year degree or higher	Ref
no 4-year degree	−1.27	0.64	0.28*	0.08	0.99
Urbanicity
Large city/suburb	Ref
small city/rural	−1.04	0.60	0.35	0.11	1.13
Omnibus χ2	25.77**
R2	0.39

^a.BIPOC includes: Black, Asian, Hispanic Middle Eastern, Native American/ Alaskan Native and Pacific Islander and mixed race

^*p< 0.05,

^**p<0.01

4. Discussion

Sexual and gender minority (SGM) young adults have higher rates of tobacco use (combustible cigarette and e-cigarette) compared to their cisgender heterosexual counterparts. Given that tobacco smoking remains the leading cause of premature death in the US, examining factors that may lead to increased use of tobacco in this population is crucial.

Our findings suggest that there is an association between experiencing distal minority stress (i.e., discrimination experiences) and current use of tobacco among SGM young adults. However, the relationship between minority stress and current tobacco use was not moderated by SGM subgroup. Exploratory analyses examining the association between minority stress and tobacco use among groups stratified by SGM status suggested that increased minority stress was associated with greater odds of currently using tobacco only for transgender young adults. This is different from previous studies examining minority stress and tobacco use that have found a positive association across all SGM (Bariola et al., 2016; Gordon et al., 2021; Wolford-Clevenger et al., 2022). However, these findings are especially salient since transgender young adults in our sample reported the highest mean minority stress scores (transgender: M = 21.77; SD: 12.64).

In the 2015 US Transgender Survey, a large non-representative survey of TGD adults, 46% of participants experienced verbal harassment in the past year due to their transgender identity and 9% were physically attacked due to their gender (James et al., 2016). More generally, transgender individuals are 2.5 times more likely to experience violent victimization compared to cisgender individuals (Truman and Morgan, 2022). Given that transgender individuals (both in general and in our study) report higher rates of minority stress than cisgender individuals, interventions aimed at reducing tobacco use among transgender individuals should focus on creating safer climates for transgender people. These may include nondiscrimination policy initiatives and their enactment, alongside interventions to bolster resiliency and connectedness among this group. Additionally, future studies examining the role of minority stress on tobacco use among SGM should employ the use of ecological momentary assessment (EMA) techniques to determine if more tobacco is used on days where they experience greater minority stress (such as physical harm or feeling unsafe due to their SGM status).

Previous research investigating the association between minority stress and lifetime cigarette use has pooled subgroups of sexual minorities. For example, in a nationally representative study of White, Black and Latino/a sexual minority adults (N = 1,500), there was a positive association between daily discrimination and lifetime use of cigarettes for all sexual minority women (Gordon et al., 2021). These results were not examined among subgroups, so we do not know if these results hold for cisgender bisexual women, cisgender lesbian women, or other sexual minority women. Qualitative studies that contextualize quantitative findings may illuminate the reasons by which minority stress is a more salient risk factor for tobacco use among subgroups of SGM young adults. To quantify these associations more thoroughly, prospective studies measuring minority stress and tobacco use should oversample for SGM subpopulations.

Some limitations of the current study include the use a survey platform (Prolific) that requires users to present photo identification (ID) to verify their identity, which could present a barrier for transgender and non-binary individuals from signing up for the service if their name and gender markers do not match those on their government issued IDs. This is especially important as approximately 34% of transgender people do not have identification that lists their correct gender marker (Herman and O’Neill, 2021). Additionally, since the parent study oversampled for tobacco use, prevalence rates of e-cigarette and cigarette use reported in this sample are not representative of prevalence rates in the US population of SGM young adults. However, by oversampling this group, we were able to examine associations between minority stress and tobacco use for SGM subgroups, which is a strength of this paper. Our work adds to the previous literature examining minority stress and tobacco use among SGM in that it further disaggregates SGM groups at the intersection of sexual orientation and gender identity and explores the unique experiences of transgender and non-binary individuals.

This study offers an important contribution to literature examining how minority stress is associated with current use of tobacco (combustible and e-cigarette) among SGM young adults. Our results point at opportunities for tailoring tobacco prevention and cessation interventions for SGM young adults most at-risk for cigarette and e-cigarette use. This work is important given the current political climate, in which federal hate crimes legislation has not been sufficient to stem the tide of violence against TGD people (HRC Foundation, n.d.) and a record-breaking number of anti-SGM bills have been introduced in state legislatures across the country (Peele, 2023). In 2022, when this data was collected, 315 anti-SGM bills were introduced in state legislatures and while only 10% were enacted as law (HRC Foundation, 2023), incidents of hate crimes and violence against SGM people have increased (Anti-Defamation League, 2023; Jones and Kishi, 2022). If tobacco use among SGM young adults is influenced, in part, by minority stress, as our findings and others suggest (Bariola et al., 2016; Gordon et al., 2021; Wolford-Clevenger et al., 2022), then efforts to promote health equity in tobacco prevention and control must also address the policy environment to discourage anti-SGM legislation and to advocate for state-level protections for SGM people. Finally, research examining the role of minority stress in smoking cessation interventions should continue to oversample for and disaggregate SGM subgroups, particularly TGD individuals, as our research indicates the association between minority stress experiences and tobacco use is not universal in this population.

Implications.

This study details the influence of minority stress on current tobacco use among sexual and gender minority (SGM) young adults. Findings underscore the need for targeted and tailored approaches to tobacco control, wherein SGM young adults most at-risk are engaged in cessation interventions that address minority stress as a contributing factor to tobacco use and which support their resilience. To promote health equity, tobacco control must address the contexts that engender minority stress. Assessment of policy impacts on SGM tobacco use and the effectiveness of interventions disseminated within SGM-supportive and discriminatory policy environments are important next steps.

Data availability statement

Data and code is available by request from the PI (JGP).