Abstract
OBJECTIVES: Sexual minority youth face health disparities and a high risk of substance abuse. This study sought to fully characterize the disparity of tobacco use and risk factors in this high-risk subpopulation.
METHODS: Using data from the 2015 Youth Risk Behavior Survey, the current use of various tobacco products (cigarettes, smokeless tobacco, cigars, and e-cigarettes) was compared according to sex and distinct sexual identities (lesbian, gay, bisexual, and unsure). Heterosexual/straight adolescents served as the control group.
RESULTS: Of 14 703 respondents, 88.8% were heterosexual/straight, 2.0% were lesbian or gay, 6.0% were bisexual, and 3.2% were unsure about their sexual identity. Sexual minorities had a higher prevalence of tobacco product use than their heterosexual/straight counterparts. Sex had a significant effect on the disparities of tobacco use. Lesbian and bisexual girls had higher odds of reporting current use of any tobacco product, cigarettes, cigars, and e-cigarettes than did straight girls, whereas sexual minority boys had similar smoking behaviors compared with straight boys. Substance use, including marijuana use, drinking, and binge drinking, was significantly associated with any tobacco use.
CONCLUSIONS: Tobacco use differs among subgroups of sexual minority youth, with lesbians and bisexual girls having a higher prevalence of tobacco use than their straight peers. Heterogeneity of tobacco use across distinct sexual identity groups underscores the need to develop evidence-based tobacco control strategies for sexual minority youth.
- aOR —
- adjusted odds ratio
- LGBT —
- lesbian, gay, bisexual, and transgender
- YRBS —
- Youth Risk Behavior Surveys
What’s Known on This Subject:
Adult studies have shown that lesbian, gay, bisexual, and transgender minorities face health disparities compared with their nonminority counterparts. Few nationally representative surveys have provided information on cigarette and other tobacco product use among sexual minority adolescents.
What This Study Adds:
Sexual minority youth, especially lesbians and bisexual girls, had a higher prevalence of tobacco use than their straight peers. For sexual minorities, high prevalence of tobacco use might start in youth, and the risk behaviors continue to develop into adulthood.
Evidence suggests that sexual minority adults, including lesbian, gay, bisexual, and transgender (LGBT) individuals, smoke at substantially higher rates than their nonminority counterparts.1–4 In addition to cigarette smoking, sexual minority adults also use other tobacco products such as hookah, cigars, and e-cigarettes at a higher prevalence than straight populations.2,5 Furthermore, disparities have been identified among these distinct groups, with sexual minority women and bisexual adults having the highest risk of smoking.2,3,6 The disparities in tobacco use are driven by certain factors, including stress due to social stigma and discrimination,7,8 and are further exacerbated by aggressive marketing from the tobacco industry and limited access to effective tobacco treatment.7,9–11 It is important to identify smoking patterns and social correlates at an early stage of life for this priority population because risky health behaviors are often established during adolescence, with 9 of 10 smokers having tried their first cigarette by age 18 years.12 Several studies have shown that LGBT youth or young adults had a higher prevalence of cigarette smoking than straight adolescents.3,13–16 For example, Austin et al13 found that the odds of smoking among lesbian and bisexual youth were 5.3 times higher than for their straight counterparts. In a Canadian study, Azagba et al16 reported that LGB adolescents were 50% to 168% more likely to smoke daily compared with non-LGB adolescents. These studies have provided evidence for researchers to advocate recognizing sexual minority communities as a priority population for tobacco prevention and cessation services.
Despite progress in addressing health disparities among sexual minorities, the data collection on sexual minority groups from national population–based samples has been limited,17,18 especially for adolescents. Most national surveys addressing tobacco use in youth do not collect information on sexual orientation, whereas localized studies, often with small sample sizes and regional differences, can yield results that might not be representative of the nation at large.4 At the same time, treating adolescent LGB as a unified population, as some current studies have done, might provide inaccurate measurements of LGB tobacco use. A few adult studies have parsed the LGB category to assess smoking prevalence according to distinct sexual identities,2,6,13 and they have found that female sexual minority adults are at higher risk of smoking than their male counterparts. However, little is known about how patterns of tobacco use differ among distinct groups of LGB adolescents at the national level. In addition, limited studies have examined LGB tobacco use beyond cigarette smoking. Johnson et al2 reported a higher prevalence of hookah and e-cigarette use among LGB adults than among straight populations. Rath et al19 reported high rates of ever-use of other tobacco products (ie, cigars, little cigars, e-cigarettes, dissolvable tobacco products, hookah) among LGB young adults (18–34 years of age) than among non-LGB young adults. However, few national studies have explored the use of alternative tobacco products according to sexual identity among adolescents. Such knowledge is critical because cigarette use among youth has declined in recent years, and the decrease in cigarette smoking attained over the last several decades is now offset by the increasing popularity of alternative tobacco products such as e-cigarettes.20–23 Furthermore, the association between youth cigarette smoking and other substance use, such as marijuana use and drinking,19,24,25 needs to be evaluated specifically among sexual minority youth.
To address the gaps in knowledge, the present study examined the differences in use of various tobacco products according to sex and sexual identity by using a nationally representative sample from the 2015 Youth Risk Behavior Survey (YRBS). This analysis specifically sought to: (1) assess the prevalence of use of cigarettes, smokeless tobacco (including chewing tobacco, snuff, and dip), cigars (including cigars, cigarillos, and little cigars), and e-cigarettes according to sex and distinct sexual identity groups; and (2) examine factors associated with tobacco use within distinct groups of sexual minorities.
Methods
Data
The data were collected from the 2015 YRBS, a national study conducted biennially in odd-numbered years since 1991. The surveys collect a wide range of priority health-risk behaviors among a nationally representative sample of US students in grades 9 through 12. The national YRBS uses a 3-stage, cluster sample design; a detailed description of the YRBS survey design and data collection can be found on the YRBS Web site.26,27 YRBS protected student privacy by allowing for anonymous participation and by following local parental permission procedures. Students completed a self-administered questionnaire during 1 class period with no physical test or examination involved.28 The 2015 national YRBS had a school response rate of 69% and a student response rate of 80%; the overall response rate was 60%. The study included a total of 15 624 respondents.
Measures
Sexual Minority Populations
The YRBS includes questions measuring 2 elements of sexual orientation: sexual identity and sex of sexual contacts. Because there was discordance between sexual identity and sex of sexual contacts,18 and the sexual identity question yielded a larger sample size compared with the question of sex of sexual contacts,29 the sexual identity question was used in this study to identify sexual minority youth.
Sexual identity was ascertained by using the following item from the 2015 YRBS: “Which of the following best describe you?” The response options included “Heterosexual (straight),” “Gay or lesbian,” “Bisexual,” and “Not sure.” A total of 921 students were excluded who did not complete this question, resulting in 14 703 respondents in the final study.
Tobacco Use
Current tobacco use was assessed by using the following 4 questions: “During the past 30 days, on how many days did you smoke cigarettes?”; “During the past 30 days, on how many days did you use chewing tobacco, snuff, or dip, such as Red Man, Levi Garrett, Beech-Nut, Skoal, Skoal Bandits, or Copenhagen?”; “During the past 30 days, on how many days did you smoke cigars, cigarillos, or little cigars?”; and “During the past 30 days, on how many days did you use an electronic vapor product?” Response options included “0 days,” “1 or 2 days,” “3 to 5 days,” “6 to 9 days,” “10 to 19 days,” “20 to 29 days,” and “all 30 days.” Students who responded >0 days were classified as current cigarette users, current smokeless tobacco users, current cigar users, or current e-cigarette users, respectively. Students who reported using any of these 4 categories of tobacco products in the past 30 days were classified as current any tobacco users.
Covariates
To control for other influences, several covariates were included in the analysis, including sex (male or female), race/ethnicity (non-Hispanic white, non-Hispanic black, Hispanic, and non-Hispanic other), age (<15 or ≥15 years), and grade (grades 9–10 and 11–12). Current marijuana users were responders who answered “≥1 time” to the question “During the past 30 days, how many times did you use marijuana?” Current drinking status was defined if responders answered “≥1 day” to the question “During the past 30 days, on how many days did you have at least one drink of alcohol?” Current binge drinking status was defined if responders answered “≥1 day” to the question “During the past 30 days, on how many days did you have 5 or more drinks of alcohol in a row, that is, within a couple of hours?”
Statistical Methods
Summary statistics of demographic characteristics and substance use according to sexual identity were reported and compared by using the Rao-Scott χ2 test. Sampling weights and survey stratum were included in all analyses to account for the complex survey design. Weighted prevalence estimates of current use of tobacco products, including any tobacco, cigarettes, smokeless tobacco, cigars, and e-cigarettes, were estimated according to sex and sexual identity. Logistic regression models were performed to compare risks of tobacco use according to sex and sexual identity and were adjusted for race, age, and grade in the multivariate analysis. Finally, separate regression analyses were performed among heterosexual, gay/lesbian, bisexual, and unsure groups to assess the associations between sex, race, age, other substance use, and current use of any tobacco products. Statistical analyses were performed by using SAS version 9.4 (SAS Institute, Inc, Cary, NC), and a P value <.05 was considered statistically significant.
Results
Among the 14 703 respondents in the analysis, 12 954 (88.8%) were heterosexual/straight, 324 (2.0%) were gay/lesbian, 922 (6.0%) were bisexual, and 503 (3.2%) were not sure about their sexual identity. The distribution of demographic characteristics and other substance use according to sexual identity is listed in Table 1. Bisexual adolescents were much more likely to be female than male (79.6% vs 20.4%) compared with other groups. The distribution of sexual identity varied according to race (P = .001), with non-Hispanic black adolescents having a relatively higher proportion of gay/lesbian subjects. Gay/lesbian (33.1%) and bisexual (31.7%) adolescents had a higher prevalence of marijuana use than their straight (20.7%) and unsure (26.0%) peers. Bisexual adolescents had the highest prevalence of drinking (41.8%) among distinct sexual identity groups.
Sample Characteristics According to Sexual Identity, 2015 YRBS (N = 14 703)
Table 2 presents the prevalence of current use of various tobacco products according to sex and sexual identity. Overall, sexual minority adolescents had a higher prevalence of reporting tobacco use than did their straight counterparts. About 40.5% of gay/lesbian and 38.5% of bisexual adolescents reporting use of any tobacco product in the last 30 days compared with 29.6% of heterosexual/straight students and 32.2% of unsure adolescents (P < .0001). Gay/lesbian adolescents and unsure adolescents had a higher prevalence of reporting use of all 4 tobacco products than did straight students. Bisexual adolescents had a higher prevalence of reporting use of cigarettes, cigars, and e-cigarettes but a lower prevalence of reporting use of smokeless tobacco products than did their straight peers.
Prevalence of Tobacco Use According to Sex and Sexual Identity, 2015 YRBS
A strong sex interaction with disparities of tobacco use was observed across sexual identity groups. Lesbian and bisexual girls had a higher prevalence of reporting current use of any tobacco products, cigarettes, cigars, and e-cigarettes than did straight girls, whereas sexual minority boys had similar smoking behaviors compared with straight boys. For heterosexual/straight and unsure adolescents, boys had a higher prevalence of reporting use of any tobacco and all 4 tobacco products than did girls. For gay/lesbians and bisexual students, girls had a higher prevalence of reporting use of any tobacco product, cigarettes, and e-cigarettes but a lower prevalence of reporting use of smokeless tobacco and cigars than did boys.
The risks of use of various tobacco products were compared among sexual identity groups (Table 3). For female adolescents, compared with their heterosexual/straight counterparts, lesbians had higher odds of reporting use of any tobacco product (adjusted odds ratio [aOR]: 2.3; P = .002), cigarettes (aOR: 2.9; P = .004), cigars (aOR: 3.2; P = .003), and e-cigarettes (aOR: 1.8; P = .045); bisexual adolescents had higher odds of reporting use of any tobacco product (aOR: 2.2; P < .0001), cigarettes (aOR: 3.7; P < .0001), cigars (aOR: 2.1; P < .0001), and e-cigarettes (aOR: 1.9; P < .0001); and unsure adolescents had higher odds of reporting use of cigars (aOR: 1.8; P = .028). For male adolescents, gay boys had lower odds of reporting use of smokeless tobacco (aOR: 0.1; P = .001) and unsure adolescents had higher odds of reporting use of cigars (aOR: 1.9; P = .044) than did their straight counterparts.
Multivariable Logistic Regression Comparing Risks of Tobacco Use According to Sex and Sexual Identity
Within each sexual identity group, factors associated with any tobacco use were assessed further (Table 4). Boys had higher odds of reporting use of any tobacco product than did girls (aOR: 1.6; P < .0001) among straight adolescents, but they were not significantly different from girls among sexual minority groups. Non-Hispanic black adolescents had lower odds of reporting use of any tobacco product than did non-Hispanic white adolescents (aOR: 0.6; P < .0001) among straight youth, but they were not significantly different from each other among sexual minorities. Use of marijuana, drinking, and binge drinking increased the odds of any tobacco use among all groups. However, the effects were different among straight adolescents and distinct sexual minority groups. For example, binge drinking (compared with no drinking) increased the odds of any tobacco use by nearly 13 times (aOR: 13.6; P < .0001) for gay/lesbian youth, but the effect dropped by more than one-half for bisexual adolescents (aOR: 5.6; P < .0001).
Multivariable Logistic Regression of Factors Associated With Any Tobacco Use Among Heterosexual and Sexual Minority Populations
Discussions
The health outcomes of sexual minorities are understudied. The Institute of Medicine called for population-based studies to better understand LGBT health.30 The present study used the first wave of the nationally representative sample from YRBS to assess the use of various tobacco products according to sex and distinct sexual identity. The findings confirm the disparities of cigarette smoking among sexual minority youth and further extend our understanding of these disparities to alternative tobacco products. A growing body of literature has examined the association between sexual orientation and smoking cigarettes.1–4,6,7,13–15,31–34 However, most of the previous studies limited their analyses to cigarette smoking or focused on adults. The present study examined differences in use of various tobacco products among sexual minority youth. Overall, lesbian/gay adolescents had the highest rate of current use of any tobacco product (40.5%), followed by bisexual adolescents (38.5%) and unsure adolescents (32.2%) compared with their straight peers (29.6%). The association between sexual orientation and tobacco use could be due to social stigma, social isolation, family rejection, discrimination, and emotional abuse.13 Because sexual minority youth face a wide range of health disparities, the findings in this study could provide evidence for public health agencies and health care organizations to formulate culturally competent tobacco prevention programs and interventions targeting LGBT youth. For instance, the Centers for Disease Control and Prevention’s “Tips from Former Smokers” campaign featured real stories from LGBT individuals to advocate smoking cessation among the sexual minority community.35 The American Lung Association has called on governmental agencies, health care systems, LGBT health advocates, and community members to develop a combination of policy change, prevention messaging campaigns, and tobacco cessation services for this vulnerable population.4
As cigarette use among youth has declined in recent years, the tobacco industry has been promoting alternative tobacco products to entice a new generation of young users. The present study found that sexual minority youth have complex patterns of various tobacco use and higher risks of alternative tobacco use than their straight peers. E-cigarettes, in particular, are gaining popularity among youth.21 To the best of my knowledge, the present study is the first national trial to report a higher prevalence of current use of e-cigarettes among lesbian (30.7%) and bisexual (30.0%) adolescents compared with straight adolescents (23.4%). Because the use of e-cigarettes has outpaced the use of traditional cigarettes among teenagers,20–23 these findings have implications in future e-cigarette research to better understand the initiation and frequency of e-cigarette use among sexual minority youth.
Previous studies of sexual minority adults have identified tobacco use disparities according to sex and sexual identity and found that sexual minority women, especially bisexual women, had the highest smoking prevalence compared with other groups.2,6 The present study assessed tobacco use according to sex and sexual identity among adolescents and similarly found that both lesbian and bisexual girls had higher odds of reporting the current use of any tobacco product, cigarettes, cigars, and e-cigarettes than did straight girls. In contrast, sexual minority boys were far more similar in smoking behaviors compared with straight boys. These findings, taken together with those of previous studies, suggest that a high prevalence of tobacco use starts at an early age among sexual minority girls, and the risk behaviors continue to develop into adulthood. The heterogeneity in tobacco use according to sex and sexual identity underscores the need to identify risk factors that are associated with the disparities of tobacco use among sexual minorities intragroup (lesbian/gay vs bisexual versus unsure) and consider between-sex differences in future studies of LGBT youth.
By analyzing factors associated with any tobacco use within each sexual identity group, substance use was found to be significantly associated with any tobacco use among all 4 sexual identity groups. Disparities in associations were reported between drinking, binge drinking, and any tobacco use among sexual minorities and their straight peers. For example, binge drinking increased the risk of any tobacco use >2 times higher for gay/lesbian youth than for bisexual youth. These findings again suggest the heterogeneity of factors associated with tobacco use among straight adolescents and distinct sexual minority groups. A one-size-fits-all approach might not work for sexual minority youth, and evidence-based strategy and tailored inventions are needed to counter the high risk of substance use among this population. Parents, health practitioners, and school educators should be aware of these risk factors associated with tobacco use among sexual minority youth and educate them about the danger of tobacco use.
These findings are subject to several limitations. First, the 2015 YRBS data are cross-sectional; thus, causal inferences could not be established. Second, because sexual identity was used as the measurement, and the study was not designed to identify transgender adolescents, the data are not inclusive of all LGBT adolescents. The inclusion of gender measures on national surveys has been limited, and enhancements for measurement methods for identifying transgender adolescents are critically needed. Third, the YRBS data are school-based surveys of representative samples of ninth- through 12th-grade students. Excluding high school dropouts, home-schooled students, and other youth not in school might lead to a sampling bias, which could be more pronounced for sexual minority youth because they are more likely to drop out of school and have poorer school attendance than their straight peers. Future research is needed to further evaluate tobacco use among sexual minority youth who are not represented in a school-based survey. Finally, some estimates of the prevalence of tobacco use also had wide confidence intervals due to the small sample size, and thus these estimates should be interpreted with appropriate caution.
Conclusions
Despite these aforementioned limitations, this population study examined differences in use of various tobacco products across LGB sexual minority youth and assessed the factors associated with tobacco use according to sex and sexual identity. Heterogeneity in use of various tobacco products was identified according to sex and sexual identity. Lesbians and bisexual adolescents had higher odds of using cigarettes, cigars, and e-cigarettes than their nonsexual minority counterparts. These findings underscore the need to develop evidence-based tobacco control policies for sexual minority youth that consider the heterogeneity within the sexual minority population.
Acknowledgments
The author thanks the Medical Writing Center at Children’s Mercy Hospital for editing the manuscript.
Footnotes
- Accepted January 3, 2017.
- Address correspondence to Hongying Dai, PhD, Health Services & Outcomes Research, Children’s Mercy Hospital, 2401 Gillham Rd, Kansas City, MO 64108. E-mail: hdai{at}cmh.edu
FINANCIAL DISCLOSURE: The author has indicated she has no financial relationships relevant to this article to disclose.
FUNDING: No external funding.
POTENTIAL CONFLICT OF INTEREST: The author has indicated she has no potential conflicts of interest to disclose.
References
- Copyright © 2017 by the American Academy of Pediatrics
