You are here

  • Home
  • Archive
  • Volume 30, Issue 1
  • Inequities in tobacco advertising exposure among young adult sexual, racial and ethnic minorities: examining intersectionality of sexual orientation with race and ethnicity

Article Text

Article menu

Download PDFPDF

Original research
Inequities in tobacco advertising exposure among young adult sexual, racial and ethnic minorities: examining intersectionality of sexual orientation with race and ethnicity
  1. Andy S L Tan1,2,
  2. Elaine P Hanby1,
  3. Ashley Sanders-Jackson3,
  4. Stella Lee1,2,
  5. Kasisomayajula Viswanath1,2,
  6. Jennifer Potter4,5
  1. 1 Department of Medical Oncology, Division of Population Sciences, Center for Community-Based Research, Dana-Farber Cancer Institute, Boston, Massachusetts, USA
  2. 2 Department of Social and Behavioral Sciences, Harvard T.H. Chan School of Public Health, Boston, Massachusetts, USA
  3. 3 Department of Advertising and Public Relations, College of Communication Arts and Science, Michigan State University, Lansing, Michigan, USA
  4. 4 The Fenway Institute at Fenway Health, Boston, Massachusetts, USA
  5. 5 Department of Medicine, Beth Israel Lahey Health, Boston, Massachusetts, USA
  1. Correspondence to Dr Andy S L Tan, Department of Medical Oncology, Division of Population Sciences, Center for Community-Based Research, Dana-Farber Cancer Institute, Boston, MA 02215, USA; andy_tan{at}dfci.harvard.edu

Abstract

Objective This study examined sexual orientation differences in encoded exposure to tobacco product ads and intersections with race and ethnicity.

Methods We analysed data from young adults (18–24) from the US Population Assessment of Tobacco and Health Study in 2013 and 2014 (N=9110). First, we compared encoded exposure to cigarette, electronic cigarette (e-cigarette), cigar and smokeless tobacco ads between sexual minorities (lesbian/gay, bisexual and something else) versus heterosexual young adults. We then analysed encoded ad exposure across sexual orientation, racial and ethnic subgroups. Analyses controlled for demographic and tobacco use variables.

Results Bisexual women had significantly higher prevalence of encoded exposure to cigarette and cigar ads compared with heterosexual women, and significantly higher prevalence of encoded e-cigarette ad exposure compared with both heterosexual and lesbian/gay women. There were no significant differences in encoded ad exposure between lesbian versus heterosexual women and between gay or bisexual men versus heterosexual men. Compared with heterosexual white counterparts, increased encoded ad exposures were reported by heterosexual black women (cigarette and cigar ads), black heterosexual men (cigar ads) and bisexual black women (cigarette and cigar ads). Compared with heterosexual non- Hispanic counterparts, increased encoded ad exposures were reported by bisexual Hispanic women (cigarette, e-cigarette and cigar ads) and heterosexual Hispanic men (cigarettes and cigar ads).

Conclusion Sexual minority women of colour and black heterosexual women and men have increased encoded exposure to certain forms of tobacco ads. Further research is needed to address the impact of tobacco ads among multiple minority individuals based on sex, sexual orientation, race and ethnicity.

  • priority/special populations
  • advertising and promotion
  • disparities
  • non-cigarette tobacco products
  • electronic nicotine delivery devices

https://doi.org/10.1136/tobaccocontrol-2019-055313

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    Sexual minorities are at increased risk of developing tobacco-related cardiovascular disease and acute respiratory illness due to higher prevalence of cigarette smoking and other forms of tobacco use compared with heterosexuals.1 2 Although there are no published data comparing tobacco-related cancer incidence between sexual minorities and heterosexual populations,3 sexual minorities are potentially at higher risk of developing tobacco-related cancers as tobacco use is the leading preventable cause of cancer and cancer deaths.4 Sexual minority young adults are particularly susceptible to initiating tobacco use due to social stigma, experiences of minority stress, norms of tobacco use in lesbian, gay, bisexual, transgender, and queer/questioning (LGBTQ) culture and processes associated with coming out.5–9 Historical macrolevel systems of oppression and marginalisation leading to pervasive racism, sexism and heterosexism, tobacco-related social norms, and the tobacco industry’s targeted marketing among sexual, racial and ethnic minorities are additional important contributory factors.9–15

    This study focuses on tobacco product ads exposure as a risk factor associated with disparities in tobacco use among sexual minority young adults because of the tobacco industry’s history of targeted marketing among sexual, racial and ethnic minorities and young adults.16–21 In addition, young adults are particularly vulnerable and receptive to tobacco product marketing, such as in bars and nightclubs.22 23 Based on data from the Washington State Behavioural Risk Factor Surveillance System (BRFSS) from 2003 to 2007, lesbian and bisexual women and gay men reported higher exposure to tobacco marketing16 (defined as owning items with a tobacco brand or logo, receiving tobacco samples and coupons, or attending tobacco industry-sponsored events and promotions) compared with heterosexuals. Exposure to marketing was associated with higher prevalence of current smoking.16 In another study using the Washington State BRFSS data (2003–2006), lesbian and bisexual women were more receptive to tobacco marketing (defined as willingness to use or wear tobacco industry merchandise) than heterosexual women.18 In comparison, gay and bisexual men were equally receptive to tobacco marketing compared with heterosexual men.18 In addition, newer forms of tobacco products, including electronic cigarettes (e-cigarettes), flavoured cigars and smokeless tobacco, are marketed online and through social media and are increasingly popular among young adult sexual minorities.24–33 Similarly, the tobacco industry has a long history of targeted marketing of tobacco products, including menthol cigarettes at black, Hispanic and low-income communities. The industry's marketing include outdoor, print and point-of-sale ads, and providing financial support to elected officials, organisations, minority-owned businesses and cultural events.19–21 34–36 African–Americans have higher exposure to tobacco ads37 and higher rates of menthol cigarette use compared with other racial and ethnic groups.38

    There is limited research describing tobacco-related health disparities among sexual minority people of colour and the intersecting influences of sexual orientation, race and ethnicity. This gap is partly due to the lack of large enough sample sizes of sexual minorities who are also from racial and ethnic minority backgrounds to conduct such analyses and studies. In a national survey of college students, Blosnich and colleagues found that black, Hispanic, Asian and multiracial sexual minorities (gay/lesbian or bisexual) reported higher cigarette smoking prevalence compared with heterosexuals from their respective racial groups.39 Black and Asian lesbian, gay and bisexual (LGB) students had lower rates of smoking cigarettes, cigars and hookah, while black LGB students had higher rates of smoking cigar/little cigars/clove cigarettes compared with white LGB students.39 Similarly, Ortiz and colleagues compared smoking prevalence among a national convenience sample of sexual minority adults and found that black and Asian sexual minorities reported lower smoking prevalence compared with white sexual minorities.40 However, these studies did not examine the intersecting influences of sexual orientation, race and ethnicity on smoking tobacco. To address this gap, Corliss and colleagues analysed interactions between sexual orientation and race/ethnicity in association with smoking behaviours among adolescents and reported that disparities in cigarette smoking between sexual minorities and heterosexuals were accentuated among black and Asian Pacific Islander gay and lesbian youth compared with disparities among white youth.41 The previously mentioned research studies suggest complex intersections of sexual orientation, race and ethnicity in influencing tobacco disparities.

    This study aimed to compare encoded exposure to ads of different tobacco products, including cigarettes, e-cigarettes, cigars and smokeless tobacco. Encoded exposure to tobacco ads is defined as having a minimal memory trace of encountering these ads.42 We analysed data from a nationally representative survey of US young adult sexual minorities and their heterosexual counterparts. We further examined the variations in encoded exposure to tobacco product ads among sexual minorities stratified by race and ethnicity to explore the association between these intersecting identities on encoded ad exposure. Results from this study may be used to identify young adult sexual minorities of colour who are at increased risk of exposure to tobacco ads and to inform regulations and interventions to mitigate adverse effects of tobacco marketing exposure on tobacco-related health disparities.

    Methods

    Sample

    Data for the current analysis were from the young adult subsample aged 18–24 years (N=9110 total, n=4496 women and n=4614 men) who participated in wave 1 of the US Population Assessment of Tobacco and Health (PATH) Study between September 2013 and December 2014.43 The weighted response rates for the household screener and adult survey in wave 1 were 54% and 74%, respectively. We used the wave 1 data of the PATH Study because measures relevant to the current analysis on tobacco product ads encoded exposure were asked in wave 1 only. Interviews were conducted using computer-assisted self-interviewing in participants’ homes. Additional details about the sample design are reported elsewhere.44

    Measures

    ​Sexual orientation

    Participants were categorised as heterosexual, lesbian or gay, bisexual or something else and were stratified by sex. Participants were asked to report their sex with the question ‘What is your sex?’ Response options were male or female. Sexual identity was determined with the question ‘Do you think of yourself as (1) lesbian or gay; (2) straight, that is, not lesbian or gay; (3) bisexual; or (4) something else?’ There were 169 participants who answered ‘something else’ and 108 participants who left this question unanswered. These participants were prompted with follow-up questions to clarify their response. Participants who indicated an alternative term (eg, queer/trisexual/omnisexual/pansexual), have not yet figured out their sexuality, do not think of themselves having a sexuality, do not use labels to identify themselves or describe themselves as something else other than gay, lesbian, bisexual or heterosexual were categorised as ‘something else’. Those who responded ‘don’t know’ or refused to answer were categorised as missing (n=101, 1.1%). We were not able to ascertain the participants’ gender identity.

    ​Tobacco product ads encoded exposure

    Participants aged 18–24 years were shown a series of 20 tobacco ads, randomly selected from a pool of tobacco ads, one at a time. The pool of tobacco ads comprised actual ads that had appeared in various media (print, television or web). At the start of wave 1 of the PATH Study, there were of 698 ads. During wave 1, the pool of ads was updated with 261 new ads, while 500 of the original ads were retired, resulting in a total of 959 unique ads used. There were four categories of tobacco product ads (cigarette, e-cigarette, cigar and smokeless tobacco) representing different brands, and sub-brands were included in these questions. Each participant saw five cigarette ads (one Marlboro ad, one Camel ad, one Newport ad and two ads of other cigarette brands), five e-cigarette ads (two television ads and three non-television ads across different brands), five cigar ads (one large cigar ad and four non-large cigar ads across different brands), and five smokeless tobacco ads (two snus ads, two oral dip ads and one chew tobacco ad across different brands). Following each tobacco ad, participants were asked, ‘In the past 12 months, have you seen this advertisement before this study?’ Participants’ responses to the 20 questions on tobacco product ads were recoded into four measures of encoded exposure: (1) recall of at least one cigarette ad, (2) recall of at least one e-cigarette ad, (3) recall of at least one cigar ad and (4) recall of at least one smokeless tobacco ad. Participants who responded ‘don’t know’ or refused to answer items for ads across these four tobacco products were coded as missing (48–88 cases across these four products, 0.53%–0.97%). Additional details of the ads used in the survey are available in the PATH Study documentation.44

    ​Demographic variables

    Participants’ race was categorised as white, black or other race (derived categorical variable). Ethnicity was categorised as Hispanic or non-Hispanic. Education was categorised as no high school diploma, high school graduate, some college (no degree) or associate degree, bachelor’s degree or advanced degree. There was no missing data on these variables.

    ​Tobacco use variables

    Participants’ use of cigarettes was categorised as (1) never smoked, (2) smoked every day, (3) smoked on some days or (4) ever smoked but do not smoke currently. E-cigarette use was categorised as (1) never vaped, (2) vaped every day, (3) vaped on some days or (4) ever vaped but do vape currently. Due to the lower prevalence of current use of other tobacco products among participants, cigar, pipe, hookah and smokeless tobacco use were categorised as ever use versus never use of each of these products. Participants who responded ‘don’t know’ or refused to answer items for using these tobacco products were coded as missing (1–19 cases across these products, 0.01%–0.21%).

    Analyses

    All analyses were conducted using STATA V.14. We conducted analyses stratified by sex rather than grouping sexual minority females and males together because of prior research showing different patterns of tobacco use disparities across female and male sexual minorities compared with their heterosexual counterparts.24 We used bivariate and multiple logistic regression models predicting encoded exposure to ads of each tobacco product category with sexual orientation. Analyses were stratified by sex, adjusting for demographic variables (race, ethnicity and highest education) and tobacco product use, and estimates were weighted to represent the US young adult subpopulation aged 18–24 years using the study replicate weights and the ‘svy’ programme in STATA. Variances were estimated using the balanced repeated replication method with Fay’s adjustment as recommended in the PATH Study user guide.43 We examined each model for potential multicollinearity using recommended guidelines.45 The proportion of missing data was under 2.7% across all the regression models, and we present the regression results using the complete case analysis approach. We obtained the marginal prevalence of encoded exposure to ads of each tobacco product using multiple regression estimates and the ‘margins’ programme in STATA in order to compare across sexual orientation category among women and men while adjusting for the demographic and tobacco use variables. We next analysed the intersections between sexual orientation with race and between sexual orientation and ethnicity in separate models by including interaction terms (sexual orientation×race and sexual orientation×ethnicity) and controlling for covariates (demographic variables and tobacco product use). We compared the marginal prevalence of encoded exposure to tobacco product ads for each sexual orientation by race category (vs heterosexual white) and for each sexual orientation by ethnicity category (vs heterosexual non-Hispanic) among women and men using the ‘pwcompare’ subcommand and adjusted with Bonferroni corrections for multiple comparisons.

    Results

    Table 1 summarises the weighted characteristics of the analytic sample of young adults. Among the participants, 1% identified as female and lesbian/gay, 4% as female and bisexual, and 1% as female and other sexual identities. Approximately 1% identified as male and gay, 1% as male and bisexual, and 1% as male and other sexual identities. The majority of young adults were white (70%); 15% were black; and 15% identified with other races. Most participants were non-Hispanic (79%). About 43% had some college education. About half of the participants (47%) never smoked cigarettes, while 15% smoked daily; 11% smoked on some days; and 28% ever smoked and did not smoke currently. Most (68%) never vaped e-cigarettes; 1% vaped daily; 8% vaped on some days; and 23% ever vaped and did not vape currently. About one in four (23%) have ever used cigars/cigarillos/filtered cigars; 13% ever used pipe tobacco; 44% ever used hookah; and 17% ever used smokeless tobacco.

    View this table:
    Table 1

    Demographic characteristics

    Encoded exposure to tobacco ads among young adult women

    Figure 1 presents the marginal prevalence of encoded exposure to tobacco ads across the four products by sexual orientation among women, adjusting for demographic and tobacco use variables. Bisexual women had higher prevalence of encoded exposure to cigarette, e-cigarette and cigar tobacco ads (52%, 51% and 26%, respectively) compared with heterosexual women (43%, 40% and 19%, respectively). Bisexual women had higher prevalence of encoded e-cigarette ad exposure (51%) compared with lesbian women (37%). Adjusting for covariates, bisexual women had higher odds of encoded exposure to cigarette ads (OR=1.44 (95% CI 1.12 to 1.84)), e-cigarette ads (OR=1.57 (95% CI 1.21 to 2.03)) and cigar ads (OR=1.50 (95% CI 1.11 to 2.01)) in the past 12 months compared with heterosexual women (online supplementary appendix 1). Bisexual women also had higher odds of encoded e-cigarette ad exposure versus lesbian/gay women (OR=1.70 (95% CI 1.03 to 2.83)) (online supplementary appendix 2).

    Figure 1
    Figure 1

    Weighted prevalence of encoded tobacco product ad exposure by sexual orientation among US adult women aged 18–24 years (2013–2014) (unweighted, n=4496).

    Figure 2 presents the marginal prevalence of encoded exposure to tobacco ads across the four products by sexual orientation and race among women, adjusting for demographic and tobacco use variables. Compared with heterosexual white women, heterosexual black women had higher encoded exposure to cigarette and cigar ads; bisexual black women had higher encoded exposure to cigarette and cigar ads; heterosexual other races had lower encoded exposure to e-cigarette ads; women who identified as something else and other races had lower encoded exposure to cigar ads; heterosexual black, heterosexual other races and lesbian black women had lower encoded exposure to smokeless tobacco ads. Online supplementary appendix 3 presents the analyses testing the interactions between sexual orientation and race among women. There was one significant interaction between sexual orientation and race for encoded exposure to smokeless ads.

    Figure 2
    Figure 2

    Weighted prevalence of encoded tobacco ad exposure by sexual orientation and race among US adult women aged 18–24 years (2013–2014) (unweighted, n=4496).

    Figure 3 presents the marginal prevalence of encoded exposure to tobacco ads across the four products by sexual orientation and ethnicity among women, adjusting for demographic and tobacco use variables. Compared with heterosexual non-Hispanic women, bisexual Hispanic women had higher encoded exposure to cigarette, e-cigarette and cigar ads. Online supplementary appendix 4 presents the analyses testing the interactions between sexual orientation and ethnicity among women. There were two significant interactions between sexual orientation and ethnicity for encoded exposure to cigar and smokeless ads.

    Figure 3
    Figure 3

    Weighted prevalence of encoded tobacco ad exposure by sexual orientation and ethnicity among US adult women aged 18–24 years (2013–2014) (unweighted, n=4496).

    Encoded exposure to tobacco ads among young adult men

    Figure 4 presents the marginal prevalence of encoded exposure to tobacco product ads among men, adjusting for demographic and tobacco use variables. Gay men had higher encoded cigarette and e-cigarette ads exposure (39% and 50%, respectively) compared with men who identified as something else (22% and 24%, respectively). Heterosexual men had higher encoded cigarette and e-cigarette ads exposure (35% and 41%, respectively) compared with men who identified as something else (22% and 24%, respectively). Controlling for covariates, we found that there were no significant differences in encoded exposure to tobacco ads between gay or bisexual men versus heterosexual men. Men who identified as something else had lower odds of encoded exposure to cigarette ads (OR=0.52 (95% CI 0.27 to 0.98)) and e-cigarette ads (OR=0.47 (95% CI 0.25 to 0.89)) versus heterosexual men (online supplementary appendix 5), and lower odds of encoded exposure to cigarette ads (OR=0.45 (95% CI 0.20 to 0.97)) and e-cigarette ads (OR=0.33 (95% CI 0.15 to 0.73)) versus gay men (online supplementary appendix 6).

    Figure 4
    Figure 4

    Weighted prevalence of encoded tobacco product ad exposure by sexual orientation among US adult men aged 18–24 years (2013–2014) (unweighted, n=4614).

    Figure 5 presents the marginal prevalence of encoded exposure to tobacco ads across the four products by sexual orientation and race among men, adjusting for demographic and tobacco use variables. Compared with heterosexual white men, heterosexual black men had higher prevalence of encoded exposure to cigar ads; men who identified as something else and white had lower prevalence of encoded exposure to cigarette ads; bisexual men of other races had lower prevalence of encoded exposure to smokeless tobacco ads. Online supplementary appendix 7 presents the analyses testing the interactions between sexual orientation and race among men. There was one significant interaction between sexual orientation and race for encoded exposure to cigarette ads.

    Figure 5
    Figure 5

    Weighted prevalence of encoded tobacco ad exposure by sexual orientation and race among US adult men aged 18–24 years (2013–2014) (unweighted, n=4614).

    Figure 6 presents the marginal prevalence of encoded exposure to tobaccos ads across the four products by sexual orientation and ethnicity among men, adjusting for demographic and tobacco use variables. Compared with heterosexual non-Hispanic men, heterosexual Hispanic men reported higher prevalence of encoded exposure to cigarette and cigar ads. Online supplementary appendix 8 presents the analyses testing the interactions between sexual orientation and ethnicity among men. There were no significant interactions between sexual orientation and ethnicity among men.

    Figure 6
    Figure 6

    Weighted prevalence of encoded tobacco ad exposure by sexual orientation and ethnicity among US adult men aged 18–24 years (2013–2014) (unweighted, n=4614).

    Discussion

    This study adds to existing literature examining tobacco marketing targeted at sexual, racial and ethnic minority populations as a risk factor associated with increased tobacco use in these populations.16 18 46–48 We compared encoded exposure to cigarette, e-cigarette, cigar and smokeless tobacco ads among a nationally representative sample of US young adult sexual minorities versus their heterosexual counterparts. In this analysis, we found that bisexual women reported higher encoded exposure to cigarette ads, e-cigarettes and cigar ads compared with heterosexual women. Men who identified as something else (ie, neither straight, gay or bisexual) reported lower encoded exposure to cigarette ads and e-cigarette ads versus heterosexual men. We further found higher levels of encoded exposure to ads among sexual minorities of colour for various tobacco products, but lower levels of encoded exposure for other tobacco product ads, corroborating the complex intersectionality between sex, sexual orientation, race and ethnicity in relation to cigarette smoking and other tobacco product use reported in the literature.24 39–41

    Young adult bisexual women reported higher encoded exposure to three out of the four forms of tobacco product marketing compared with heterosexual women, adjusting for demographic and current or ever use of tobacco products. These results correspond to a report by Wheldon and colleagues, who found that young adult bisexual women had 4.0 to 5.5 times increased odds of regularly using cigarettes, e-cigarettes, and cigars compared with heterosexual young adult women.24 These findings suggest that increased tobacco product marketing exposure may be an important risk factor associated with tobacco use disparities among young adult bisexual women. The reasons for increased marketing exposure among bisexual women are unclear and may be due to increased opportunities of exposure via targeted social media marketing, point-of-sale promotions, free tobacco samples or coupons at concerts or special events, and tobacco industry sponsorship at lesbian, gay, bisexual and transgender events.18 47 Although we are not able to determine the causal direction between increased marketing exposure and tobacco use disparities in this population, other studies have examined the mediating role of tobacco marketing in influencing tobacco use behaviours among sexual minority women.16 Collectively, these findings warrant increased research and regulatory scrutiny over the marketing practices of tobacco manufacturers, specifically cigarettes, cigars and e-cigarettes, in targeting young adult bisexual women. Additional research will also be necessary to identify the major sources of increased exposure to tobacco marketing among bisexual women and their receptivity to marketing messages.

    In comparison, we found distinctly different patterns of encoded tobacco marketing exposure among young adult men by sexual orientation. Gay or bisexual men did not differ from heterosexual men in encoded exposure to tobacco product marketing. However, men who identified as something else had lower odds of encoded exposure to cigarette and e-cigarette ads than heterosexual men and gay men. Wheldon and colleagues reported that tobacco product use varied less consistently by sexual orientation among men than among women.24 Gay young adult men were more likely to be cigarette smokers than heterosexual men. However, gay and bisexual men were less likely to use smokeless tobacco than heterosexual men, and tobacco use among men who identified with other sexual orientations was not significantly different from heterosexual men.24 These contrasting findings between women and men and across sexual minority groups underscore the importance in future research of considering women and men and different sexual minority groups separately when comparing tobacco use behaviours and risk factors.

    We further observed a complex pattern of varying exposure to different tobacco product ads among sexual, racial and ethnic minorities compared with heterosexual white and non-Hispanic young adults. These findings reinforce the importance of research to increase our understanding of the macro and structural-level systems of privilege and oppression underpinning health disparities from tobacco use among multiple oppressed and marginalised populations across multiple social categories, including race, ethnicity, gender and sexual orientation. Among young adult women, heterosexual black women and bisexual women of colour may be at increased risk of tobacco ads exposure from cigarette, e-cigarette and cigar companies compared with heterosexual white and non-Hispanic women. Conversely, sexual minorities and women of colour reported lower encoded exposure to smokeless tobacco ads versus heterosexual white women. Among young adult men, heterosexual black and Hispanic men were at increased risk of tobacco ads exposure from cigarette and cigar ads versus heterosexual white and non-Hispanic men. Conversely, men who identified as something other than straight, gay or bisexual and white and bisexual men of other races had lower exposure to cigarette and smokeless ads, respectively, compared with heterosexual white men. Further research is needed to assess specific sources of e-cigarette ad exposure and structural risk factors that contribute to higher encoded exposure to some forms of tobacco ads among specific sexual, racial and ethnic minority young adults, as well as potential protective factors that relate to lower exposure to tobacco ads among other minority populations. The heterogeneity in encoded exposure to tobacco ads across sexual orientation, race and ethnicity indicates the need for future surveillance, research and targeted interventions among sexual minorities of colour to mitigate the influence of marketing in these subgroups. The findings from this study further suggest the need for greater scrutiny and restriction of tobacco industry marketing practices targeting sexual, racial and ethnic minorities to reduce tobacco-related health disparities in these populations.

    This study was limited in several ways. Participants were asked to report their sex as either male or female and were not asked about their gender identity and sex assigned at birth. This limited the ability to assess inequities in tobacco ads exposure among gender minorities compared with cisgender individuals. Although we observed variability in the prevalence of exposure to tobacco ads across sexual, racial and ethnic minority subgroups, there were few statistically significant interactions detected in the regression analyses, which may be due to small numbers of participants within certain subgroups, including women who identified as lesbian and from other races, something else (ie, not straight, lesbian/gay or bisexual) and other races, something else and black, something else and Hispanic, and men who identified as sexual minority and black or Hispanic. Therefore, the estimates from these subgroups were less precise. The derived race variable categorised Asian, American Indian/Alaska Native, Native Hawaiian and Pacific Islanders within ‘other races’. This analysis was not able to assess differences between these heterogenous subgroups due to small sample sizes. Future research involving larger samples of sexual minorities of colour from different racial and ethnic backgrounds would be needed to replicate this study. The study outcomes were based on self-reported recognition of ads of tobacco products, which may be prone to recall bias. In particular, participants with greater experience with tobacco products or greater susceptibility may selectively recall being exposed to tobacco ads. However, the approach of using recognition tasks to measure encoded exposure to tobacco-related content has been validated in previous tobacco control research and found to be predictive of subsequent tobacco use behaviour.42 49 Participants were shown a subset of 20 tobacco ads from a larger pool of ads within each category, and there may be under-reporting of actual ad exposure. We were not able to assess frequency of exposure to marketing, only whether they had seen each type of tobacco ads in the last 12 months. We were also not able to compare the effects of ad modality (eg, static vs video) and placement (eg, print, email or point-of-sale) using this data. Future research could consider comparing the effects of ad modality and placement on encoded exposure, ad receptivity and tobacco use behaviours among young adult sexual minorities and examining the intersections between sexual orientation, race and ethnicity. There was no available information in the PATH Study on the designated market area (regions within the USA used to define television and radio markets) in which participants were located, which limited the ability to analyse tobacco marketing distribution across different media markets and associations with distributions of sexual, racial and minority populations within these markets.

    In sum, this study demonstrates differences in tobacco ad exposure by sex, sexual orientation, race and ethnicity among US young adults. These findings inform several future research directions and potential targets for regulatory and public health education interventions to address tobacco use disparities among sexual minorities. First, research is needed to carefully examine intersections between sexual orientation, gender identity, and other social identities and contextual factors in explaining observed inequities in ad exposure. Second, greater scrutiny and restriction of tobacco industry marketing practices targeting sexual minorities are recommended. Third, efforts to develop effective and culturally tailored countermarketing health messages and cessation interventions for young adult sexual minorities of colour are clearly warranted.

    What this paper adds

    • This study examines the intersectionality between multiple social identities, including sex, sexual orientation, race and ethnicity in association with encoded exposure to tobacco product ads, a risk factor associated with increased tobacco use.

    • Bisexual young adult women report higher levels of encoded exposure to cigarette, e-cigarette and cigar ads compared with heterosexual young adult women.

    • We further found higher levels of encoded exposure to various tobacco product ads among bisexual women of colour, heterosexual black women and men, and heterosexual Hispanic men compared with young adults who were heterosexual white or non-Hispanic.

    • Results from this study may inform future research, surveillance of industry practices and public health education interventions to address inequities in exposure to tobacco marketing and tobacco use among multiple minority individuals based on sex, sexual orientation, race and ethnicity.

    References

    1. ↵
      2. Caceres BA ,
      3. Brody A ,
      4. Luscombe RE , et al
      . A systematic review of cardiovascular disease in sexual minorities. Am J Public Health 2017;107:e13–21.doi:10.2105/AJPH.2016.303630
      OpenUrl
    2. ↵
      2. Blosnich J ,
      3. Jarrett T ,
      4. Horn K
      . Disparities in smoking and acute respiratory illnesses among sexual minority young adults. Lung 2010;188:401–7.doi:10.1007/s00408-010-9244-5
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      2. Quinn GP ,
      3. Sanchez JA ,
      4. Sutton SK , et al
      . Cancer and lesbian, gay, bisexual, transgender/transsexual, and queer/questioning (LGBTQ) populations. CA Cancer J Clin 2015;65:384–400.doi:10.3322/caac.21288
      OpenUrlPubMed
    4. ↵
      1. Centers for Disease Control and Prevention
      . The health consequences of smoking - 50 years of progress: a report of the Surgeon General. Atlanta, GA, 2014.
    5. ↵
      2. Youatt EJ ,
      3. Johns MM ,
      4. Pingel ES , et al
      . Exploring young adult sexual minority women's perspectives on LGBTQ smoking. J LGBT Youth 2015;12:323–42.doi:10.1080/19361653.2015.1022242
      OpenUrl
    6. ↵
      2. Blosnich J ,
      3. Lee JGL ,
      4. Horn K
      . A systematic review of the aetiology of tobacco disparities for sexual minorities. Tob Control 2013;22:66–73.doi:10.1136/tobaccocontrol-2011-050181
      OpenUrlAbstract/FREE Full Text
    7. ↵
      2. Brewster KL ,
      3. Tillman KH
      . Sexual orientation and substance use among adolescents and young adults. Am J Public Health 2012;102:1168–76.doi:10.2105/AJPH.2011.300261
      OpenUrlCrossRefPubMed
    8. ↵
      1. American Lung Association
      . Smoking out a deadly threat: tobacco use in LGBT communities. Washington, DC: American Lung Association, 2010. https://www.lung.org/assets/documents/research/lgbt-report.pdf
    9. ↵
      2. Hinds JT ,
      3. Loukas A ,
      4. Perry CL
      . Explaining sexual minority young adult cigarette smoking disparities. Psychol Addict Behav 2019;33:371–81.doi:10.1037/adb0000465
      OpenUrl
    10. ↵
      2. Pachankis JE ,
      3. Hatzenbuehler ML ,
      4. Starks TJ
      . The influence of structural stigma and rejection sensitivity on young sexual minority men's daily tobacco and alcohol use. Soc Sci Med 2014;103:67–75.doi:10.1016/j.socscimed.2013.10.005
      OpenUrlCrossRefPubMed
    11. ↵
      2. Hatzenbuehler ML ,
      3. Wieringa NF ,
      4. Keyes KM
      . Community-Level determinants of tobacco use disparities in lesbian, gay, and bisexual youth: results from a population-based study. Arch Pediatr Adolesc Med 2011;165.doi:10.1001/archpediatrics.2011.64
    12. ↵
      2. Hatzenbuehler ML ,
      3. Keyes KM ,
      4. Hamilton A , et al
      . State-level tobacco environments and sexual orientation disparities in tobacco use and dependence in the USA. Tob Control 2014;23:e127–32.doi:10.1136/tobaccocontrol-2013-051279
      OpenUrlAbstract/FREE Full Text
    13. ↵
      2. Matthews AK ,
      3. Hotton A ,
      4. DuBois S , et al
      . Demographic, psychosocial, and contextual correlates of tobacco use in sexual minority women. Res Nurs Health 2011;34:141–52.doi:10.1002/nur.20427
      OpenUrlPubMed
    14. ↵
      2. Matthews AK ,
      3. Steffen A ,
      4. Hughes T , et al
      . Demographic, healthcare, and contextual factors associated with smoking status among sexual minority women. LGBT Health 2017;4:17–23.doi:10.1089/lgbt.2016.0039
      OpenUrl
    15. ↵
      2. Bowleg L
      . The problem with the phrase women and minorities: intersectionality-an important theoretical framework for public health. Am J Public Health 2012;102:1267–73.doi:10.2105/AJPH.2012.300750
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      2. Balsam KF ,
      3. Beadnell B ,
      4. Riggs KR
      . Understanding sexual orientation health disparities in smoking: a population-based analysis. Am J Orthopsychiatry 2012;82:482–93.doi:10.1111/j.1939-0025.2012.01186.x
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      2. Emory K ,
      3. Buchting FO ,
      4. Trinidad DR , et al
      . Lesbian, gay, bisexual, and transgender (LGBT) view it differently than non-LGBT: exposure to tobacco-related couponing, e-cigarette advertisements, and anti-tobacco messages on social and traditional media. Nicotine Tob Res 2019;21:513–22.doi:10.1093/ntr/nty049
      OpenUrlPubMed
    18. ↵
      2. Dilley JA ,
      3. Spigner C ,
      4. Boysun MJ , et al
      . Does tobacco industry marketing excessively impact lesbian, gay and bisexual communities? Tob Control 2008;17:385–90.doi:10.1136/tc.2007.024216
      OpenUrlAbstract/FREE Full Text
    19. ↵
      2. Yerger VB ,
      3. Przewoznik J ,
      4. Malone RE , et al
      . Racialized geography, corporate activity, and health disparities: tobacco industry targeting of inner cities. J Health Care Poor Underserved 2007;18:10–38.doi:10.1353/hpu.2007.0120
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      2. Moore DJ ,
      3. Williams JD ,
      4. Qualls WJ
      . Target marketing of tobacco and alcohol-related products to ethnic minority groups in the United States. Ethn Dis 1996;6:83–98.
      OpenUrlPubMed
    21. ↵
      2. Robinson RG ,
      3. Barry M ,
      4. Bloch M , et al
      . Report of the tobacco policy research group on marketing and promotions targeted at African Americans, Latinos, and women. Tob Control 1992;1:S24–30.doi:10.1136/tc.1.suppl1.S24
      OpenUrlFREE Full Text
    22. ↵
      2. Biener L ,
      3. Albers AB
      . Young adults: vulnerable new targets of tobacco marketing. Am J Public Health 2004;94:326–30.doi:10.2105/AJPH.94.2.326
      OpenUrlCrossRefPubMedWeb of Science
    23. ↵
      2. Sepe E ,
      3. Ling PM ,
      4. Glantz SA
      . Smooth moves: bar and nightclub tobacco promotions that target young adults. Am J Public Health 2002;92:414–9.doi:10.2105/AJPH.92.3.414
      OpenUrlCrossRefPubMedWeb of Science
    24. ↵
      2. Wheldon CW ,
      3. Kaufman AR ,
      4. Kasza KA , et al
      . Tobacco use among adults by sexual orientation: findings from the population assessment of tobacco and health study. LGBT Health 2018;5:33–44.doi:10.1089/lgbt.2017.0175
      OpenUrlPubMed
    25. ↵
      2. Richardson A ,
      3. Williams V ,
      4. Rath J , et al
      . The next generation of users: prevalence and longitudinal patterns of tobacco use among US young adults. Am J Public Health 2014;104:1429–36.doi:10.2105/AJPH.2013.301802
      OpenUrlCrossRefPubMedWeb of Science
    26. ↵
      2. Villanti AC ,
      3. Richardson A ,
      4. Vallone DM , et al
      . Flavored tobacco product use among U.S. young adults. Am J Prev Med 2013;44:388–91.doi:10.1016/j.amepre.2012.11.031
      OpenUrlCrossRefPubMed
    27. ↵
      2. Richardson A ,
      3. Pearson J ,
      4. Xiao H , et al
      . Prevalence, harm perceptions, and reasons for using noncombustible tobacco products among current and former smokers. Am J Public Health 2014;104:1437–44.doi:10.2105/AJPH.2013.301804
      OpenUrlCrossRefPubMed
    28. ↵
      2. Biener L ,
      3. Song E ,
      4. Sutfin EL , et al
      . Electronic cigarette trial and use among young adults: reasons for trial and cessation of Vaping. Int J Environ Res Public Health 2015;12:16019–26.doi:10.3390/ijerph121215039
      OpenUrl
    29. ↵
      2. Kasza KA ,
      3. Ambrose BK ,
      4. Conway KP , et al
      . Tobacco-product use by adults and youths in the United States in 2013 and 2014. N Engl J Med 2017;376:342–53.doi:10.1056/NEJMsa1607538
      OpenUrlPubMed
    30. ↵
      2. Ganz O ,
      3. Johnson AL ,
      4. Cohn AM , et al
      . Tobacco harm perceptions and use among sexual and gender minorities: findings from a national sample of young adults in the United States. Addict Behav 2018;81:104–8.doi:10.1016/j.addbeh.2018.01.032
      OpenUrl
    31. ↵
      2. Chu K-H ,
      3. Sidhu AK ,
      4. Valente TW
      . Electronic cigarette marketing online: a multi-site, Multi-Product comparison. JMIR Public Health Surveill 2015;1.doi:10.2196/publichealth.4777
    32. ↵
      2. Cobb NK ,
      3. Brookover J ,
      4. Cobb CO
      . Forensic analysis of online marketing for electronic nicotine delivery systems. Tob Control 2015;24:128–31.doi:10.1136/tobaccocontrol-2013-051185
      OpenUrlAbstract/FREE Full Text
    33. ↵
      2. Moran MB ,
      3. Chen-Sankey JC ,
      4. Tan AS , et al
      . Sources of awareness of JUUL e-cigarettes in 2 surveys of adults in the United States. Am J Health Behav 2019;43:279–86.doi:10.5993/AJHB.43.2.5
      OpenUrlPubMed
    34. ↵
      2. Anderson SJ
      . Marketing of menthol cigarettes and consumer perceptions: a review of tobacco industry documents. Tob Control 2011;20(Suppl 2):ii20–8.doi:10.1136/tc.2010.041939
      OpenUrlPubMed
    35. ↵
      2. Balbach ED ,
      3. Gasior RJ ,
      4. Barbeau EM
      . R.J. Reynolds' targeting of African Americans: 1988-2000. Am J Public Health 2003;93:822–7.doi:10.2105/AJPH.93.5.822
      OpenUrlCrossRefPubMedWeb of Science
    36. ↵
      2. Mills SD ,
      3. Henriksen L ,
      4. Golden SD , et al
      . Disparities in retail marketing for menthol cigarettes in the United States, 2015. Health Place 2018;53:62–70.doi:10.1016/j.healthplace.2018.06.011
      OpenUrl
    37. ↵
      2. Primack BA ,
      3. Bost JE ,
      4. Land SR , et al
      . Volume of tobacco advertising in African American markets: systematic review and meta-analysis. Public Health Rep 2007;122:607–15.doi:10.1177/003335490712200508
      OpenUrlCrossRefPubMed
    38. ↵
      2. Weinberger AH ,
      3. Giovenco DP ,
      4. Zhu J , et al
      . Racial/Ethnic differences in daily, nondaily, and menthol cigarette use and smoking quit ratios in the United States: 2002 to 2016. Prev Med 2019;125:32–9.doi:10.1016/j.ypmed.2019.04.009
      OpenUrl
    39. ↵
      2. Blosnich JR ,
      3. Jarrett T ,
      4. Horn K
      . Racial and ethnic differences in current use of cigarettes, cigars, and hookahs among lesbian, gay, and bisexual young adults. Nicotine Tob Res 2011;13:487–91.doi:10.1093/ntr/ntq261
      OpenUrlCrossRefPubMedWeb of Science
    40. ↵
      2. Ortiz KS ,
      3. Duncan DT ,
      4. Blosnich JR , et al
      . Smoking among sexual minorities: are there racial differences? Nicotine Tob Res 2015;17:1362–8.doi:10.1093/ntr/ntv001
      OpenUrlCrossRefPubMed
    41. ↵
      2. Corliss HL ,
      3. Rosario M ,
      4. Birkett MA , et al
      . Sexual orientation disparities in adolescent cigarette smoking: intersections with race/ethnicity, gender, and age. Am J Public Health 2014;104:1137–47.doi:10.2105/AJPH.2013.301819
      OpenUrlCrossRefPubMed
    42. ↵
      2. Southwell BG ,
      3. Barmada CH ,
      4. Hornik RC , et al
      . Can we measure encoded exposure? validation evidence from a national campaign. J Health Commun 2002;7:445–53.doi:10.1080/10810730290001800
      OpenUrlCrossRefPubMedWeb of Science
    43. ↵
      1. United States Department of Health and Human Services. National Institutes of Health. National Institute on Drug Abuse
      2. United States Department of Health and Human Services. Food and Drug Administration. Center for Tobacco Products
      . Population assessment of tobacco and health (path) study wave 2 questionnaire. Popul Assess Tob Health PATH Study Ser 2018.
    44. ↵
      2. Hyland A ,
      3. Ambrose BK ,
      4. Conway KP , et al
      . Design and methods of the population assessment of tobacco and health (path) study. Tob Control 2017;26:371–8.doi:10.1136/tobaccocontrol-2016-052934
      OpenUrlAbstract/FREE Full Text
    45. ↵
      2. Allison P
      . When can you safely ignore Multicollinearity? 2012. Available: https://statisticalhorizons.com/multicollinearity [Accessed 16 Nov 2018].
    46. ↵
      2. Smith EA ,
      3. Malone RE
      . The outing of Philip Morris: advertising tobacco to gay men. Am J Public Health 2003;93:988–93.doi:10.2105/AJPH.93.6.988
      OpenUrlCrossRefPubMedWeb of Science
    47. ↵
      2. Stevens P ,
      3. Carlson LM ,
      4. Hinman JM
      . An analysis of tobacco industry marketing to lesbian, gay, bisexual, and transgender (LGBT) populations: strategies for mainstream tobacco control and prevention. Health Promot Pract 2004;5:129S–34.doi:10.1177/1524839904264617
      OpenUrlCrossRefPubMed
    48. ↵
      2. Matthews AK ,
      3. Balsam K ,
      4. Hotton A , et al
      . Awareness of media-based antitobacco messages among a community sample of LGBT individuals. Health Promot Pract 2014;15:857–66.doi:10.1177/1524839914533343
      OpenUrlCrossRefPubMed
    49. ↵
      2. Depue JB ,
      3. Southwell BG ,
      4. Betzner AE , et al
      . Encoded exposure to tobacco use in social media predicts subsequent smoking behavior. Am J Health Promot 2015;29.doi:10.4278/ajhp.130214-ARB-69

    Footnotes

    • Contributors All authors contributed to the conceptualisation of the study. ASLT conducted the data analyses and was responsible for the overall content as guarantor. ASLT and EPH led the initial drafting of the paper. All authors contributed to the writing, and reviewed and approved the final manuscript.

    • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

    • Competing interests None declared.

    • Patient consent for publication Not required.

    • Ethics approval Ethics review for the Population Assessment of Tobacco and Health Study was obtained by Westat. The current analysis was reviewed and determined to be exempt by our university’s institutional review board.

    • Provenance and peer review Not commissioned; externally peer reviewed.