Abstract
Background
Exposure to tobacco product marketing increases tobacco use among young adults, especially those from vulnerable communities (VCs).
Purpose
This study examined real-time tobacco marketing exposure among young adults from vulnerable and non-vulnerable communities using Ecological Momentary Assessment (EMA).
Methods
This study used EMA data to assess context (e.g., location and activity) of tobacco marketing exposure using four text-messaging surveys per day over 2 weeks. Young adult non-current tobacco users living in Washington, D.C. (n = 146; ages 18–24) recorded 5,285 surveys, including 20 participants (13.2%) from VCs with high proportions of lower income and racial/ethnic minorities, and high smoking rates. Unadjusted and adjusted multilevel logistic regressions were used to assess the associations between exposure to any and flavored tobacco marketing, VC residence, and real-time context.
Results
Fifty-nine participants (40.4%) reported at least one tobacco marketing exposure and recorded 94 exposure moments. In adjusted models, odds of exposure were higher among VC residents (AOR = 2.6, 95% CI = 1.2–5.4), in the presence of anyone using tobacco versus no use (AOR = 4.0, 95% CI = 2.4–6.7), at store/retail (AOR = 17.0, 95% CI = 6.4–44.8), or outside/in transit (AOR = 4.1, 95% CI = 2.1–7.8) versus at home. VC residence (AOR = 7.2, 95% CI = 2.3–22.2) was the strongest predictor of flavored tobacco marketing exposure among all covariates examined.
Conclusions
Young adults are predominantly exposed to tobacco marketing in their daily lives through retail advertisements. Young adults from VCs are at increased risks of seeing any tobacco and especially flavored tobacco marketing. Policies that curtail tobacco retailer density and advertisement displays may reduce overall and differential tobacco marketing exposure.
Keywords: Tobacco marketing, Flavored tobacco marketing, Ecological momentary assessment, Young adults, Community health, Health disparities
Individuals living in vulnerable communities were more likely to be exposed to any tobacco and flavored tobacco marketing than those who lived outside these communities.
Introduction
In recent years, the uptake of tobacco products among young adults has increased in the USA, with overall tobacco use initiation showing significant increases for both e-cigarettes and cigarettes [1, 2]. Specifically, e-cigarette use prevalence among young adults increased from 5.1% to 7.6% from 2014 to 2018 [3], while the proportion of ever smokers who initiated cigarettes in young adulthood more than doubled from 2002 to 2018 [2]. According to state and national surveillance systems, young adults have become more likely to initiate tobacco products than youth (those aged 17 and under) [4] and have the highest rates of current tobacco use compared to any other age group [5]. This trend of increased tobacco use among young adults is alarming, given that tobacco product use among young adults can lead to nicotine addiction and increases the risks for developing cancers and cardiovascular diseases later in life [6].
One important risk factor for tobacco initiation and regular use among young adults is exposure to tobacco marketing. Mounting evidence shows that exposure to tobacco marketing materials in one’s physical environment (e.g., advertisements in brick-and-mortar retail outlets) leads to perceptions of tobacco use as favorable and normative, while stimulating impulse purchases of tobacco, tobacco product initiation and regular use [7–9]. Social media and websites have also continued to serve as a significant venue in which young adults receive regular exposure to tobacco marketing messages [10, 11]. Moreover, young adults may be more likely to recall or show favorable attitudes toward tobacco marketing materials than older adults [11–13], further increasing their risks of being influenced by the tobacco industry’s marketing efforts.
Additionally, racial and ethnic minorities and those from low socioeconomic statuses are more likely to use tobacco products such as cigarettes and cigars [14, 15] and flavored tobacco products such as menthol cigarettes than their counterparts from less vulnerable groups [16–18]. One potential reason for this disparity is the tobacco industry’s targeted marketing to the vulnerable communities characterized by having a high density of residents from racial/ethnic minority backgrounds and/or lower-socioeconomic statuses. For example, the density of tobacco retailers and the amount of tobacco advertising materials found in each retailer is higher in communities with higher percentages of racial/ethnic minorities populations of low-socioeconomic statuses [19–21]. Other evidence suggests that the tobacco industry attempts to sell and promote tobacco products by selectively targeting vulnerable, disadvantaged neighborhoods [22, 23]. This disproportionate exposure to tobacco marketing among young adults living in vulnerable communities may prompt their tobacco initiation and regular use during early adulthood, which may further translate to increased disparities in tobacco-use-related outcomes, including the development of chronic respiratory problems, cardiovascular disease, and tobacco-related cancers [6]. Therefore, investigating the individual and community-level risk factors of tobacco marketing exposure will help inform evidence-based strategies to reduce tobacco marketing among members of such vulnerable communities.
Previous studies examining the risk factors and context of tobacco marketing exposure have had several limitations. First, instead of real-time exposure, many of these studies measured retrospective recall of exposure [22, 23], a method that may lead to recall errors and subjective biases. Second, the effects of exposure specifically to flavored tobacco marketing, as opposed to tobacco marketing in general, have largely been unexamined. Given the prevalence of flavored tobacco use is on the rise among young adults, and the link between flavored tobacco product use with initiation and regular use in this age group [16–18], it is critical to assess the context and risk factors for exposure to flavored tobacco marketing. Third, previous observational studies [24–26] using survey responses have mostly focused on assessing individuals’ socio-demographics (e.g., race/ethnicity and socioeconomic statuses) as the main risk factors of tobacco marketing exposure, without assessing individuals’ physical residence in vulnerable communities. Additional research may help assess the associations between young adults’ residency in vulnerable communities and their exposure to tobacco marketing, as they are likely exposed throughout their normal daily activities.
To address these limitations, this study used data from an Ecological Momentary Assessment (EMA) study that examined real-time and real-world exposure to tobacco and flavored tobacco marketing materials and its associated context, making use of participants’ current experiences and behaviors in their natural contexts [27]. EMA provides tracking closer to the moment of exposure [27] and is well-suited for collecting information on the infrequent occurrences of pro-tobacco marketing materials and messages [28–34]. For example, former studies used EMA to capture young adults’ momentary encounters of tobacco advertising [31] and pro-smoking media [32]. These studies, however, did not specifically examine participants’ real-time exposure to flavored tobacco marketing or their physical residence in vulnerable communities as one of the risk factors for tobacco marketing exposure. Therefore, the goal of this study was to use EMA to assess the associations between vulnerable community residence, the real-time context of activities and locations, and momentary exposure to tobacco and flavored tobacco marketing among young adult non-current tobacco users.
Methods
Participants and Recruitment
The study recruited young adult non-current tobacco users in Washington, D.C., from 2017 to 2019. Recruitment occurred through posting on community, neighborhood, and university listservs and websites (e.g., Craigslist) and posting flyers in community locations (e.g., libraries or coffee shops), as well as community events (e.g., health fairs) around the city. Recruitment materials directed potential respondents to an online eligibility survey. Eligibility requirements included: (a) aged 18–24; (b) not having used any tobacco product (i.e., cigarettes, premium cigars, little cigars/cigarillos, e-cigarettes, smokeless tobacco, snus, hookah/shisha/waterpipe) in the past 30 days; (c) residing, working, or attending school in Washington, D.C.; and (d) owning an Android or iPhone smartphone with an unlimited text plan. Eligible respondents from the online screening survey were re-screened for eligibility via phone, attended an enrollment session with study staff where they provided consent to participate, and were then oriented to the EMA protocol. After orientation, participants completed a baseline survey, resulting in 152 participants who enrolled in the two-week EMA study. Six respondents had technical difficulties in responding to the EMA and therefore were dropped from the analysis leaving an analytic sample of 146. The study design and protocol were approved by the Advarra IRB (formerly Chesapeake IRB) (Protocol: Pro00020538) and the University of Kentucky IRB (Protocol: 51579).
Study Procedures
For this study, EMA data were gathered over a 2-week period for each respondent (August 2017 and April 2019). Prior to beginning EMA, participants completed an in-person or telephone baseline session, including a baseline survey and training on completing the EMA. All baseline participants were enrolled in the EMA. Study participants responded to random prompts regarding their momentary exposures to tobacco marketing via an app or text links sent to their smartphones four times per day. Initially, the EMA was conducted via an app (Illumivu.com). However, this app proved cumbersome for participants with limited space on their phones and had other technical challenges. As a result, we switched to sending EMA links in text messages via Survey Signal (surveysignal.com Chicago, IL), allowing the remaining respondents to complete a Qualtrics mini-survey on their smartphone via their phone’s browser. A total of 42 respondents completed EMA using Illumivu and the remainder via Survey Signal. On average, the EMA survey took 52 s for participants to complete. Participants received $25 for completing their baseline survey. Participants also received $20 upon starting EMA and an increased amount of incentives for completing more EMA assessments: 0–14 surveys completed $0, 15–24 surveys $5, 25–34 surveys $10, 35–44 surveys $15, 45–49 surveys $20, and 50+ surveys $25. Respondents began the 2-week EMA period the day after completing their baseline survey with random scheduled times in four time block intervals during awake hours (typically 8 am to 11 pm). Once participants were alerted, the survey was available up to 60 min and they received one reminder after 10 min. This particular analysis used data from this study’s baseline survey and respondents’ subsequent two weeks of EMA response.
Measures
EMA time-varying tobacco marketing exposure
Answers to the random EMA surveys captured current tobacco marketing exposure at the time of assessment. Exposure to tobacco marketing (yes/no) was measured by the question “In your current location, do you see any advertisements or signs promoting tobacco products?” Exposure to flavored tobacco marketing (yes/no) was measured by the question, “In your current location, looking at the biggest advertisement you see promoting tobacco products, is it for (a) a non-flavored/tobacco flavored product, (b) a flavored tobacco product, such as menthol, candy or fruit flavor, and (c) both a flavored and non-flavored tobacco product(s).” Those who reported seeing flavored tobacco products (options 2 and 3) in the advertisements were considered as having flavored tobacco marketing exposure.
EMA time-varying covariates
Answers to the random EMA surveys captured current location and activity at the time of assessment. The following categories were used for momentary location and activity: (home, work/school, party/club/other social places, bar/restaurant, outside/in transit, store or other retail location, online/social media, or other location). In the presence of anyone using tobacco in the current location at the time of assessment (yes/no) was measured by the question “Is anyone using tobacco products in your current location?”
Baseline covariates
Socio-demographics and tobacco use history were measured by the baseline survey. Demographics were collected from the baseline survey regarding age (18–20 vs. 21–24), biological sex (male vs. female), race/ethnicity (non-Hispanic White, non-Hispanic Black, non-Hispanic other race, vs. Hispanic), and highest completed education level (high school degree/GED or currently enrolled in high school, some college or in college, vs. college degree and higher). Ever tobacco use status was assessed using several items from the PhenX Tool kit for ever use of cigarettes, cigars, little cigars/cigarillos, e-cigarettes, smokeless tobacco, snus, hookah/shisha/waterpipe [35]. For analysis, any tobacco use among this sample of baseline past-30-day non-tobacco users was measured dichotomously as never use or ever use. Washington, D.C., wards 5, 7, and 8 were considered as vulnerable communities mainly because they have high proportions of residents who currently smoked cigarettes (17.1–24.0% compared with 14.3% in the city as a whole in 2017) [36] and they are among the wards with the lowest median household income [37]. Additionally, these communities have the D.C. area’s highest proportions of non-Hispanic Black/African American residents [37], a group often targeted by the tobacco industry to market tobacco products [21, 22].
Statistical Analysis
Our descriptive statistics analysis was conducted to assess baseline participant demographics and momentary reports of seeing others using tobacco and seeing tobacco and flavored tobacco marketing. Using SAS v9.4 software, we conducted chi-square (χ 2) tests to determine if there were any differences in demographics between those who reported seeing tobacco marketing and those who did not. We also described EMA completion by demographic characteristics of the participants and conducted t-tests or ANOVA to determine if there were differences in percentage completion by demographic characteristics. Finally, we calculated unadjusted and adjusted multilevel logistic regression models to assess individual and observation specific correlates of reporting momentary exposure to tobacco marketing and flavored tobacco marketing [38]. We implemented random intercepts in the regression models, which assumed a within-subject compound symmetry correlation structure [39, 40]. Due to the structure of the data, we nested EMA observations (level one) within individual respondents (level two). These models allow for different numbers of observations across individuals and account for non-independence of EMA responses by the same individual. We include both level one EMA time-varying correlates (tobacco use at location, and location and activity) and level two baseline correlates (socio-demographics, ever tobacco use, and vulnerable community residence). Additionally, we conducted a sensitivity analysis that restricted the sample to the participants who had an EMA compliance of 20% and higher [41] in order to compare whether the time-varying correlates and baseline correlates vary based on participants’ EMA compliance.
Results
Participant Characteristics and EMA Compliance
Table 1 shows that among 146 participants who completed more than one random EMA survey, the majority of the participants were female (n = 116, 79.5%) and between the ages of 21 and 24 (n = 107; 74.8%). Almost 40% of respondents (n = 57, 39.0%) had never used tobacco products before, and 20 participants were from vulnerable communities (13.7%). Although only 28.1% (n = 41) of the overall sample was non-Hispanic Black individuals, 45.0% (n = 9) of those from vulnerable communities were non-Hispanic Black individuals. Table 2 shows overall EMA compliance (65.5%) and by participant characteristics. Overall, participants completed 5,285 surveys (36.2 surveys per person) during the 2-week assessment. EMA compliance was higher among females than males (68.0% vs. 56.0%; p = 0.013) and was not significantly associated with tobacco use status or other demographic characteristics. Nine participants were excluded from the sensitivity analysis (n = 137) since they had an EMA compliance of lower than 20%. The overall EMA compliance for this new sample was 69.3%.
Table 1.
Participant characteristics by seeing tobacco marketing (n = 146)
| Overall | Report seeing any tobacco marketing | ||||||
|---|---|---|---|---|---|---|---|
| Yes | No | ||||||
| N | % | N | % | N | % | P-valuea,b | |
| Biological sex | 0.48 | ||||||
| Male | 30 | 20.5 | 14 | 24.6 | 16 | 18.0 | |
| Female | 116 | 79.5 | 43 | 65.4 | 73 | 82.0 | |
| Age | 0.30 | ||||||
| 18–20 | 39 | 26.7 | 20 | 33.3 | 19 | 22.1 | |
| 21–24 | 107 | 73.3 | 40 | 66.7 | 67 | 77.9 | |
| Race/ethnicity | 0.17 | ||||||
| Non-Hispanic White | 57 | 39.0 | 18 | 31.6 | 39 | 43.8 | |
| Non-Hispanic Black | 41 | 28.1 | 16 | 28.1 | 25 | 28.1 | |
| Hispanic | 18 | 12.3 | 11 | 19.3 | 7 | 7.9 | |
| Non-Hispanic other | 30 | 20.6 | 12 | 21.0 | 18 | 20.2 | |
| Highest completed education level | 0.33 | ||||||
| High school degree or less | 21 | 14.4 | 10 | 17.5 | 11 | 12.4 | |
| Some college or in college | 50 | 34.2 | 22 | 38.6 | 28 | 31.5 | |
| College degree and higher | 75 | 51.4 | 25 | 43.9 | 78 | 87.6 | |
| Ever tobacco use | 0.72 | ||||||
| Never used | 57 | 39.0 | 21 | 36.8 | 36 | 40.5 | |
| Ever used | 89 | 61.0 | 36 | 63.2 | 53 | 59.5 | |
| Vulnerable community residence | 0.002 | ||||||
| No | 126 | 86.3 | 43 | 75.4 | 83 | 93.3 | |
| Yes | 20 | 13.7 | 14 | 24.6 | 6 | 6.7 | |
a Results are from chi-square tests.
b Bolded p-values are statistically significant p < 0.05.
Table 2.
EMA study completion and study participant characteristics (n = 146)
| EMA completion % Random EMA surveys completed | |||
|---|---|---|---|
| % | SD | P-valuea,b | |
| Biological sex | 0.01 | ||
| Male | 56.0 | 0.27 | |
| Female | 68.0 | 0.22 | |
| Age | 0.61 | ||
| 18–20 | 67.1 | 0.28 | |
| 21–24 | 64.8 | 0.23 | |
| Race/ethnicity | 0.32 | ||
| Non-Hispanic White | 0.67 | 0.24 | |
| Non-Hispanic Black | 0.60 | 0.25 | |
| Hispanic | 0.64 | 0.25 | |
| Non-Hispanic other | 0.70 | 0.20 | |
| Highest completed education level | 0.55 | ||
| High school degree or less | 0.67 | 0.30 | |
| Some college or in college | 0.63 | 0.24 | |
| College degree and higher | 0.67 | 0.21 | |
| Ever tobacco use | 0.47 | ||
| Never used | 67.2 | 0.26 | |
| Ever used | 64.4 | 0.22 | |
| Vulnerable community residence | 0.17 | ||
| No | 66.6 | 0.23 | |
| Yes | 58.8 | 0.26 | |
a Results are from t-tests (for variables of two categories) and ANOVA tests (for variables of more than two categories).
b Bolded p-values are statistically significant p < 0.05.
Real-Time Exposure to Tobacco and Flavored Tobacco Marketing
Among all completed surveys, 94 (1.8%) indicated momentary exposure to tobacco marketing and 45 (0.9%) indicated momentary exposure to flavored tobacco marketing. Among all participants, 56 (38.1%) and 30 (20.4%) reported seeing tobacco marketing and flavored tobacco marketing at least once during the 2-week assessment, respectively. Additionally, 20 participants (13.6%) reported seeing tobacco marketing, with 9 participants (6.1%) reporting seeing flavored tobacco marketing more than once during the 2-week assessment.
Predictors of Real-Time Tobacco Marketing Exposure
Table 3 shows the unadjusted and adjusted analyses for predicting real-time exposure to tobacco marketing. The unadjusted model shows that certain groups, including non-Hispanic Black (OR = 2.1, 95% CI = 1.0–4.4) and Hispanic (OR = 3.6, 95% CI = 1.5–8.6) individuals as well as those from vulnerable communities (OR = 3.5, 95% CI = 1.7–7.2), were more likely to see tobacco marketing than non-Hispanic Whites and those not from vulnerable communities. Observing others using tobacco products in the current location (OR = 6.3, 95% CI = 3.9–10.2) was associated with higher odds of seeing tobacco marketing in that location in real time than not observing others using tobacco. Being at bars/restaurants (OR = 2.9, 95% CI = 1.2–7.1), outside/in transit (OR = 5.5, 95% CI = 3.0–10.3), at store/other retail (OR = 25.2, 95% CI = 9.8–64.9), online/on social media (OR = 2.4, 95% CI = 1.1–5.1), and other locations (OR = 6.0, 95% CI = 1.9–18.9) were also associated with higher odds of seeing tobacco marketing in real time than being at home.
Table 3.
Participant characteristics and real-time risk factors for momentary tobacco marketing exposure (i = 5,285 surveys, n = 146)
| ORa | 95% CI | P-value | AORa,b | 95% CI | P-value | |
|---|---|---|---|---|---|---|
| Baseline covariates | ||||||
| Biological sex | 0.3047 | 0.3251 | ||||
| Male | Reference | Reference | ||||
| Female | 0.7 | 0.3–1.5 | 0.7 | 0.3–1.4 | ||
| Age | 0.4050 | 0.9588 | ||||
| 18–20 | 1.3 | 0.7–2.7 | 1.1 | 0.6–2.3 | ||
| 21–24 | Reference | Reference | ||||
| Race/ethnicity | 0.0296 | 0.1873 | ||||
| Non-Hispanic White | Reference | Reference | ||||
| Non-Hispanic Black | 2.1 | 1.0–4.4 | 1.8 | 0.9–3.7 | ||
| Hispanic | 3.6 | 1.5–8.6 | 2.1 | 0.9–5.2 | ||
| Non-Hispanic other | 1.4 | 0.6–3.2 | 1.2 | 0.5–2.7 | ||
| Highest completed education level | 0.5644 | 0.7895 | ||||
| High school degree or less | 1.6 | 0.6–3.8 | 3.5 | 1.0–12.0 | ||
| Some college or in college | 1.6 | 0.8–3.1 | 1.9 | 0.7–5.2 | ||
| College degree and higher | Reference | Reference | ||||
| Ever tobacco use | 0.6560 | 0.2997 | ||||
| Never used | Reference | Reference | ||||
| Ever used | 1.2 | 0.6–2.2 | 1.4 | 0.7–2.5 | ||
| Vulnerable community residence | 0.0007 | 0.0138 | ||||
| No | Reference | Reference | ||||
| Yes | 3.5 | 1.7–7.2 | 2.6 | 1.2–5.4 | ||
| EMA time-varying covariates | ||||||
| Saw anyone using tobacco products | 0.0001 | <0.0001 | ||||
| No | Reference | Reference | ||||
| Yes | 6.3 | 3.9–10.2 | 4.0 | 2.4–6.7 | ||
| Current location and activity | 0.0001 | <0.0001 | ||||
| Home | Reference | Reference | ||||
| Work/school/chores | 1.1 | 0.5–2.2 | 1.1 | 0.5–2.2 | ||
| Party/club/other social places | 2.0 | 0.7–6.2 | 1.6 | 0.5–4.9 | ||
| Bar/restaurant | 2.9 | 1.2–7.1 | 2.6 | 1.0–6.5 | ||
| Outside/in transit | 5.5 | 3.0–10.3 | 4.1 | 2.1–7.8 | ||
| Store/other retail | 25.2 | 9.8–64.9 | 17.0 | 6.4–44.8 | ||
| Online/social mediac | 2.4 | 1.1–5.1 | 2.0 | 0.9–4.4 | ||
| Other location | 6.0 | 1.9–18.9 | 5.4 | 1.7–17.1 |
a Bolded p-values are statistically significant p < 0.05.
b The intraclass correlation for the null models was 0.28.
c Respondents reported being online or using social media at the time of the prompt regardless of their physical location.
The adjusted model shows that participants with a high school degree or less compared with those with a college degree or higher (AOR = 3.5, 1.0–12.0) and those living in vulnerable communities compared with those living outside such communities (AOR = 2.6, 95% CI = 1.2–5.4) were more likely to see tobacco marketing. Being in locations with others using tobacco products (AOR = 4.0, 95% CI = 2.4–6.7) was associated with higher odds of seeing tobacco marketing in those locations in real time than in locations with no tobacco use in sight. In adjusted models, those at bars/restaurants (AOR = 2.6, 95% CI = 1.0–6.5), being outside/in transit (AOR = 4.1, 95% CI = 2.1–7.8), and at store/other retail (AOR = 17.0, 95% CI = 6.4–44.8) were more likely to see tobacco marketing in real time than those at home.
Sensitivity analysis results (Electronic Supplementary Material-Table S1) show similar findings. The only identified differences are that in the unadjusted model, non-Hispanic Black participants no longer reported a higher likelihood of real-time tobacco marketing exposure than non-Hispanic White participants; while in the adjusted model, Hispanic participants were more likely to report a higher likelihood of real-time tobacco marketing exposure than non-Hispanic White participants (AOR = 2.6, 95% CI = 1.1–6.4)
Predictors of Real-Time Flavored Tobacco Marketing Exposure
Table 4 shows the unadjusted and adjusted analyses for predicting real-time exposure to flavored tobacco marketing. Correlates of seeing flavored tobacco marketing exposure were similar to those for seeing any tobacco marketing, but odds were higher for younger age (18–20), Hispanic ethnicity, vulnerable community residence, in the presence of anyone using tobacco, and outside/in transit. Those of younger age (OR = 2.8, 95% CI = 1.0–7.3), Hispanics (OR = 4.4, 95% CI = 1.2–15.8), and those from vulnerable communities (OR = 7.2, 95% CI = 2.7–18.7) were more likely to see flavored tobacco marketing than those of older age, non-Hispanic Whites, and those not from vulnerable communities, respectively. Being outside/in transit (OR = 7.6, 95% CI = 3.2–18.0), at store/other retail (AOR = 8.1, 95% CI = 1.5–43.6), or online/on social media (AOR = 2.8, 95% CI = 1.0–8.2) was also associated with higher odds of seeing flavored tobacco marketing than being at home.
Table 4.
Participant characteristics and real-time risk factors for momentary flavored tobacco marketing exposure (i = 5,285 surveys, n = 146)
| ORa | 95% CI | P-value | AORa,b | 95% CI | P-value | |
|---|---|---|---|---|---|---|
| Baseline covariates | ||||||
| Biological sex | 0.4253 | 0.2752 | ||||
| Male | Reference | Reference | ||||
| Female | 0.6 | 0.2–1.9 | 0.5 | 0.2–1.7 | ||
| Age | 0.0415 | 0.1873 | ||||
| 18–20 | 2.8 | 1.0–7.3 | 2.3 | 0.7–8.0 | ||
| 21–24 | Reference | Reference | ||||
| Race/ethnicity | 0.0866 | 0.6387 | ||||
| Non-Hispanic White | Reference | Reference | ||||
| Non-Hispanic Black | 2.1 | 0.7–6.4 | 1.7 | 0.6–5.1 | ||
| Hispanic | 4.4 | 1.2–15.8 | 1.9 | 0.5–7.2 | ||
| Non-Hispanic other | 0.9 | 0.3–3.5 | 0.9 | 0.2–3.4 | ||
| Highest completed education level | 0.0903 | 0.7816 | ||||
| High school degree or less | 3.5 | 1.0–12.1 | 1.0 | 0.2–4.4 | ||
| Some college or in college | 1.9 | 0.7–5.2 | 0.7 | 0.2–2.3 | ||
| College degree and higher | Reference | Reference | ||||
| Ever tobacco use | 0.9800 | 0.3044 | ||||
| Never used | Reference | Reference | ||||
| Ever used | 1.0 | 0.4–2.5 | 1.6 | 0.6–4.1 | ||
| Vulnerable community residence | 0.0001 | 0.0007 | ||||
| No | Reference | Reference | ||||
| Yes | 7.2 | 2.7–18.7 | 7.2 | 2.3–22.2 | ||
| EMA time-varying covariates | ||||||
| In the presence anyone using tobacco | 0.0404 | 0.0011 | ||||
| No | Reference | Reference | ||||
| Yes | 7.3 | 3.8–14.3 | 4.3 | 2.1–8.9 | ||
| Current location and activity | 0.0001 | 0.0114 | ||||
| Home | Reference | Reference | ||||
| Work/school/chores | 0.9 | 0.3–2.7 | 0.8 | 0.3–2.6 | ||
| Party/club/other social places | 3.6 | 0.9–14.1 | 2.6 | 0.6–10.5 | ||
| Bar/restaurant | 2.4 | 0.6–9.1 | 1.9 | 0.5–8.1 | ||
| Outside/in transit | 7.6 | 3.2–18.0 | 5.6 | 2.2–14.0 | ||
| Store/other retail | 8.1 | 1.5–43.6 | 4.3 | 0.8–24.4 | ||
| Online/social mediac | 2.8 | 1.0–8.2 | 2.3 | 0.8–6.7 | ||
| Other location | 2.7 | 0.3–23.4 | 2.3 | 0.3–20.7 |
a Bolded p-values are statistically significant p < 0.05.
b The intraclass correlation for the null models was 0.39.
c Respondents reported being online or using social media at the time of the prompt regardless of their physical location.
In adjusted model results, odds of flavored tobacco marketing exposure were higher than those for any tobacco marketing exposure when considering several risk factors, including vulnerable community residence, in the presence of anyone using tobacco, and outside/in transit. Participants from vulnerable communities (AOR = 5.6, 95% CI = 2.2–14.0) are more likely to see flavored tobacco marketing than those not from vulnerable communities. In the presence of anyone using tobacco products (AOR = 4.3, 95% CI = 2.1–8.9) is more likely to be associated with seeing flavored tobacco marketing in real time than not seeing tobacco use by others. Being outside/in transit (AOR = 5.6, 95% CI = 2.2–14.0) was also associated with higher odds of seeing flavored tobacco marketing in real time than being at home. Sensitivity analysis results (Supplementary Table S2) show similar findings. The only identified difference is that participants of younger age (18–20) were no longer more likely to report real-time exposure to flavored tobacco marketing compared to those of older age (21–24) in the unadjusted model.
Sensitivity analysis results (Electronic Supplementary Material-Table S2) show similar findings. The only identified difference is that participants of younger age (18–20) were no longer more likely to report real-time exposure to flavored tobacco marketing compared to those of older age (21–24) in the unadjusted model.
Discussion
This study adds to the literature on using EMA to capture real-time exposures to tobacco marketing and flavored tobacco marketing among young adults who were not current tobacco users. The study highlights a series of key correlates and risk factors for real-time tobacco marketing exposure, including living in a vulnerable community, in the presence of others using tobacco, and being in public spaces where tobacco marketing may be more prevalent such as at stores/retailers, outside/in transit, and at bars/restaurants. The results of this study have strong implications for tobacco prevention and policymaking by minimizing the presence of tobacco marketing among young adults, especially those from vulnerable communities.
Our study underscores that the most common location for seeing tobacco marketing in real time is retail locations for the young adults in our study. Since the participants are non-current tobacco users, these retail exposures are unlikely related to purchasing tobacco but rather are for other reasons such as buying groceries, alcohol, or other non-tobacco products. This exposure to tobacco marketing materials may increase young people’s likelihood of developing pro-tobacco perceptions that normalizes tobacco use and reduce their harm perceptions about tobacco use, both of which may ultimately lead to experimenting with tobacco products [42–44]. Evidence has already shown that the frequency of walking to school and stopping at tobacco-selling retailers, mostly gas stations and convenience stores, was strongly associated with current use of cigarettes, e-cigarettes, and cigars among young people [44]. Other research demonstrates that more frequent visiting of convenience stores among tobacco-naïve youth was associated with increased odds of subsequent e-cigarette initiation [42, 43]. Additionally, our results present a unique opportunity to compare the risks of real-time tobacco marketing exposure through online sources and traditional marketing locations. Although evidence has indicated that retail websites and social media have become an increasingly important source of tobacco marketing exposure for young adults [10, 11], our results and former studies [29, 30] reveal that point-of-sale retailers may still be the most prominent source of exposure for this group.
The study results reveal that in the unadjusted regression models, participants who were Hispanic, non-Hispanic Black, and/or with high school degree or less reported higher odds of seeing tobacco or flavored tobacco marketing in real time as compared to their counterparts. In particular, Hispanic young adults were more likely than non-Hispanic White young adults to be exposed to any tobacco and flavored tobacco marketing in the unadjusted models. These results highlight the increased risks of being targeted by tobacco marketing and promotion among Hispanic young people, echoing previous studies that documented similar strategies targeting Hispanic communities [45, 46]. These forementioned associations, however, were no longer statistically significant after adjusting for other confounders. In particular, vulnerable community residence is strongly associated with increased exposure to tobacco marketing and flavored tobacco marketing, even after holding other known individual and contextual risk factors (e.g., race/ethnicity, education, momentary location, and activity) constant. We suspect that this may be due to well-documented marketing tactics used by the tobacco industry targeting those communities [47]. As a result, these vulnerable communities might have a higher density and proportion of retailers selling tobacco or a higher number of tobacco marketing materials per retailer, leading to their community members’ increased exposure to tobacco marketing regardless of their own individual risk factors for such exposure. Additional research is needed to understand how best to mitigate escalated tobacco marketing in vulnerable communities.
About half of the tobacco marketing exposure moments reported in the study were for flavored tobacco products. This is concerning given that flavored tobacco marketing may glamorize tobacco products and make tobacco use seem more attractive, less harmful, and easier to use [48, 49]. We also found that vulnerable community residence is the strongest predictor for flavored tobacco marketing, and its association is much stronger than that of tobacco marketing exposure in general. These findings once again reflect that the tobacco industry targets vulnerable communities in selling and promoting flavored tobacco products, which may contribute to higher flavored tobacco use prevalence among certain population groups, including Black residents and individuals of low-socioeconomic statuses [50, 51]. This finding is also consistent with other studies finding that exterior menthol cigarette advertising is more prevalent in communities with more Black residents and in neighborhoods with the household lowest income [52]. This differential use may further increase health disparities since flavored tobacco use, including smoking menthol cigarettes and flavored cigarillos, is associated with higher tobacco product use satisfaction, frequency, and duration as compared to non-flavored product use [47, 53, 54].
Seeing others using tobacco in some locations, an indicator of pro-tobacco-using environments and potential gaps in smoke-free policies, is also strongly associated with real-time tobacco and flavored tobacco marketing exposure. This may indicate that tobacco use is common in locations (e.g., retailers/public transits/restaurants/bars) where tobacco marketing, promotion, and sales are concurrently present. Although Washington, D.C., has widespread indoor workplace smoke-free policy protections including at restaurants and retail stores, such policies do not extend to outdoor or exterior locations at such businesses [55]. As such, these locations may serve as community “hotspots” [56] of exposure to pro-tobacco messages, tobacco use cues, and tobacco sales and promotion, which may greatly shape young people’s perceptions and behavior related to tobacco use. Future studies are recommended to use EMA equipped with geo-tracking technologies for examining those community-level hotspots to further inform the advancement of local tobacco control policies.
Strengths
The advantage of using EMA random surveys, rather than participant-or event-initiated surveys, to capture tobacco marketing exposure is that they discourage participants from consciously or unconsciously seeking tobacco marketing during the EMA study period. Focusing on non-current tobacco users who are not exposed to tobacco marketing due to the purchase of tobacco products is also different from prior studies that included both users and non-users [31, 33]. Therefore, the tobacco marketing moments captured by this study may closely represent real-time exposure in natural contexts among young adults who were not current tobacco users. The EMA approach used by this study may be replicated by future research investigating real-time tobacco and flavored tobacco marketing exposure among socially and economically diverse populations. Additionally, study participants were recruited from Washington, D.C., an ideal location to examine place-based disparities in tobacco marketing exposure as this locality has highest community inequity in smoking prevalence among largest cities in country [57].
Limitations
This study has several limitations. First, the EMA survey questions did not differentiate specific tobacco products when assessing marketing exposure, limiting our ability to detect the occurrences of, or the risk factors for, marketing exposure of various tobacco products. Second, the compliance rate of the study is moderate as compared to other EMA studies assessing tobacco marketing among young adults [32, 33]. This may be partially caused by the switching of platforms from Illumivu to Survey Signal during data collection. However, our sensitivity analysis showed that the results from the models of assessing baseline and real-time correlates only slightly differed using the full sample and the sample with an EMA compliance of 20% and higher. Third, the tobacco marketing exposure reported by the participants may have occurred outside of participants’ communities of residence. Future studies are recommended to use EMA equipped with geo-tracking technologies [58] for investigating the exact locations of tobacco marketing exposure and community-level “hotspots” to further inform the advancement of local tobacco control policies. Additionally, this study did not examine exposure to other types of tobacco marketing materials such as tobacco packages. Future research is warranted to capture the momentary exposure to more types of tobacco marketing materials as they may also increase young adults’ curiosity and intention of trying tobacco products. Finally, future studies may consider increase study compensation in order to improve EMA compliance. This is particularly important for equitable research with minoritized participants.
Policy Implications
Our study has significant implications for tobacco marketing regulations. First and foremost, policy initiatives aimed at directly reducing the presence of tobacco marketing in retail settings, such as prohibiting the exterior display of tobacco advertisements [59, 60], should be prioritized. Local policies to reduce tobacco retailer density through limiting tobacco retailer licensing [33] in vulnerable communities may also be instrumental in minimizing young people’s exposure to tobacco products and marketing. Additionally, strategies to curtail flavored tobacco marketing and product displays, including flavored tobacco sales restrictions, are greatly needed among localities with high concentrations of vulnerable populations [61, 62]. The FDA’s recently proposed plan on banning menthol cigarettes and flavored cigar products from the market [63] may help reduce the exposure to flavored tobacco products and marketing among members of vulnerable communities. These policy strategies may also reduce tobacco-related disparities associated with tobacco use, given vulnerable communities’ differential exposure to tobacco and flavored tobacco marketing.
Conclusion
For young adults who are non-current tobacco users, real-time exposure to tobacco marketing tends to occur with increased frequency among residents of vulnerable communities and at retailers and locations where people are using tobacco products. Residing in vulnerable communities put young adults at great risks of flavored tobacco marketing exposure and thereby future tobacco initiation and escalated use. Tobacco control measures and policies that act to reduce the amount of tobacco marketing particularly among vulnerable communities with high concentrations of Black residents and individuals with low-socioeconomic statuses are greatly needed. Those actions may diminish the risk of tobacco exposure and tobacco use for young adults and enhance health equity.
Supplementary Material
Acknowledgments
The authors thank the Truth’s research associates who helped develop the first round of the study: Ashley Mayo, MPH, and Emily Harvey, MA. The authors also thank interns and other Truth staff supporting with recruitment including flyering and distributing recruitment materials all over D.C.: Talene Appleton, Jordan Schneider, Allyson Buffett, Nicole Gonzalez, Justin Haigler.
This study was funded by a grant through the National Cancer Institute (R21CA208206). JCS is supported by the Division Intramural Research, National Institute on Minority Health and Health Disparities, and a grant from the National Cancer Institute and the Food and Drug Administration (K99CA242589). AMC is partially supported by FY21 Oklahoma Tobacco Settlement Endowment Trust (TSET) contract number R21-02 and a grant from the National Cancer Institute (P30CA225520) awarded to the Stephenson Cancer Center. The content is solely the responsibility of the authors and does not necessarily represent the official views of the US Government, Department of Health and Human Services, National Institutes of Health, National Institute on Minority Health and Health Disparities, or Food and Drug Administration. Although author AAR is a Food and Drug Administration employee, this work was not done as part of his official duties. This publication reflects the views of the author and should not be construed to reflect the Food and Drug Administration’s views or policies.
Contributor Information
Julia C Chen-Sankey, Division of Intramural Research, National Institute on Minority Health and Health Disparities, Bethesda, MD, USA.
Judy van de Venne, Center for Health Equity Transformation, Department of Behavioral Science, College of Medicine, University of Kentucky, Lexington, KY, USA.
Susan Westneat, Center for Health Equity Transformation, Department of Behavioral Science, College of Medicine, University of Kentucky, Lexington, KY, USA.
Basmah Rahman, Truth Initiative Schroeder Institute®, Washington, DC, USA.
Shanell Folger, Truth Initiative Schroeder Institute®, Washington, DC, USA.
Andrew Anesetti-Rothermel, Center for Health Equity Transformation, Department of Behavioral Science, College of Medicine, University of Kentucky, Lexington, KY, USA; Center for Tobacco Products, US Food and Drug Administration, Silver Spring, MD, USA.
Charles Debnam, Community Wellness Alliance, Washington, DC, USA.
Kurt M Ribisl, Department of Health Behavior, Gillings School of Global Public Health, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA.
Amy Cohn, TSET Health Promotion Research Center, University of Oklahoma Health Sciences Center, Oklahoma City, OK, USA; Department of Pediatrics, University of Oklahoma Health Sciences Center, Oklahoma City, OK, USA.
Shyanika W Rose, Center for Health Equity Transformation, Department of Behavioral Science, College of Medicine, University of Kentucky, Lexington, KY, USA; Markey Cancer Center, University of Kentucky, Lexington, KY, USA.
Compliance with Ethical Standards
Conflicts of Interest Dr. Kurt Ribisl is a paid expert consultant in litigation against electronic cigarette companies. Although author Dr. Andrew Anesetti-Rothermel is an employee of the U.S. Food and Drug Administration (FDA), this work was not done as part of his official duties at FDA. This publication reflects the views of the author and should not be construed to reflect the FDA/CTP’s views or policies. Other authors have no conflicts of interest to declare.
Ethical Adherence Study procedures were approved by the Advarra IRB (formerly Chesapeake IRB; Protocol: Pro00020538) and the University of Kentucky IRB (Protocol: 51579).
References
- 1. Stanton CA, Sharma E, Seaman EL, et al. Initiation of any tobacco and five tobacco products across 3 years among youth, young adults and adults in the USA: findings from the PATH Study Waves 1-3 (2013-2016). Tob Control. 2020;29:s178–s190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Barrington-Trimis JL, Braymiller JL, Unger JB, et al. Trends in the age of cigarette smoking initiation among young adults in the US from 2002 to 2018. JAMA Netw Open. 2020;3:e2019022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Dai H, Leventhal AM. Prevalence of e-cigarette use among adults in the United States, 2014-2018. JAMA. 2019;322:1824–1827. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Perry CL, Pérez A, Bluestein M, et al. Youth or young adults: which group is at highest risk for tobacco use onset? J Adolesc Health. 2018;63:413–420. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Kasza KA, Ambrose BK, Conway KP, et al. Tobacco-product use by adults and youths in the United States in 2013 and 2014. N Engl J Med. 2017;376:342–353. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Centers for Disease Control and Prevention. A Report of the Surgeon General: Preventing Tobacco Use Among Youth and Young Adults.2012. Accessed from: https://www.ncbi.nlm.nih.gov/books/NBK99237/
- 7. Henriksen L, Schleicher NC, Feighery EC, Fortmann SP. A longitudinal study of exposure to retail cigarette advertising and smoking initiation. Pediatrics. 2010;126:232–238. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Chen-Sankey JC, Unger JB, Bansal-Travers M, Niederdeppe J, Bernat E, Choi K. E-cigarette marketing exposure and subsequent experimentation among youth and young adults. Pediatrics. 2019; 144:e20191119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Hafez N, Ling PM. How Philip Morris built Marlboro into a global brand for young adults: implications for international tobacco control. Tob Control. 2005;14:262–271. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Freeman B. New media and tobacco control. Tob Control. 2012;21:139–144. [DOI] [PubMed] [Google Scholar]
- 11. Gilpin EA, Pierce JP, Rosbrook B. Are adolescents receptive to current sales promotion practices of the tobacco industry? Prev Med. 1997;26:14–21. [DOI] [PubMed] [Google Scholar]
- 12. Feighery E, Borzekowski DL, Schooler C, Flora J. Seeing, wanting, owning: the relationship between receptivity to tobacco marketing and smoking susceptibility in young people. Tob Control. 1998;7:123–128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Biener L, Albers AB. Young adults: vulnerable new targets of tobacco marketing. Am J Public Health. 2004;94:326–330. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. U.S. Department of Health and Human Services. The health consequences of smoking—50 years of progress: a report of the Surgeon General. Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2014: 17. [Google Scholar]
- 15. Chen-Sankey JC, Mead EL, Le D, Quisenberry A, Delnevo C, Choi K. Cigar smoking patterns and disparities by product type and race/ethnicity—a nationally representative study among U.S. adults. Am J Prev Med. 2021; 60 (1), 87–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Chen-Sankey JC, Choi K, Kirchner TR, Feldman RH, Butler J 3rd, Mead EL. Flavored cigar smoking among African American young adult dual users: an ecological momentary assessment. Drug Alcohol Depend. 2019;196:79–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Chen JC, Green KM, Arria AM, Borzekowski DLG. Prospective predictors of flavored e-cigarette use: a one-year longitudinal study of young adults in the U.S. Drug Alcohol Depend. 2018;191:279–285. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Villanti AC, Richardson A, Vallone DM, Rath JM. Flavored tobacco product use among U.S. young adults. Am J Prev Med. 2013;44:388–391. [DOI] [PubMed] [Google Scholar]
- 19. Ribisl KM, D’Angelo H, Feld AL, et al. Disparities in tobacco marketing and product availability at the point of sale: results of a national study. Prev Med. 2017;105:381–388. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Roberts ME, Berman ML, Slater MD, Hinton A, Ferketich AK. Point-of-sale tobacco marketing in rural and urban Ohio: could the new landscape of tobacco products widen inequalities? Prev Med. 2015;81:232–235. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Yerger VB, Przewoznik J, Malone RE. Racialized geography, corporate activity, and health disparities: tobacco industry targeting of inner cities. J Health Care Poor Underserved. 2007;18:10–38. [DOI] [PubMed] [Google Scholar]
- 22. Cruz TB, Wright LT, Crawford G. The menthol marketing mix: targeted promotions for focus communities in the United States. Nicotine Tob Res. 2010;12Suppl 2:S147–S153. [DOI] [PubMed] [Google Scholar]
- 23. Burton S, Clark L, Jackson K. The association between seeing retail displays of tobacco and tobacco smoking and purchase: findings from a diary-style survey. Addiction. 2012;107:169–175. [DOI] [PubMed] [Google Scholar]
- 24. Moran MB, Heley K, Pierce JP, Niaura R, Strong D, Abrams D. Ethnic and socioeconomic disparities in recalled exposure to and self-reported impact of tobacco marketing and promotions. Health Commun. 2019;34:280–289. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Lienemann BA, Rose SW, Unger JB, et al. Tobacco advertisement liking, vulnerability factors, and tobacco use among young adults. Nicotine Tob Res. 2019;21:300–308. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Assari S. Association of educational attainment and race/ethnicity with exposure to tobacco advertisement among US young adults. JAMA Netw Open. 2020;3:e1919393. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Burke LE, Shiffman S, Music E, et al. Ecological momentary assessment in behavioral research: addressing technological and human participant challenges. J Med Internet Res. 2017;19:e77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Scharf DM, Martino SC, Setodji CM, Staplefoote BL, Shadel WG. Middle and high school students’ exposure to alcohol- and smoking-related media: a pilot study using ecological momentary assessment. Psychol Addict Behav. 2013;27:1201–1206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Roberts ME, Keller-Hamilton B, Hinton A, et al. The magnitude and impact of tobacco marketing exposure in adolescents’ day-to-day lives: an ecological momentary assessment (EMA) study. Addict Behav. 2019;88:144–149. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Hébert ET, Vandewater EA, Businelle MS, Harrell MB, Kelder SH, Perry CL. Feasibility and reliability of a mobile tool to evaluate exposure to tobacco product marketing and messages using ecological momentary assessment. Addict Behav. 2017;73:105–110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Roberts ME, Lu B, Browning CR, Ferketich AK. Tracking young adults’ attitudes toward tobacco marketing using ecological momentary assessment (EMA). Subst Use Misuse. 2017;52:1219–1224. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Shadel WG, Martino SC, Setodji C, Scharf D. Momentary effects of exposure to prosmoking media on college students’ future smoking risk. Health Psychol. 2012;31:460–466. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Martino SC, Scharf DM, Setodji CM, Shadel WG. Measuring exposure to protobacco marketing and media: a field study using ecological momentary assessment. Nicotine Tob Res. 2012;14:398–406. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Rose SW, Anesetti-Rothermel A, Elmasry H, Niaura R. Young adult non-smokers’ exposure to real-world tobacco marketing: results of an ecological momentary assessment pilot study. BMC Res Notes. 2017;10:435. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Hamilton CM, Strader LC, Pratt JG, et al. The PhenX Toolkit: get the most from your measures. Am J Epidemiol. 2011;174:253–260. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. District of Columbia Department of Health. District of Columbia Behavioral Risk Factor Surveillance System (BRFSS) Annual Health Report.2018. Accessed from: https://dchealth.dc.gov/sites/default/files/dc/sites/doh/publication/attachments/BRFSS%202017%20Annual%20Report%20Final.pdf
- 37. Centers for Disease Control and Prevention. BRFSS Prevalence & Trends Data. Accessed from: https://www.cdc.gov/brfss/brfssprevalence/index.html.
- 38. Bryk AS, Raudenbush SW.. Hierarchical linear models: applications and data analysis methods. New York, NY: Sage Publications, Inc, 1992. [Google Scholar]
- 39. Wolfinger R, O’connell M. Generalized linear mixed models a pseudo-likelihood approach. J Stat Comput Simul. 1993; 48:233–243. [Google Scholar]
- 40. Breslow NE, Clayton DG. Approximate inference in generalized linear mixed models. J Am Stat Assoc. 1993; 88:9–25. [Google Scholar]
- 41. Jones A, Remmerswaal D, Verveer I, et al. Compliance with ecological momentary assessment protocols in substance users: a meta-analysis. Addiction. 2019;114:609–619. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. King JL, Wagoner KG, Reboussin BA, Sutfin EL. Youth and young adult convenience store behaviors: findings from intercept surveys. Nicotine Tob Res. 2020;22:2092–2097. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. D’Angelo H, Patel M, Rose SW. Convenience store access and e-cigarette advertising exposure is associated with future e-cigarette Initiation among tobacco-naïve youth in the PATH study (2013-2016). J Adolesc Health. 2021;68:794–800. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Trapl E, Anesetti-Rothermel A, Moore SP, Gittleman H. Association between school-based tobacco retailer exposures and young adolescent cigarette, cigar and e-cigarette use. Tob Control. 2020. doi: 10.1136/tobaccocontrol-2020-055764 [DOI] [PubMed] [Google Scholar]
- 45. Iglesias-Rios L, Parascandola M. A historical review of R.J. Reynolds’ strategies for marketing tobacco to Hispanics in the United States. Am J Public Health. 2013;103:e15–e27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46. Laws MB, Whitman J, Bowser DM, Krech L. Tobacco availability and point of sale marketing in demographically contrasting districts of Massachusetts. Tob Control. 2002;11Suppl 2:ii71–ii73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47. Cruz TB, Rose SW, Lienemann BA, et al. Pro-tobacco marketing and anti-tobacco campaigns aimed at vulnerable populations: a review of the literature. Tob Induc Dis. 2019;17:68. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48. Camenga DR, Fiellin LE, Pendergrass T, Miller E, Pentz MA, Hieftje K. Adolescents’ perceptions of flavored tobacco products, including e-cigarettes: a qualitative study to inform FDA tobacco education efforts through videogames. Addict Behav. 2018;82:189–194. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49. Chen-Sankey JC, Kong G, Choi K. Perceived ease of flavored e-cigarette use and e-cigarette use progression among youth never tobacco users. Plos One. 2019;14:e0212353. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50. Harrell MB, Loukas A, Jackson CD, Marti CN, Perry CL. Flavored tobacco product use among youth and young adults: what if flavors didn’t exist? Tob Regul Sci. 2017;3: 168–173. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51. Gardiner PS. The African Americanization of menthol cigarette use in the United States. Nicotine Tob Res. 2004;6Suppl 1:S55–S65. [DOI] [PubMed] [Google Scholar]
- 52. Mills SD, Henriksen L, Golden SD, et al. Disparities in retail marketing for menthol cigarettes in the United States, 2015. Health Place. 2018;53:62–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53. Villanti AC, Johnson AL, Glasser AM, et al. Association of flavored tobacco use with tobacco initiation and subsequent use among US youth and adults, 2013–2015. JAMA Netw Open. 2019; 2:e1913804–e1913804. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54. Cohn AM, Ganz O, Dennhardt AA, et al. Menthol cigarette smoking is associated with greater subjective reward, satisfaction, and “throat hit”, but not greater behavioral economic demand. Addict Behav. 2020;101:106108. [DOI] [PubMed] [Google Scholar]
- 55. Smoke-Free and Tobacco Laws. District of Columbia Department of Health. Accessed from: https://dchealth.dc.gov/service/smoke-free-and-tobacco-laws
- 56. Li W, Land T, Zhang Z, Keithly L, Kelsey JL. Small-area estimation and prioritizing communities for tobacco control efforts in Massachusetts. Am J Public Health. 2009;99: 470–479. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57. Leas EC, Schleicher NC, Prochaska JJ, Henriksen L. Place-based inequity in smoking prevalence in the largest cities in the United States. JAMA Intern Med. 2019;179:442–444. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58. Kirchner TR, Shiffman S. Spatio-temporal determinants of mental health and well-being: advances in geographically-explicit ecological momentary assessment (GEMA). Soc Psychiatry Psychiatr Epidemiol. 2016;51:1211–1223. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59. Kong AY, King BA. Boosting the tobacco control vaccine: recognizing the role of the retail environment in addressing tobacco use and disparities. Tob Control. doi: 10.1136/tobaccocontrol-2020-055722 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 60. Henriksen L. Comprehensive tobacco marketing restrictions: promotion, packaging, price and place. Tob Control. 2012;21:147–153. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 61. Rose SW, Amato MS, Anesetti-Rothermel A, et al. Characteristics and reach equity of policies restricting flavored tobacco product sales in the United States. Health Promot Pract. 2020;21:44S–53S. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 62. Chen JC, Green KM, Chen J, Hoke KS, Borzekowski DL. Restricting the sale of flavored e-cigarettes in the US: an examination of local regulations. Tob Regul Sci. 2018; 4:32–40. [Google Scholar]
- 63. The U.S. Food and Drug Administration. FDA commits to evidence-based actions aimed at saving lives and preventing future generations of smokers. 2021. Accessed from: https://www.fda.gov/news-events/press-announcements/fda-commits-evidence-based-actions-aimed-saving-lives-and-preventing-future-generations-smokers
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.