pediatrics
June 2017, VOLUME139 /ISSUE 6

# Receptivity to Tobacco Advertising and Susceptibility to Tobacco Products

1. John P. Pierce, PhDa,b,
2. James D. Sargent, MDc,
3. Martha M. White, MSa,
4. Nicolette Borek, PhDd,
5. David B. Portnoy, PhD, MPHd,
6. Victoria R. Green, BAe,f,
7. Annette R. Kaufman, PhD, MPHg,
8. Cassandra A. Stanton, PhDh,i,
9. Maansi Bansal-Travers, PhDj,
10. David R. Strong, PhDa,b,
11. Jennifer L. Pearson, PhD, MPHk,l,
12. Blair N. Coleman, PhD, MPHd,
13. Eric Leas, MPHa,b,
16. Meghan B. Moran, PhDk,
17. Charles Carusi, PhDh,
18. Andrew Hyland, PhDj, and
19. Karen Messer, PhDa,b
1. aCancer Prevention Program, Moores Cancer Center, and
2. bDepartment of Family Medicine and Public Health, University of California, San Diego, La Jolla, California;
3. cC. Everett Koop Institute, Norris Cotton Cancer Center, Geisel School of Medicine at Dartmouth, Hanover, New Hampshire;
4. dCenter for Tobacco Products, US Food and Drug Administration, Silver Spring, Maryland;
5. eNational Institute on Drug Abuse, National Institutes of Health, Bethesda, Maryland;
6. fKelly Government Solutions, Rockville, Maryland;
7. gTobacco Control Research Branch, Behavioral Research Program, Division of Cancer Control and Population Sciences, National Cancer Institute, National Institutes of Health, Rockville, Maryland;
8. hWestat, Rockville, Maryland;
9. iDepartment of Oncology, Cancer Prevention and Control Program, Lombardi Comprehensive Cancer Center, Georgetown University Medical Center, Washington, District of Columbia;
10. jDepartment of Health Behavior, Roswell Park Cancer Institute, Buffalo, New York;
11. kDepartment of Health, Behavior, and Society, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland; and
12. lSchroeder Institute for Tobacco Research and Policy Studies at Truth Initiative, Washington, District of Columbia
1. Dr Pierce conceptualized and designed the study and drafted the initial manuscript; Dr Sargent conceptualized and designed the study including the study instruments and critically reviewed the manuscript; Ms White carried out the initial analyses and reviewed and revised the manuscript; Dr Messer designed the analyses and reviewed and revised the manuscript; Drs Carusi and Hyland lead the PATH project, including the design of survey instruments, and reviewed and revised the manuscript; and Drs Borek, Portnoy, Kaufman, Stanton, Bansal-Travers, Strong, Pearson, Coleman, Trinidad, and Moran and Ms Green, Ms Noble, and Mr Leas reviewed and approved the initial proposed analyses and reviewed and revised the manuscript; and all authors approved the final manuscript as submitted.

## Abstract

BACKGROUND AND OBJECTIVES: Non–cigarette tobacco marketing is less regulated and may promote cigarette smoking among adolescents. We quantified receptivity to advertising for multiple tobacco products and hypothesized associations with susceptibility to cigarette smoking.

METHODS: Wave 1 of the nationally representative PATH (Population Assessment of Tobacco and Health) study interviewed 10 751 adolescents who had never used tobacco. A stratified random selection of 5 advertisements for each of cigarettes, e-cigarettes, smokeless products, and cigars were shown from 959 recent tobacco advertisements. Aided recall was classified as low receptivity, and image-liking or favorite ad as higher receptivity. The main dependent variable was susceptibility to cigarette smoking.

CONCLUSIONS: A large proportion of US adolescent never tobacco users are receptive to tobacco advertising, with television advertising for e-cigarettes having the highest recall. Receptivity to advertising for each non–cigarette tobacco product was associated with susceptibility to smoke cigarettes.

• Abbreviations:
CI
confidence interval
aOR
P1
period 1
P2
period 2
PATH
Population Assessment of Tobacco and Health
• #### What’s Known on This Subject:

The consensus that cigarette marketing is 1 cause of adolescent smoking is the basis for marketing constraints imposed on these products in the United States and elsewhere. Little is known about the influence of marketing for non–cigarette tobacco products.

Advertising for non–cigarette tobacco products is reaching non–tobacco-using US adolescents, especially through television. High percentages of non–tobacco-using adolescents recognize tobacco ad images, including e-cigarette ads. Receptivity to these ads is associated with susceptibility to future cigarette smoking.

Susceptibility to smoking is a validated measure that predicts the risk of smoking initiation as many as 3 to 4 years before first experimentation.2325 A series of questions identifies “committed never users” as those who have never been curious about use, have strong intentions not to use, and who would resist an offer to use from a best friend. All other never users are considered susceptible. Susceptibility and subsequent experimentation vary across sociodemographic variables, receptivity to tobacco marketing,16,17 exposure to other tobacco users,26 use of another psychoactive substance,27 use of another tobacco product,28 and psychosocial variables such as externalizing problem behaviors (eg, rule-breaking, aggression),29 internalizing problem behaviors (eg, depression, anxiety),30 and sensation-seeking.31

This study explores levels of receptivity to the marketing of e-cigarettes, cigarettes, cigars, and smokeless tobacco products among never-using 12 to 17 year olds, with the use of data from the national Population Assessment of Tobacco and Health (PATH) Study. The study design allowed us to assess adolescent receptivity to advertising for multiple tobacco products and to summarize this in a measure of receptivity to any tobacco advertising. We investigated whether this general receptivity to tobacco advertising was associated with susceptibility to use any tobacco product. Finally, we investigated the associations of advertising receptivity to each product with susceptibility to cigarette smoking.

## Methods

Data are from wave 1 of the PATH study, which is a nationally representative sample of the civilian, noninstitutionalized US population, aged ≥12 years between September 2013 and December 2014.32,33 With oversight from the National Institute of Drug Abuse and the Food and Drug Administration, Westat collected data with the use of audio-computer–assisted self-interviews. Westat’s Institutional Review Board approved the study design/protocol, and the Office of Management and Budget approved the data collection. Households were identified by using an address-based, area-probability sampling method, and a screener survey enumerated household members (response rate = 54%). Generally, all youth aged 12 to 17 years (maximum 2 per household) were selected for interview, and parental consent and youth assent were obtained. Interviews were completed for 78.4% of selected youth. The data were weighted to adjust for the complex sample design and nonresponse to allow population estimates. If the respondent did not answer regarding their age, sex, race, or Hispanic ethnicity, these were obtained from the household screener (n = 704 or 6.6%) or, if not available, by using statistical imputation methods (n = 143 or 1.3%).34 In this article, our analyses were restricted to the 10 751 of the youth sample (N = 13 651) who reported that they had never used any tobacco product.

### Dependent Variable: Susceptibility To Use Tobacco Products and Cigarettes

Following previous research,23,24 never users who had heard of the product were classified as either susceptible to use or committed never users on the basis of their responses to 3 questions assessing their curiosity about the product, intention to try it in the near future, and likely response if a best friend were to offer them the product (see Supplemental Information). Only those with the strongest rejection to all 3 questions were categorized as committed never users to each of the following 8 products (cigarettes, e-cigarettes, pipes, cigars, hookahs, smokeless tobacco, dissolvable tobacco, and bidis/kreteks). The strongest rejection to all questions for all products classified an individual as a committed never tobacco user; all others were considered to be susceptible to use ≥1 products. For this analysis, those who had never heard of a product were considered to be committed never users.

### Key Independent Variable: Receptivity to Tobacco Product Advertising

Respondents were asked “What is the brand of your favorite tobacco advertisement?” A list of brands was provided, with an option to nominate another brand. We assumed that identifying Marlboro or Camel as favorite referred to cigarettes, unless snus was specifically mentioned.

#### Aided Recall and Liking of Contemporary Ads

The above sampling scheme allowed some respondents to see >1 ad for a brand within a category, particularly for the little cigars and cigarillos set in P2. When >1 ad was shown for a brand, aided recall was assessed only on the first ad that was displayed. We computed the estimated recall rate as the weighted proportion of respondents who recalled seeing an ad for the brand out of the total number of respondents who were shown an ad for the brand. Before ranking brands on the frequency of recall, we removed brands with advertisements that were shown to <500 participants. Thus, for the rank order of aided recall, we considered a total of 55 tobacco brands (9 cigarette brands, 9 e-cigarette brands with television ads, 14 e-cigarette brands with print ads, 9 smokeless brands, and 14 cigar brands).

### Other Covariates of Tobacco Use in Adolescence

#### Exposure to Tobacco Use

All respondents were asked: “Does anyone who lives with you now use any of the following: [list of tobacco products]” and “During the past 7 days, about how many hours were you around others who were smoking? Include time in your home, in a car, at school or outdoors.” We report 2 binary variables: 1 for household exposure (no users of tobacco in the household versus any) and 1 for general exposure (no exposure to smoking by others in the past 7 days versus any).

#### Use of Other Drugs or Alcohol

Ever use was ascertained for alcohol, marijuana, as well as for misuse of prescription drugs (ie, Ritalin/Adderall, painkillers, sedatives, tranquilizers), cocaine or crack, methamphetamine or speed, heroin, inhalants, solvents, and hallucinogens by a series of questions: “Have you ever used [substance]?” (See Supplemental Information for the full list of questions). Adolescents who reported ever use of any of these were classified as “any use”; all others were classified as “no use.”

#### Psychosocial Predictors of Tobacco Use

We adjusted for mental health and substance use problems by using scales from the Global Appraisal of Individual Needs–Short Screener.35 The internalizing subscale (α = .81) included 4 items of depressive and anxiety symptoms. The externalizing subscale (α = .70) included experience with 5 conduct and behavioral items. The substance use problems subscale (α = .67) asked 7 questions about problems associated with alcohol or drug use. Adolescents were scored on how many items they had experienced in the past month or past 2 to 12 months (see Supplemental Information for questions). In addition, sensation seeking (α = .74) was assessed by 3 items modified from the Brief Sensation Seeking Scale.36 On all scales, respondents were scored as having no symptoms (0), low to moderate symptom levels (1–2), or high symptom levels (≥3).

### Statistical Analyses

Analyses were performed by using SAS Software, version 9.3 (SAS Institute, Cary, NC).37 We used the survey weights supplied with the data, and computed variances and P values by using the recommended Balanced Repeated Replication method with Fay = 0.3.27,33,38 Weighted percentages were calculated with SAS version 9.3 PROC SURVEYFREQ. Modified Wilson confidence limits for proportions were calculated on the weighted estimates by using PROC SURVEYFREQ. Separate logistic regression models tested the association of receptivity to any tobacco advertising with susceptibility to any tobacco product, as well as with susceptibility to cigarette smoking. These models controlled for the above-mentioned covariates. χ2 Tests were also conducted to test for significant differences between percentage recall of pairs of selected ad brands or modes. Odds ratios, confidence limits, and P values are reported from the weighted, adjusted model (calculated via PROC SURVEYLOGISTIC).

## Results

### Demographic Characteristics

This sample of adolescent never tobacco users was balanced on sex and had a high proportion of younger adolescents (12–13 years: 39.5%; 14–15 years: 34.1%; 16–17 years: 26.4%) (Table 1). Just over half were non-Hispanic white (53.6%), 22.4% were Hispanic, 14.4% were non-Hispanic African American, 5.2% were Asian, and 4.5% were other races/ethnicities. Most adolescents (82.0%) had parents who completed high school, and 60.7% had parents with at least some college education. Never tobacco users who were not receptive to any tobacco advertising (54%) had the lowest susceptibility to any tobacco use (34.3%; 95% confidence interval [CI]: 32.9%–35.6%). Among never users with low receptivity (34%), 50.4% (95% CI: 48.5%–52.2%) were susceptible. Among the 10% of never users who had a moderate receptivity, 65.4% (95% CI: 62.2%–68.4%) were susceptible. Among the <2% of never users with high receptivity to tobacco advertising, 87.7% (95% CI: 81.2%–92.2%) were susceptible to any tobacco product.

TABLE 1

Logistic Regression Predicting Susceptibility to Any Tobacco Product Among Never-Using Adolescents

Overall, 41.0% of 12 to 13 year olds and approximately half of both 14 to 15 year olds and 16 to 17 year olds were classified as being receptive to any tobacco advertising (Table 2). Approximately one-third of each age group had a low level of receptivity. There were significantly fewer 12 to 13 year olds with moderate or high receptivity (8.8%) compared with 16- to 17-year-old adolescents (15.0%) (P < .001).

TABLE 2

Receptivity to Tobacco Advertising by Product Type Among Never Tobacco Users

In multivariable logistic regression (Table 1) controlling for potential confounding variables, even low receptivity to any tobacco ads was significantly associated with increased concurrent susceptibility to use any tobacco product (adjusted odds ratio [aOR]: 1.38; 95% CI: 1.25–1.53). Moderate and high receptivity levels had higher and significant odds ratios (moderate: aOR = 2.39, P < .001; high: aOR = 6.73, P < .001).

### Aided Recall for Tobacco Advertising by Brand

TABLE 3

Ads for Tobacco Brands With the Highest Aided Recall by Age Group Among US Never Tobacco Users

### Association of Advertising Receptivity With Susceptibility to Cigarette Smoking

In our multivariable logistic regression controlling for potential covariates, moderate to high receptivity to cigarettes (aOR: 1.57; 95% CI: 1.25–1.98), e-cigarettes (aOR: 1.53; 95% CI: 1.21–1.94), and smokeless tobacco (aOR: 1.58; CI: 1.23–2.03) was significantly associated with concurrent susceptibility to smoke cigarettes (Table 4), which was not the case for moderate to high receptivity to cigar advertising. A low level of receptivity to any of the 4 forms of tobacco advertising was not associated with concurrent susceptibility to smoke cigarettes.

TABLE 4

Variables Associated With Susceptibility to Cigarette Smoking Among Adolescent Never Tobacco Users

## Discussion

Despite marketing restrictions on cigarettes and other products, tobacco advertising continues to reach one-third to one-half of adolescents, depending on age. E-cigarette advertising generated the highest reach. This dominance of e-cigarettes is in stark contrast to the advertising expenditures for each product over the study period: >$9 billion/year for cigarettes,39$415 million/year for smokeless products,40 and only \$60 million/year for e-cigarettes4 (although this latter amount appears to be growing substantially each year). One explanation for receptivity being highest for the product with the lowest marketing budget4 is that the product with minimal marketing restrictions may have a significant advantage. Products with less-restricted marketing can use marketing synergies across multiple media channels, which are likely more cost-effective than campaigns for products with significant marketing restrictions.41 It has been noted that television is still among the most effective advertising platforms in the United States.42 In this study, 4 of the top 10 ads recalled were e-cigarette television ads, which made up only a portion of the e-cigarette marketing expenditures, a budget that was 2 orders of magnitude below the cigarette marketing budget.4,39 Our results suggest that, even though cigarettes are not allowed to be advertised on television, interest in them may be increased through observing ads for other tobacco products. This topic, along with how receptivity may affect harm perceptions, will be explored in separate longitudinal analyses of PATH study data.

Two major study strengths are its large nationally representative sample of adolescents and the use of a stratified random sample from a near census of recent tobacco advertising images to gauge reach and receptivity. A limitation is that our findings are cross-sectional and the directionality of associations cannot be determined. However, susceptibility has been a consistent precursor of risk of smoking initiation2 and, in the PATH study wave 1, susceptibility to use tobacco among 12 to 17 year olds appears to be equivalent to the prevalence of experimentation for those aged a few years older.43 Moreover, this sample will be followed longitudinally, allowing us to confirm whether both low and moderate/high receptivity to tobacco advertising predicts later tobacco use behavior among these never smokers.

## Conclusions

In this survey, 41% of US 12 to 13 year olds, as well as half of older adolescents who had never used tobacco, were receptive to tobacco adverting and this receptivity was associated with increased susceptibility to cigarette smoking, regardless of the type of tobacco product advertised. Indeed, there was no difference in the association between receptivity for advertising for the different tobacco products and susceptibility to cigarette smoking. E-cigarette advertising was the most recalled by US adolescents, particularly ads that were shown on television. This high level of recall was achieved despite e-cigarette advertising expenditures being a fraction of those for cigarette marketing.

## Footnotes

• Accepted March 15, 2017.
• Address correspondence to John P. Pierce, PhD, Department of Family Medicine and Public Health, Moores Cancer Center, 3855 Health Sciences Dr, MC 0901, University of California, San Diego, La Jolla, CA 92093-0901. E-mail: jppierce{at}ucsd.edu
• The views and opinions expressed in this presentation are those of the authors only and do not necessarily represent the views, official policy, or position of the US Department of Health and Human Services or any of its affiliated institutions or agencies.

• FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.

• FUNDING: This study is supported with federal funds from the National Institute on Drug Abuse, National Institutes of Health, and the Food and Drug Administration, Department of Health and Human Services, under a contract to Westat (contract HHSN271201100027C). Funded by the National Institutes of Health (NIH).

• POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.