Geographic Differences in JUUL Use and Risk Perceptions

Abstract

Purpose

The aim of this study was to compare prevalence of JUUL use and JUUL risk perceptions between adolescents living in an urban area versus Appalachian areas of the U.S.

Methods

Data were drawn from a prospective cohort study of adolescent males (the Buckeye Teen Health Study, or BTHS); our cross-sectional analysis used data from one timepoint, collected between January and December 2019 (N = 873). Chi-square tests and multivariable logistic regression compared JUUL use prevalence and risk perceptions between participants in an urban Ohio county and nine predominantly rural Appalachian Ohio counties.

Results

Over a quarter of the sample (29.2%) had ever used JUUL. In the unadjusted model, prevalence of JUUL use was similar between regions but Appalachian participants perceived JUUL as more harmful (p < .001) and more addictive (p = .04) than urban participants. In the adjusted model, region was not significantly related to current JUUL use (OR: 1.20, 95% CI: 0.77, 1.87) or ever JUUL use (OR: 1.15, 95% CI: 0.83, 1.60).

Conclusions

JUUL use was similar between urban and Appalachian participants despite regional differences in risk perceptions. Interventions that only target risk perceptions may not be sufficient to prevent adolescent e-cigarette use, particularly in rural communities.

Keywords:

• Adolescents
• e-cigarettes
• prevalence
• risk perception
• Appalachia

Introduction

Although combustible tobacco product use among U.S. adolescents has continued to decrease over the past decade (Gentzke et al., 2019), e-cigarette use has increased from 1.5% in 2011 to 11.3% in 2021 (Park-Lee et al., 2021). JUUL, a pod-based e-cigarette company whose initial marketing strategy targeted a younger audience (Huang et al., 2019; Kim et al., 2019), was so popular that in 2018 it accounted for almost 70% of all e-cigarette retail sales (Fargo, 2018). While adolescent JUUL use varies across demographic characteristics (Hammond et al., 2020), less is known about differences in use by urbanicity. Yet understanding whether there are geographic differences is important given established disparities for other tobacco product use, which contribute to substantial urban/rural health disparities (Zahnd et al., 2018). Studies on youth e-cigarette prevalence find similar rates in urban and rural areas; however, these studies report data that were collected prior to JUUL entering, and subsequently dominating, the market (Pesko & Robarts, 2017; Wiggins et al., 2020). Although JUUL discontinued the sale of most flavored pods during the study time period, it remains the most commonly reported usual brand among high school students (Wang et al., 2021).

The current study evaluated whether there are regional differences in JUUL use prevalence among adolescent males. Male users are a particularly important group to examine, as young adult data indicate that JUUL use may be twice as common among males compared to females (Ickes et al., 2020). Because risk perceptions are well established predictors of tobacco use, we also examined JUUL-related risk perceptions for regional differences (Parker et al., 2018; Russell et al., 2020; Song et al., 2009). These findings will provide information to guide the development of e-cigarette prevention programs for adolescents in different geographic regions.

Materials and methods

The Buckeye Teen Health Study is a longitudinal cohort study comprised of adolescent males in urban and Appalachian Ohio. Boys between the ages of 11 and 16 years living in urban Franklin County (which contains the state capital, Columbus) or in one of nine rural Appalachian Ohio counties were enrolled in the study between January 2015 and June 2016. Only males were included in this cohort because the goal of the parent study was to examine dual use of cigarettes and smokeless tobacco, and smokeless tobacco is rarely used by females (Wang et al., 2018). The majority of participants (N = 991) were recruited through address-based sampling and the remainder of participants (N = 229) were enrolled through various convenience sampling techniques such as attending community fairs and using respondent-driven sampling. Supplementing address-based recruitment with convenience sampling allowed us to achieve sufficient sample size and increase the diversity of the sample. After receiving written assent from the participant and written permission from their parent or guardian, an in-person interview was conducted; follow-up surveys were completed every 6-months in-person or by phone. To address the surge in JUUL use that was occurring nationally, items specifically about JUUL were added to the follow-up surveys in January 2019. The present study is a cross-sectional analysis from one follow-up timepoint during 2019. All procedures were approved by The Ohio State University’s Institutional Review Board. A detailed description of study methodology has previously been published (Friedman et al., 2018).

Of the 1220 adolescent males who entered the study, 873 completed at least one survey during 2019; if participants completed more than one survey, their most recent survey response was used. Participants’ demographic characteristics were collected at baseline as were parents’ self-reported level of education. The outcomes of interest, JUUL use prevalence and JUUL perceptions, were measured in 2019.

To assess prevalence, all participants were asked whether they had ever tried JUUL. Those who had tried JUUL and used a JUUL in the past 30 days were considered a current user; those who had tried JUUL but had not used a JUUL in the past 30 days were designated an ever user (Gentzke et al., 2019). To assess risk perceptions of JUUL, we asked two questions about JUUL’s harmfulness and addictiveness, drawn from the PATH study: “How much do you think people harm themselves when they use JUULs?” and “How likely is someone to become addicted to JUULs?” (United States Department of Health and Human Services. National Institutes of Health. National Institute on Drug Abuse & United States Department of Health and Human Services. Food and Drug Administration. Center for Tobacco Products, 2021). We additionally added an exploratory question about relative risk: “Compared to one pack of cigarettes, how much nicotine do you think is in one JUUL pod?” Response options and response frequencies to the questions are presented in Table 1.

Table 1. Sample demographics, JUUL-use behaviors and perceptions, Ohio Buckeye Teen Health Study (BTHS) male participants, 2019.

	Urban n = 531 (60.8%)	Appalachian n = 342 (39.2%)	Total n = 873	p-value^a
Age				.52
14–17 years	56.3% (n = 299)	54.1% (n = 185)	55.4% (n = 484)
18–21 years	43.7% (n = 232)	45.9% (n = 157)	44.6% (n = 389)
Race/Ethnicity				<.001
Non-Hispanic White	68.2% (n = 362)	92.7% (n = 317)	77.8% (n = 679)
Other	31.8% (n = 169)	7.3% (n = 25)	22.2% (n = 194)
Parental education				<.001
High school Degree/GED or less	8.1% (n = 43)	16.8% (n = 57)	11.5% (n = 100)
Some college/associate degree	25.5% (n = 135)	34.4% (n = 117)	29.0% (n = 252)
Bachelor’s degree or higher	66.4% (n = 352)	48.8% (n = 166)	59.5% (n = 518)
Ever JUUL use				.57
Never tried JUUL	70.4% (n = 369)	71.3% (n = 241)	70.8% (n = 610)
Ever tried JUUL	29.6% (n = 155)	28.7% (n = 97)	29.2% (n = 252)
Current JUUL use^b
Yes	40.6% (n = 63)	43.3% (n = 42)	41.7% (n = 105)	.68
No	59.4% (n = 92)	56.7% (n = 55)	58.3% (n = 147)
Perception of JUUL harm				<.001
No harm/little harm	9.2% (n = 47)	7.9% (n = 26)	8.6% (n = 73)
Some harm	55.8% (n = 286)	43.2% (n = 143)	50.8% (n = 429)
A lot of harm	35.1% (n = 180)	48.9% (n = 162)	40.5% (n = 342)
Perception of JUUL addictiveness				.04
Not likely	8.0% (n = 40)	9.3% (n = 30)	8.5% (n = 70)
Somewhat likely	43.4% (n = 218)	34.6% (n = 111)	40.0% (n = 329)
Very likely	48.6% (n = 244)	56.1% (n = 180)	51.5% (n = 424)
Perception of JUUL nicotine compared to cigarettes				.24
More nicotine in cigarette pack	9.8% (n = 47)	8.9% (n = 27)	9.4% (n = 74)
More nicotine in JUUL pod	34.4% (n = 165)	40.3% (n = 123)	36.7% (n = 288)
About the same	55.8% (n = 268)	50.8% (n = 155)	53.9% (n = 423)

a p-value indicated results of chi-square tests assessing whether the distribution of response vary by urban vs. Appalachian status.

b Of those who have ever tried JUUL.

The sample was first summarized using descriptive statistics and chi-square analyses; we also tested for urban/rural differences in our predictor and outcome variables. A series of univariate logistic regressions were next used to identify significant predictors of JUUL use (not shown). A final multivariable logistic regression was then used to compare JUUL use prevalence between urban and Appalachian participants while controlling for significant predictors from the univariate analyses predicting JUUL use. An alpha level of .05 was used to determine statistical significance.

Results

The 873 adolescent boys who completed a survey during 2019 were between the ages of 14 and 21 years. Most respondents were white (77.8%) and lived in an urban area (60.8%). Over a quarter of participants (29.2%) reported ever trying a JUUL and 12.0% reported currently using a JUUL. JUUL use prevalence did not differ between urban and Appalachian participants (Table 1).

Overall, less than 10% of participants thought JUUL caused little or no harm. However, Appalachian participants perceived JUUL as significantly more harmful (p<.001) than urban participants (e.g., 48.9% of Appalachian vs. 35.1% of urban youth selected “a lot of harm”). The majority of participants (51.5%) thought it was very likely someone would become addicted to JUUL, but perceived addictiveness also differed between urban and Appalachian participants (p=.04). Among urban participants, 48.6% thought it was “very likely” someone would become addicted to JUUL compared to 56.1% of Appalachian participants. More than half of participants (53.9%) correctly stated the amount of nicotine in a JUUL pod; no difference was found between regions (Table 1).

Results from our multivariable model (which included predictors significantly associated with current or ever JUUL use in univariate analyses) are presented in Table 2. As in the unadjusted model, region was not significantly associated with current or ever JUUL use. Participants who were older (aged 18-21 vs. 14-17) and who perceived JUUL as less (vs. more) harmful had higher odds of JUUL use.

Table 2. Odds ratios (ORs) and 95% confidence intervals (CI) for multivariable logistic regressions testing predictors of current and ever JUUL use, Ohio Buckeye Teen Health Study (BTHS) male participants, 2019.

	Current JUUL use		Ever JUUL use
	OR	95% CI	OR	95% CI
Region
Urban	Ref		Ref
Appalachian	1.20	0.77, 1.87	1.15	0.83, 1.60
Age
14–17	Ref		Ref
18–21	2.10	1.35, 3.27	2.20	1.60, 3.04
Perception of JUUL harm
No harm/some harm	Ref		Ref
Some harm	0.64	0.35, 1.18	0.72	0.42, 1.23
A lot of harm	0.14	0.06, 0.30	0.30	0.17, 0.54
Perception of JUUL addictiveness
Not likely	Ref		–	–
Somewhat likely	0.42	0.22, 0.80	–	–
Very likely	0.43	0.22, 0.85	–	–
Perception of JUUL nicotine compared to cigarettes
More in cigarette pack	–	–	Ref
More in JUUL pod	–	–	1.61	0.86, 3.02
About the same	–	–	2.27	1.25, 4.12

Bold indicates significance at alpha=.05.

– This variable was not significant in the univariate analysis and was therefore not included as a predictor for the multivariable analysis.

Note: Variables included in the multivariable model were significant predictors of JUUL use in univariate logistic regression analyses (not shown), not from the chi-square tests of urban/rural differences in Table 1.

Discussion

Although participants in Appalachian counties perceived JUUL to cause more harm and be more addictive than their urban counterparts, they were equally likely to use JUUL, which is consistent with prior research (Pesko & Robarts, 2017; Wiggins et al., 2020). The current study adds a novel finding at the intrapersonal level: Appalachian adolescents appear to perceive JUUL to be a riskier product than urban adolescents do—but, importantly, these differences did not translate to a reduced prevalence of use. While lower risk perceptions are an established predictor of smoking initiation among adolescents (Parker et al., 2018; Song et al., 2009), a recent study found lower risk perceptions were only predictive of e-cigarette initiation among nonsmokers (Cooper et al., 2018). Therefore, in rural areas where the prevalence of adolescent cigarette smoking is known to be higher (Wiggins et al., 2020), increased risk perceptions may not be as protective against e-cigarette use and could explain the similar use prevalence between regions in our study. Interventions addressing more than one level of influence can be more effective, and results can be maintained longer, than interventions that only focus on individual characteristics, like risk perceptions (Bond et al., 2004; DiClemente et al., 2007; Wen et al., 2009). This information combined with our results suggest that an intervention that only targets individual beliefs may be insufficient to reduce use, particularly among rural adolescents. However, further research is needed to determine effective evidence-based interventions that are equitable for both urban and rural youth.

Limitations

There are several limitations to note. First, our sample consisted of adolescent males living in Ohio at the time of study enrollment, which limits the generalizability of the findings to other locations and genders. Second, while JUUL was the most popular e-cigarette brand at the implementation of the survey, and prefilled pods/cartridges remained the most popular e-cigarette category in 2020, use of disposable e-cigarettes increased dramatically (Wang et al., 2020). It is possible that our findings are not generalizable to other brands or styles of e-cigarettes that have recently become more common for young people to use. Finally, we did not include external factors such as parental use or strength of tobacco control policies that would have helped to better contextualize the impact of risk perceptions.

Conclusions

While JUUL prevalence did not differ between urban and Appalachian adolescents, factors that are consistently associated with tobacco use differed based on location. This suggests that there may be unique, place-specific reasons driving youth to try emerging tobacco products. Although this study specifically examined JUUL, the results may translate to other e-cigarettes that are gaining popularity among youth. More research is required to determine which risk factors are most important to target and how the influence of these factors varies across geographic regions.

Declaration of interest statement

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

Additional information

Funding

This research was supported by grant P50CA180908 from the National Cancer Institute of the National Institutes of Health (NIH) and the Food and Drug Administration (FDA) Center for Tobacco Products. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH or the FDA.