Discordant device/brand reporting among adolescents who used e-cigarettes in the National Youth Tobacco Survey

In a recent issue of Nicotine & Tobacco Research, Kaplan et al.,¹ using data from the Population Assessment of Tobacco and Health (PATH) Study of US adults, found that a significant proportion (18.2%) of respondents who used electronic nicotine delivery systems (ENDS) gave reports of the brand they used and the device type they used that were discordant with each other. For example, Kaplan et al. report that 17.4% of adults who reported use of JUUL-brand ENDS, a closed prefilled pod system, did not correctly indicate that their device uses replaceable prefilled cartridges. We agree with Kaplan et al. that accurate monitoring of characteristics of ENDS use and device types in tobacco surveys is crucial for regulatory and health authorities to assess the public health impact of these products, and to target interventions. Kaplan et al. reported on data from US adults.

An important question is whether brand/device reporting discordance also applies to adolescents. According to Omrani et al.,² “[m]ethodological knowledge on surveying young adolescents is scarce and researchers often rely on theories and methodological studies based on adult respondents. However, young adolescents are in the process of developing their cognitive, psychological, emotional and social skills, therefore present a unique set of considerations.” Thus, it is not immediately clear whether the discordance observed by Kaplan et al. among adults would also be observed in a survey of adolescents. This question is important because if brand/device discordance is apparent in adolescent surveys, it may impair the clarity and practical utility of brand- and device-specific estimates of adolescent e-cigarette use, in what is an extremely diverse market.

In this Letter, we extend the findings of Kaplan et al. to US adolescents (middle and high school students) using data from the National Youth Tobacco Survey (NYTS), a nationally representative, probabilistic, cross-sectional survey of tobacco use among US adolescents fielded by the US Centers for Disease Control and Prevention and supported by the Center for Tobacco Products of the US Food and Drug Administration. We combined the 2021 and 2022 NYTS surveys (with identical questionnaires; findings were similar in each year) to obtain larger sample sizes (N = 20 413 in 2021 and N = 28 291 in 2022). Unweighted sample sizes are reported; statistics are weighted with NYTS poststratification weights to be nationally representative. Analyses were conducted in SAS v. 9.4 with survey procedures to take the complex survey design into account. The NYTS data and our statistical analysis code are freely available online at https://www.cdc.gov/tobacco/data_statistics/surveys/nyts/ and https://doi.org/10.17605/OSF.IO/FV8CY.

Respondents in NYTS were asked “During the past 30 days, what brand of e-cigarettes did you usually use? (Choose only one answer)”; and “Which of the following best describes the type of e-cigarette you have used in the past 30 days? If you have used more than one type, please think about the one you use most often” (response options were: “An e-cigarette that uses pre-filled or refillable pods or cartridges,” “A disposable e-cigarette,” “An e-cigarette with a tank that you refill with liquids (including mod systems that can be customized by the user),” or “I don’t know the type”). Following the methods of Kaplan et al., we cross-tabulated the brand of ENDS a respondent reported “usually” using and the device type they reported using “most often” to assess the concordance between self-reported brand and device type. We necessarily excluded brand names under which multiple device types were sold in the United States at the time of the survey (eg, NYTS assesses use of the NJOY brand but does not distinguish between disposable and pod/cartridge NJOY devices). NYTS did not explicitly assess exclusively tank-based brands.

Among adolescents who reported “usually” using an exclusively pod/cartridge-based ENDS brand (JUUL, Logic, or Vuse; N = 722), 63.4% reported that their “most often” used device type was a pod/cartridge device. The remaining 36.6% of these respondents instead reported that their “most often” used device type was disposable (26.1%), tank (5.3%), or that they don’t know (5.1%).

Among adolescents who reported “usually” using an exclusively disposable ENDS brand (Mojo, Posh, Puff Bar, or Stig; N = 867), 80.7% reported that their “most often” used device type was disposable. The remaining 19.3% of these respondents instead reported that their “most often” used device type was a pod/cartridge device (11.6%), tank (4.5%), or that they “don’t know” (3.2%).

From the above it is clear that discordance between reported ENDS brand and device type is not unique to adults nor to the PATH dataset; ENDS brand/device discordance is also abundant among adolescents in the NYTS dataset. While the sources of this discordance are not exactly known, some discordance could reflect noise from nonserious answering behavior, which could be addressed with seriousness checks,³ and other discordance could reflect respondents either not knowing what is meant by each device type or applying incorrect names to the brand of e-cigarette they used. We support the recommendations made by Kaplan et al. that including images of brand-specific devices may help to classify ENDS device types more accurately in tobacco surveys, including PATH and NYTS.

Funding

The preparation of this work was funded by Juul Labs Inc.

Declaration of Interests

Through Pinney Associates, FF provides consulting services on tobacco harm reduction on an exclusive basis to Juul Labs Inc. (JLI). The preparation of this work was supported by JLI, and JLI reviewed and provided comments on a draft manuscript. The content and the decision to publish are the responsibility of the author.

Author Contributions

Floe Foxon (Conceptualization [lead], Data curation [lead], Formal analysis [lead], Investigation [lead], Methodology [lead], Software [lead], Writing—original draft [lead])

Data Availability

The dataset supporting the conclusions of this article are freely available in the National Youth Tobacco Survey repository, https://www.cdc.gov/tobacco/data_statistics/surveys/nyts/. Statistical analysis code is made available in the online supplementary materials at doi:10.17605/OSF.IO/FV8CY.