The need to advance the measurement of exposure to digital tobacco content

The recent article by Mongilio et al. titled, “Patterns of ENDS-related content exposure on social media among U.S. adolescents and young adults,”1 examined how often adolescents and young adults reported seeing different types of electronic nicotine delivery systems (ENDS)-related content on social media. Participants were asked how often they encountered ENDS-related content on social media in the past week, and if so, what types of content they remembered, such as promotional posts, user-generated content, or public health messages. Using self-reported responses, the authors grouped individuals into exposure profiles (ie, based on content theme and source) using latent class analysis, identifying seven classes that reflected distinct patterns of such content exposure. This study suggested that exposure to ENDS-related content was heterogeneous, rather than uniform, with individuals differing in the types and frequency of content they encountered on social media. We believe this study underscores the need to further advance the measurement of exposure to digital tobacco content. However, without distinguishing exposure (eg, recall of tobacco-related digital content) from receptivity (ie, psychological engagement with such content),2 and without applying psychometric methods to validate behavioral constructs,3 it remains difficult to determine whether observed exposure patterns reflect passive contact, algorithmic targeting,4 or psychologically meaningful engagement with tobacco-related digital content. In this Letter to the Editor, we highlight the importance of distinguishing exposure from receptivity in digital tobacco marketing research and argue for the development and use of psychometrically validated measures to strengthen inference and inform tobacco control efforts. There is a need to develop and validate a measure of digital tobacco exposure. Validated measures are essential for reliability, quantifying exposure and receptivity to digital tobacco content, and for evaluating their causal associations with tobacco use. In the study by Mongilio et al.,1 the authors used latent class analysis to describe patterns in self-reported indicators,5 an approach that identifies data-driven exposure profiles but does not establish a behavioral construct. Rather, exposure was derived from participants’ recall of ENDS-related content seen in the past week, the frequency with which such content was encountered, and the types of content recalled. These responses were then grouped into mutually exclusive exposure profiles, without evidence that the underlying indicators reliably or validly measured the same construct across individuals or over time. In other words, the identified profiles described groups of people who reported seeing similar kinds of content, but they did not provide a reliable or validated way to measure content exposure, nor did they establish whether these distinctions held consistently across individuals, platforms, or contexts. A central challenge facing digital tobacco research is how to accurately measure exposure to, and receptivity toward,6 tobacco-related content encountered in the digital environment. To date, most research has relied on cross-sectional, self-reported, and single-item recall measures of exposure (eg, a survey item asking participants whether they remembered seeing tobacco content online).7,8 Single-item indicators may suffer from self-report bias and lack content validity.9 As such, greater specificity is needed to reflect the full range of platforms, content types (eg, influencer content), and levels of engagement that characterize digital tobacco-related content. Recognition-based approaches (eg, cued recall using representative digital stimuli) may also improve exposure measurement by prompting memory retrieval and enabling assessment of familiarity with specific marketing content rather than relying on unaided recall. Furthermore, the field lacks a measure that captures psychological engagement with digital tobacco content. Exposure alone does not indicate whether individuals attend to, process, or respond favorably to pro-tobacco content that they encounter online. These processes are central to psychological receptivity and the development of pro-tobacco-related attitudes and behaviors.2 These measurement gaps have implications for research, practice, and policy interpretations. While existing regulations and enforcement efforts have successfully targeted overt marketing violations (eg, cartoons, youth-oriented models, or explicit product placement), much of contemporary digital tobacco marketing is subtle and adaptive in nature. As a result, recall-based exposure indicators may be insufficient to guide targeted regulatory action. Regulators require measurement approaches that can reliably distinguish meaningful marketing signals from background noise, identify high-risk content features, and detect shifts in marketing practices over time. Rigorous validation should follow established psychometric best practices, including evaluation of dimensionality, reliability, and construct validity, and, where appropriate, the application of item response theory models to enhance measurement precision and comparability across populations and platforms. The study by Mongilio et al.1 provides a relevant point of departure for advancing measurement in digital tobacco research. Continued investment in rigorous measurement development will be essential for improving surveillance, strengthening causal inference, and informing effective tobacco control policies in the digital environment. In particular, progress will require measures that can capture exposure across platforms, content types, and formats; distinguish exposure from receptivity and other psychological responses to digital tobacco content; and reliably track changes in marketing practices and their associations with pro-tobacco-related attitudes and behaviors over time. Complementary research efforts have attempted to assess digital tobacco exposure using exogenous indicators (eg, aggregated tweet volume),10 which can inform psychometric approaches by capturing broader patterns of tobacco-related activity in the digital environment. Achieving such advances will improve the field’s ability to identify high-risk marketing practices, evaluate the impact of regulatory actions, and design more effective tobacco control interventions in an increasingly complex digital media environment. This publication was funded by the California Department of Public Health through contract #23-10422 Tobacco Industry Monitoring Evaluation (TIME), and in part by grant U01CA278695 from the National Cancer Institute (NCI) of the National Institutes of Health (NIH) and the U.S. Food and Drug Administration (FDA) Center for Tobacco Products. The authors declare no competing interests. Scott Ian Donaldson (Conceptualization [equal], Writing—original draft [equal], Writing—review & editing [equal]), and Jon-Patrick Allem (Conceptualization [equal], Funding acquisition [equal], Writing—original draft [equal], Writing—review & editing [equal]) The findings and conclusions in this article are those of the authors and do not necessarily represent the views or opinions of the California Department of Public Health, California Health and Human Services Agency, NIH, or FDA.