Updated Population-Level Counterfactual Trend Modelling to Examine the Relationship between Smoking Prevalence and E-Cigarette Use among US Adults and Adolescents

Categories: Posters

Poster #135

Author: Foxon F


Background Concerns have arisen that e-cigarettes (ECs) may act as a gateway to combustible cigarettes (CC) among adolescents and may not facilitate switching away from CCs among adults. Population-level trend studies examine whether smoking prevalence consistent with those concerns is observed. Previous US analyses (Foxon & Selya 2020; Foxon et al. 2022) support that ECs facilitate switching and are not a gateway, but analyses may need updating.
Aims To update these analyses, population-level trends are re-examined with the most recently available national data to understand whether the latest estimates provide evidence for/against a gateway among adolescents, and/or suggest ECs are ineffective among adults.
Methods Annual US CC and EC use prevalence were derived from the 1999-2022 National Youth Tobacco Survey (adolescents) and the 1990-2021 National Health Interview Survey (adults). Inflection points were identified algorithmically, and exponential decay functions were fitted to model the counterfactual scenario; what may have happened to CC prevalence if ECs were not introduced at the inflection point. Discrepancies in CC prevalence were calculated by comparing actual survey-estimated prevalence post-inflection to counterfactual projections from pre-inflection trends, after adjusting for the impact of the 2009 Family Smoking Prevention and Tobacco Control Act (FSPTCA). Correlations between CC discrepancies and actual survey-estimated EC prevalence were investigated.
Results Through 2021/2, survey-estimated CC prevalence post-inflection was lower than expected based on pre-inflection trends, even after adjusting for the FSPTCA. CC discrepancies were correlated with EC use such that increasing EC use was associated with decreasing CC use. Discrepancies were greater in cohorts with greater EC use (high schoolers and young adults). These findings were robust to sensitivity tests.
Conclusion Updated analyses support prior findings that observed CC prevalence among adolescents and adults may be lower than in the counterfactual scenario in which ECs were not introduced; data do not support the hypothesis that ECs act as a gateway to CCs among adolescents or are ineffective when used by adults for switching.

Disclosure Through PinneyAssociates, FF consults for Juul Labs Inc. JLI supported this work and reviewed a draft. FF is responsible for the content and decision to publish.