Longitudinal transitions in e-cigarette and cigarette use among US adults: prospective cohort study.

Background: To support tobacco control efforts, this study sought to characterize longitudinal transitions in use of electronic nicotine delivery systems (ENDS) and cigarettes.
Methods: Participants were nationally representative samples of 53,729 US adults from Waves 3-5 (2015-2019) of the Population Assessment of Tobacco and Health Study. We examined behavioral transitions (initiation, relapse, progression, and cessation) in ENDS and cigarette use across waves. Weighted generalized estimating equation models adjusted for sociodemographic variables.
Findings: Of never ENDS users at baseline, an estimated 1.7% reported initiating ENDS use by follow-up. Of former ENDS users, an estimated 12.1% relapsed into ENDS use. Of periodic ENDS users at baseline, 13% progressed to established ENDS use. Of baseline current ENDS users, 46.3% discontinued ENDS use. The corresponding transitions for cigarette smoking were 1.6% (initiation), 4.8% (relapse), 21.1% (progression), and 14% (discontinuation). Adults aged 18-24 (vs. older age), Hispanics (vs. non-Hispanic white), and past 12-month cannabis users were more likely to initiate ENDS or cigarettes (all p  < 0.05). Having any internalizing mental health symptoms increased the odds of ENDS initiation, while externalizing symptoms increased the odds of cigarette initiation. Those who perceived nicotine as very harmful (vs. none/low harm) were more likely to discontinue ENDS. Current cigarette users (vs. non-users) at baseline were more likely to initiate, relapse, or discontinue ENDS (all p  < 0.05) and vice versa.
Interpretation: We observed high changeability in ENDS and cigarette use among US adults over time. In absolute terms, ENDS use grew, while smoking fell. Tobacco control programs should focus on priority populations, including young adults and people with internalizing and externalizing mental health symptoms.
Funding: National Institutes of Health, R01-CA246606-01A1, R01-DA048390.
Competing Interests: We declare no competing interests.
(© 2023 The Author(s).)