Longitudinal e-Cigarette and Cigarette Use Among US Youth in the PATH Study (2013–2015)
Open Access
- 27 January 2019
- journal article
- research article
- Published by Oxford University Press (OUP) in JNCI: Journal of the National Cancer Institute
- Vol. 111 (10) , 1088-1096
Abstract
Evidence is accumulating that youth who try Electronic Nicotine Delivery Systems (ENDS, e-cigarettes) may go on to try cigarettes. This analysis examines the bidirectional patterns of ENDS and cigarette use among US youth over one year and uses propensity score matching (PSM) to examine frequency of ENDS use on changes in cigarette smoking. Our analysis included 11 996 participants who had two waves of available data (Wave 1 [W1] 2013–2014; Wave 2 [W2] 2014–2015) drawn from the longitudinal Population Assessment of Tobacco and Health Study. Cross-sectional weighted prevalence estimates are reported for cigarettes and ENDS. We used PSM to estimate the likelihood of ENDS use at W1 and to draw matched analytic samples, then used regression (logistic or linear) models to examine the effect of W1 ENDS use on W2 cigarette smoking. All statistical tests were two-sided. In weighted analyses, 69.3% of W1 past-30-day cigarette smokers exhibited past-30-day smoking at W2; 42.2% of W1 past-30-day ENDS users were using ENDS at W2. W1 ever use of either product was similarly associated with W2 new use of the other product. Unweighted PSM models indicated W1 cigarette-naïve ENDS use was associated with W2 ever-cigarette smoking (n = 676; adjusted odds ratio = 3.21, 95% confidence interval [CI] = 1.95 to 5.45, P < .001); W1 ever-ENDS use did not affect change in cigarette frequency at W2 (n = 1020, beta = 0.31, 95% CI = −0.76 to 1.39, P = .57); 1–5 days ENDS use compared with ever, no past-30-day ENDS use was associated with a statistically significant decrease of W2 smoking days (n = 256, beta = −2.64, 95% CI = −4.96 to −0.32; P = .03); and W1 6+ day ENDS users did not show a decrease in frequency of cigarette smoking. Ever-ENDS use predicts future cigarette smoking, and frequency of ENDS use has a differential impact on subsequent cigarette smoking uptake or reduction. These results suggest that both cigarettes and ENDS should be targeted in early tobacco prevention efforts with youth.Keywords
Funding Information
- National Institute on Drug Abuse
- National Institutes of Health
- Center for Tobacco Products (HHSN271201100027C)
- Food and Drug Administration
- Department of Health and Human Services
This publication has 30 references indexed in Scilit:
- Association of Electronic Cigarette Use With Initiation of Combustible Tobacco Product Smoking in Early AdolescenceJAMA, 2015
- E-Cigarette Uptake Amongst UK Youth: Experimentation, but Little or No Regular Use in NonsmokersNicotine & Tobacco Research, 2015
- How to define e-cigarette prevalence? Finding clues in the use frequency distributionTobacco Control, 2015
- Electronic cigarettes and nicotine dependence: evolving products, evolving problemsBMC Medicine, 2015
- Propensity Score Analysis with Survey Weighted DataJournal of Causal Inference, 2015
- A comparison of adolescent smoking initiation measures on predicting future smoking behaviorPreventive Medicine Reports, 2015
- Does one cigarette make an adolescent smoker, and is it influenced by age and age of smoking initiation? Evidence of association from the U.S. Youth Risk Behavior Surveillance System (2011)Preventive Medicine, 2014
- Generalizing Observational Study Results: Applying Propensity Score Methods to Complex SurveysHealth Services Research, 2013
- Matching as Nonparametric Preprocessing for Reducing Model Dependence in Parametric Causal InferencePolitical Analysis, 2007
- The central role of the propensity score in observational studies for causal effectsBiometrika, 1983