Abstract
To examine and compare trends in past 30-day cigarette smoking among adolescents in the US from 2002 to 2019, before and after the onset of the "e-cigarette era" in 2014. Using National Youth Tobacco Survey (NYTS) data from 2002 to 2019, we modeled the prevalence of past 30-day cigarette smoking over time. A total of n = 274,551 (weighted N = 340,403,754) middle and high school students were included in this study. Interrupted time series analyses were used to examine changes in cigarette smoking over time and compare trends in cigarette smoking pre- and post-2014. Models were applied to the full sample and stratified by middle (6th-8th grade) and high school (9th-12th grade). The observed number of current adolescent cigarette smokers post-2014 was compared to the predicted number, given trends in cigarette smoking prevalence observed pre-2014. Among all students, past 30-day cigarette smoking declined by approximately 0.75% per year from 2002 to 2013 (p < 0.001). Following a significant drop in prevalence from 2013 to 2014 (1.64%; p < 0.001), the decline in past 30-day cigarette smoking slowed significantly to approximately 0.37% per year (p < 0.001), from 2015 to 2019. We estimate that the onset of the "e-cigarette era" in 2014 corresponded to over 1.66 million (95% CI: 1.57 m - 1.75 m) more past 30-day cigarette smokers from 2015 to 2019, cumulatively. The rate of decline in past 30-day cigarette smoking prevalence among adolescents observed since 2002 slowed with the onset of the "e-cigarette era" in 2014, providing evidence at a population-level for the "gateway effect."