E-Cigs in Early Adolescence Tied to Heavier Smoking Later On

Use of electronic cigarettes during early adolescence appeared to increase the odds of any smoking and more frequent tobacco cigarette use in later adolescence, according to longitudinal data from two large-scale cohorts in the U.K. and U.S.

Among youths who began smoking before age 15, the odds of later adolescent smoking were significantly higher for those who used e-cigarettes relative to those who had not used e-cigarettes, both in the U.K. (adjusted OR 1.45, 95% CI 1.05-2.01) and the U.S. (aOR 2.55, 95% CI 1.59-4.08), reported Jeremy Staff, PhD, of Pennsylvania State University in University Park, and co-authors.

Among U.S. participants, those who used e-cigarettes early in adolescence were also more likely to later be a frequent smoker (aOR 5.11, 95% CI 2.73-9.55), an infrequent smoker (aOR 2.42, 95% CI 1.66-3.51), or someone who exclusively uses e-cigarettes (aOR 3.69, 95% CI 2.39-5.70) compared with non-users, they wrote in Tobacco Control.

Early smoking in e-cigarette users also had higher odds of being a frequent smoker relative to being an infrequent smoker (aOR 2.11, 95% CI 1.03-4.34).

In the U.K., early adolescent e-cigarette users were also more likely to later be a frequent smoker relative to a non-user (aOR 2.01, 95% CI 1.33-3.05), and to be a frequent smoker relative to an infrequent smoker (aOR 1.67, 95% CI 1.06-2.63).

“Youth in the USA and the U.K. show similar long-term declines in tobacco cigarette smoking and recent increases in e-cigarette use, although with stronger shifts towards vaping among U.S. youth,” Staff and team wrote. “Yet, while numerous studies suggest that e-cigarettes in adolescence may be a catalyst for later tobacco use, less well considered is how vaping may shape tobacco consumption among youth who have already initiated smoking tobacco cigarettes early in adolescence.”

In the U.S. cohort, 42% of early e-cigarette users continued smoking in late adolescence, while only 24% of participants who did not use e-cigarettes later smoked. Among U.K. participants, 61% of early e-cigarette users continued to smoke in late adolescence, compared with 50% of smokers who did not use e-cigarettes.

Also of note was that participants from both the U.K. and U.S. were more likely to not use nicotine products at all if they had not used e-cigarettes.

Data from 2022 showed that 3.3% of U.S. middle school students and 14.1% of high school students were currently using e-cigarettes, as were 7% of 11- to 17-year-old students in the U.K. E-cigarettes have only been available on the U.S. market for 16 years, and usage has increased, declined, and then increased again over this time. Staff and colleagues noted that today’s young people have grown up around e-cigarettes, using them at young ages despite health risks.

“Youth potentially face considerable health and economic costs from tobacco smoking and are in need of interventions for e-cigarette use. Youth in these cohorts were the first generation to be exposed as adolescents to e-cigarettes as a new nicotine delivery product (‘guinea pig generation’),” they wrote. “Among youth who started smoking early in adolescence, early e-cigarette adopters were more likely to become entrenched into tobacco use and in heavier smoking than those who smoked but had not used e-cigarettes. Tobacco control efforts aimed at adolescents should incorporate the risks posed by e-cigarettes for early smoking youth.”

For this study, Staff and team used data from the ongoing U.K. Millennium Cohort Study (MCS; n=1,090) and the U.S. Population Assessment of Tobacco and Health (PATH; n=803) study. Some demographics differed between the two cohorts, including a higher male population among PATH participants (54% vs 44%), as well as a much larger white population among MCS participants (90% vs 64%).

Frequent smoking was defined as “usually smoked more than six cigarettes per week” or 27 or more cigarettes in the previous month.

This study does have limitations, Staff and team noted. For example, some study data were self-reported and therefore could be subject to bias, and some early life confounders were not measured in the study. This, paired with the observational design of the study, limited causal inference.