UK teens who currently vape are as likely to take up smoking as their peers in the 1970s, despite a substantial fall in the prevalence of teenage smoking over the past 50 years, suggests a long term intergenerational study published online in the journal Tobacco Control.
The likelihood of starting to smoke among teens who don't vape was around 1.5%, but 33% among those who do, the findings indicate.
It's not entirely clear if the rise in popularity of e-cigarettes (vapes) among teens threatens the steady decline in the prevalence of cigarette smoking in this age group, say the researchers, as the published research is somewhat equivocal.
And if historic declines are slowing, it's not clear how the risk of cigarette smoking among today's young people—especially those who vape—compares to that of previous generations who came of age before notable tobacco control legislation and the advent of e-cigarettes, they add.
To try and find out, they drew on intergenerational data from three nationally representative birth cohorts of UK teens* born in 1958 ( National Child Development Study ; NCDS), 1970 ( The British Cohort Study ; BCS), and 2001 ( Millennium Cohort Study ; MCS).
As well as baseline surveys at either birth or 9 months, MCS participants provided data at ages 3, 5, 7, 11, 14, 17 and 23; BCS participants at ages 5, 10, 16, 26, 30, 34, 38, 42, 46 and 51; and NCDS participants at ages 7, 11, 16, 23, 33, 42, 46, 50, 55 and 62.
The prevalence of teen smoking was assessed in 1974 among 11,969 NCDS participants, in 1986 among 6222 BCS participants, and in 2018 among 9733 MCS participants. The odds of smoking among 16 and 17 year olds were estimated, based on a common set of childhood risk and protective factors; teen vaping was included as a predictor in the MCS.
Risk/protective factors included whether they had ever drunk alcohol by age 16 or 17; how engaged they were with education at school; the extent of externalising behaviours reported by the mother or main caregiver at ages 10 or 11; and parental occupation, education, and smoking behaviour— including during pregnancy.
Analysis of the intergenerational data revealed a steep decline in the prevalence of cigarette smoking among teens, falling from 33% in 1974, to 25% in 1986, and to 12% in 2018. Around half of the MCS participants hadn't vaped by the time they were 17; 41% said they had previously vaped; and 11% reported current vaping.
The decline in prevalence of teen smoking can be attributed to a mix of tobacco control legislation, better public understanding of the health consequences of smoking, and a shift away from the perception of smoking as socially acceptable, suggest the researchers.
Risk/protective factors also changed over time. For example, the percentage of teens who had started drinking by the age of 16 or 17 fell from 94% in the NCDS to 83% in the MCS.
The average age at which mothers left education also rose from 15.5 in the NCDS to 17 in the MCS. Similarly, the prevalence of parental smoking fell from over 70% in the NCDS to 27% in the MCS; and fewer mothers continued smoking while pregnant in the MCS than in the NCDS and the BCS.
Many risk factors for teen smoking were similar across the cohorts—drinking before the age of 17 and greater externalising behaviours, for example—as were some child level protective factors—greater engagement with education, for example.
To illustrate the likelihood of cigarette smoking for an 'average' teen (16-17) over time, the researchers worked out predicted probabilities of cigarette smoking with all risk factors included from the intergenerational data.
This probability was 30% in the NCDS and 22% in the BCS. Among those who had never vaped in the MCS this was around 1.5%, but 33% for the teens who reported current vaping.
"This probability is especially concerning given the recent increases in e-cigarette use prevalence among UK youth, despite some initial assurances that e-cigarettes would have little appeal to [them]," say the researchers.
The researchers acknowledge that they were unable to account for some sociodemographic characteristics, including race and ethnicity, due to insufficient sample sizes in the earlier cohorts. And they emphasise that estimates of the associations between vaping and smoking in the MCS aren't causal and shouldn't be interpreted as such, especially as the temporal sequencing of smoking and vaping isn't explicit.
Nevertheless they conclude: "While our research shows that the historic decline in the likelihood of youth cigarette smoking has continued in this recent cohort of UK youth, overall, we find that this is not the case among e-cigarette users.
"Youth who had never used e-cigarettes had an estimated less than 1 in 50 chance of reporting weekly cigarette use at age 17, while those who had previously used e-cigarettes had over a 1 in 10 chance. Youth who reported current e-cigarette use had an almost 1 in 3 chance of also reporting current cigarette use.
"As such, the decline in the likelihood of cigarette smoking is waning for youth who have used e-cigarettes—about half of our sample—and has reversed for those currently using e-cigarettes.
"Among contemporary youth, efforts to reduce cigarette smoking should focus both on those who are currently using e-cigarettes and on the prevention of e-cigarette use among youth, to maintain the promising declines in youth nicotine use in years to come."
