Are e-cigarettes a gateway to tobacco smoking in youth?

Arguments about electronic cigarettes (e-cigarettes) and what do about them rumble on. I won’t wade into the broader debate, but I direct you to a recent viewpoint article (McKee and Capewell, 2015), and accompanying commentaries (particularly McNeill et al., 2015).

The main arguments involve the safety of e-cigarettes (relative to smoking regular cigarettes and relative to, well, breathing air), their effect on renormalisation of tobacco smoking when its prevalence is declining, their role as smoking cessation aids, and their use by young people.

The present article is related to the latter issue, specifically the worrying possibility that use of e-cigarettes by young people may make them more likely to progress to smoking regular cigarettes in the future.

Last year I wrote about a cross-sectional study which showed that American youth who used e-cigarettes were more likely to smoke regular cigarettes and less likely to quit smoking than others who did not use e-cigarettes (Field, 2014). This study caused quite a stink at the time, largely because it was interpreted as evidence for the claim that e-cigarettes may be a ‘gateway’ into tobacco smoking for young people, although various commentators were quick to point out that the findings didn’t imply that at all. A subsequent longitudinal study (Leventhal et al., 2015) demonstrated that use of e-cigarettes among American high school students was associated with subsequent progression to use of regular cigarettes, although the authors of that study emphasised that this does not mean that e-cigarettes play a causal role.

The present study (Primack et al., 2015) investigated a related issue using a similar research design among slightly older participants.

Currently there is no evidence showing a causal link between e-cigarette use and subsequent progression to use of regular cigarettes.

Methods

This is a longitudinal cohort study in which 694 non-smoking participants aged between 16 and 26 (mean age = 20 years; 54% female) provided information about their use of e-cigarettes and regular cigarettes, and their ‘susceptibility to smoking’, via a telephone survey.

‘Susceptibility to smoking’ was determined based on responses to questions about whether they might smoke a (regular) cigarette in the next year.

Participants provided this information at a baseline assessment that took place between October 2012 and May 2014, and then provided it again at follow-up, one year later.

Results

At the baseline assessment, 16 participants (2.3%) had ever used e-cigarettes, and were thereafter referred to as ‘E-Cig users’. Of these, six people (37.5%) progressed to being a ‘cigarette smoking initiator’ (defined as taking at least one puff on a regular cigarette in their lifetime) at follow-up, as opposed to only 65 individuals (9.6%) who were not E-Cig ‘users’ at baseline. An additional five individuals (31.3%) who were e-cigarette users at baseline were classified as ‘increasingly susceptible to smoking’ at follow-up, as opposed to 63 individuals (9.3%) who were not e-cigarette users at baseline.

The predictive validity of e-cigarette use at baseline for regular cigarette smoking and susceptibility to smoking at follow-up were confirmed with a series of multivariable multinomial logistic regression analyses that used different methods to deal with missing data and inconsistent responding from participants.

In these analyses, progression along the cigarette smoking trajectory from baseline to follow-up was the dependent variable, and predictor variables were e-cigarette use at baseline alongside age, gender, race / ethnicity, maternal education, sensation-seeking, and cigarette smoking in parents and friends. These analyses confirmed that:

Individuals who smoked e-cigarettes at baseline had larger point estimates of progressing to ‘susceptible to smoking’ at follow-up, compared to those who did not use e-cigarettes at baseline (Adjusted Odds Ratio = 8.5, 95% CI = 1.3 to 57.2)
Similarly, individuals who smoked e-cigarettes at baseline had larger point estimates of progressing to be a ‘cigarette smoking initiator’ at follow-up, compared to those who did not use e-cigarettes at baseline (Adjusted Odds Ratio = 8.3, 95% CI = 1.2 to 58.6).
Younger age, sensation-seeking tendencies, and having friends who smoke were also associated with progression along the cigarette smoking trajectory.

This study of 694 non-smoking 16-26 year olds, used telephone surveys to gather information about e-cigarette and regular cigarette use and intention to use.

Discussion

The headline finding is certainly alarming: among young non-smokers, use of E-cigarettes was associated with more than an 8-fold increase in the likelihood of progressing to smoking regular cigarettes one year later. This finding is consistent with findings reported in another recent study (Leventhal et al., 2015) and both sets of results go beyond findings from a cross-sectional study that I wrote about last year, because they establish temporal precedence (some people are using E-Cigs before regular cigarettes) and because they control for sensation-seeking, a variable that is likely to influence use of both E-Cigs and regular cigarettes.

Primack et al’s discussion of their findings is appropriately cautious, and they make the valid observation that attempts to market E-Cigs to youth may be cynically motivated by the desire to get the next generation addicted to nicotine and ensure their progression to smoking regular cigarettes.

However, an accompanying editorial in the same issue of the journal (Klein, 2015) threw caution to the wind and concluded that:

these products (e-cigarettes) harm nonsmokers and result in a net harm to society and public health.

And:

At a time when many claim to be uncertain about harms and benefits of e-cigarettes and argue for more studies, these data provide strong longitudinal evidence that e-cigarette use leads to smoking, most likely owing to nicotine addiction. We no not need more research on this question; we have the evidence base…

The interpretation of this data by some commentators has been misplaced and alarmist.

Limitations

Klein’s comments are ridiculous. While this study makes an important contribution, it does not provide strong longitudinal evidence that e-cigarette use causes smoking. These are some of the reasons why:

The findings are effectively based on only 16 individuals, that is the 2.3% of the sample who had ever used an e-cigarette at the baseline assessment. It is true that the incidence of e-cigarette use has dramatically increased in the time since this baseline assessment (October 2012 to May 2014). Nonetheless, it requires an enormous leap of faith to conclude that non-smokers who try e-cigarettes will be more than 8 times more likely to progress to regular cigarettes in the future compared to people who do not try e-cigarette, based on only 16 people. More research is needed here, and if the incidence of E-Cig use among youth who do not currently smoke regular cigarettes is as high as claimed by some (see another of our blog posts), it should not be difficult to investigate this research question.
As acknowledged by the authors, the primary outcome measure was simply taking at least one puff of a regular cigarette. Although there is a well-defined developmental trajectory of smoking (with changes in attitudes preceding experimentation which precedes regular smoking), this study does not demonstrate that use of E-Cigs leads to regular cigarette smoking.
The study does not distinguish between regular use of e-cigarettes versus experimenting with them.
It is worth pointing out that the average age of participants in this study was 20 years old, and the age range was 16-26. Note that much of the debate on this topic is centred on the protection of children and adolescents, so we should be cautious before generalizing findings from 20 year olds to children. Note, however, that the longitudinal cohort study reported by Leventhal et al. (2015) tested high school students with an average age of 14, and reached similar conclusions to the present study. However, neither of these studies demonstrate that the use of e-cigarettes plays a causal role in progression to use of regular cigarettes.