Electronic cigarettes (or e-cigarettes) are electronic devices that can deliver nicotine in aerosol form without burning tobacco. Their use is increasing worldwide [1
] and is more common among current smokers and former smokers than never-smokers [5
]. The prevalence of e-cigarette use differs between countries, according to national policies. In France, 26% of 15–75 years old had tried e-cigarettes in 2014; 3% were daily users [9
]. Their promotion and sale are partially regulated in France. The liquid that is vaporized is considered neither to be a tobacco product nor a medicinal drug but as a product for regular consumption that can be sold only in specialist shops. However, its sale is forbidden to people under the age of 18. Advertising for electronic devices is prohibited as is their use in certain places like schools and closed forms of public transport.
In Europe, smoking prevalence in 2014 was higher among adults aged 25–39 years (33%) or 40–54 years (34%) than in subjects aged 15 to 24 years (25%) or 55 and over (17%) [10
]. Smoking prevalence in European young adults was higher than in the U.S. population for the same age [11
]. Nicotine replacement therapy and electronic cigarettes were the most common tools used to stop tobacco without assistance by European people [10
]. Indeed, the most frequently reported reason for e-cigarette use in previous studies was smoking reduction or cessation even though its safety and efficacy as a support to smoking cessation are still under study [2
]. Furthermore, it is mainly adolescents and young adults who use them [7
]. While several studies have explored the beliefs and perceptions of young populations about e-cigarettes [16
] or their awareness related to their harmfulness [17
], few have focused specially on the reasons for using them. In a survey of adults in the USA in 2013, the reasons for experimenting with e-cigarettes differed with age [19
]. Among people 18–34 years old, the most common reason was curiosity. The frequency of this reason decreased with age, the most frequent reason for trying in those aged 55 being the desire to reduce or stop smoking.
Since tobacco and e-cigarette use levels were both high in young adults, we wondered whether they tried e-cigarettes for reasons related or not with smoking (cessation or decrease). We also wanted to establish the characteristics of young smokers who use e-cigarettes, in terms of tobacco addiction and motivation to quit smoking. We, therefore, sought to find out which young adults had tried e-cigarettes and why. The main objective of this study was to describe the relationship between electronic cigarette use and tobacco-smoking in a large sample of French students. Another objective was to describe the reasons for experimenting with e-cigarettes in these young adults. As with several previous studies which were conducted in student populations [20
], we chose to recruit volunteers to participate in an online survey. This allowed us to quickly collect data among the target population.
In this study, two out of five students (40%) had tried an e-cigarette at least once, but only 9% remained current users. There was a statistically significant relationship between e-cigarette use and tobacco-smoking; experimentation as well as current or daily use of e-cigarette were more common in former smokers than in current smokers. Among never smokers, 13% had tried an e-cigarette. The three main reasons for trying e-cigarettes in decreasing order were curiosity, someone offering one, and the attraction of new flavors. In smokers, previous attempts to quit smoking and a strong desire to stop tobacco were reported more among current e-cigarette users than former users. There was no statistically significant association between e-cigarette use and nicotine dependence.
Most students who had tried an e-cigarette finally gave it up, whatever their smoking status. Almost nine in ten students were smokers or former smokers at least one year before trying an e-cigarette. In several previous studies in young people, e-cigarette experimentation as well as current use were more frequent in current tobacco smokers (called dual users) than in former smokers [17
]. This was not the case in our study as former smokers were more likely current e-cigarette users than current smokers.
Thirteen percent of never-smokers had tried an e-cigarette but 97% of them had discontinued it at the time of the survey. A few of them continued to use it non-daily, alternating e-liquids with or without nicotine. Thus, a minority of never-smokers had an e-cigarette use which might have durably exposed them to nicotine. Although this was a minor phenomenon in our study, this exposure to nicotine by e-cigarette use could be a gateway into cigarette smoking. A recent meta-analysis investigated the initiation of tobacco smoking depending on the experimentation with e-cigarettes among adolescents and young adults [32
]. The tobacco smoking initiation was more common among ever e-cigarette users than those who had never tried it.
The most common reason for trying an e-cigarette was curiosity. Findings suggested that never-smokers tried an e-cigarette mainly out of curiosity and then gave up it. To our knowledge, no study to date has specifically focused on motives of e-cigarette use in young adults who never smoked. However, our findings are similar to those found in never-smoker adolescents. For example, 2 to 5% of non-smoker adolescents in the United Kingdom had tried an e-cigarette in 2014 but fewer than 1% used it monthly [33
]. Some studies in adolescents have shown that the intention to smoke cigarettes later was higher in non-smokers who had tried an e-cigarette than in non-smokers who had never tried one [34
]. Longitudinal or qualitative research studies also are needed in young adults who do not smoke to understand why they use e-cigarettes and how this may accelerate the onset of tobacco-smoking.
We also found that previous attempts to quit smoking among current smokers were reported more by current users of e-cigarettes than by former or never e-cigarette users. They also more frequently reported being in the process of stopping smoking and having a stronger motivation for doing so. Thus, some of the smokers in our population might use e-cigarettes to change their tobacco consumption, which would be consistent with previous observations [14
]. In absence of demonstrated efficiency of e-cigarettes in smoking cessation, caregivers must continue to propose to young adults some drugs indicated as a cessation aid: transdermal patches of nicotine, varenicline, and bupropion in France [38
]. These treatments may be completed, where appropriate, with the use of e-cigarettes.
This cross-sectional analysis included a large number of participants. To our knowledge, this is the first study in French-speaking students who smoke to analyze the relationship between e-cigarette use and nicotine dependence or previous attempts to quit smoking. It is also the first study exploring the presence of nicotine in e-liquids among young French e-cigarette users.
This study had several limitations. First, the voluntary participation of students and the moderate response rate may have introduced a selection bias. Due to this voluntary participation, the constitution of our sample was quicker and easier. Nevertheless, as shown in the Supplementary Materials (Table S1)
, participants in the survey on e-cigarettes were more often female than male and were younger than those who did not participate. They went to Bordeaux campuses and trained in healthcare related studies more than non-participants. Thus, a selection bias was plausible. It was not also excluded that former smokers might have participated more than current or never-smokers in our study. All these might overestimate the prevalence of e-cigarette use in former smokers and underestimate it among never-smokers. Second, data available on the prevalence of tobacco-smoking among French-speaking young adults came from the Health Barometer 2010 survey [39
]. This cross-sectional phone survey conducted among a representative sample of the French population including 6400 young people aged 15 to 30 years of age, 2276 of whom were students. This survey found that the prevalence of current smoking was 44% in 15–30-year-olds. However, that sample was not comparable to ours. Since we lacked recent data, it was impossible to know whether our proportions were close to reality. With a 48% prevalence of current smoking in our study, we cannot rule out an over-representation of smokers by self-selection bias. Representativeness and self-selection bias could have been partially managed by empirical or randomly selected sampling before the beginning of our study and by weighting during the data analysis. These management techniques of selection bias could not be applied because data about the characteristics of the target student population were not available. Third, the ratio of women to men in our sample was 4 to 1. This proportion of women was very high and remains incompletely explained. In France, women represented 57% of students enrolled in a university for the academic year 2015–2016 (Paris campuses: 58%; Bordeaux campuses: 56%, Nice campus: 55%, Versailles campuses: 52%) [40
]. More than 75% of students included were French-speaking students enrolled in these four French universities. In 2016, women were overrepresented among the French undergraduate students in the most common majors in our study: Literature, arts, humanities and social (73.1%), healthcare (71.7%) [41
]. Thus, our prevalence of female students could be explained by the female predominance in the university majors of the three main universities of i-Share (Bordeaux, Versailles, and Nice). It could also be due to a selection bias, with female students participating more often than male students in clinical research studies. In other recent studies in French student populations, there also was a female predominance [42
] but in a lower rate. Fourth, analyses performed in the smokers' subgroup concerned only the 585 smokers who answered correctly about their smoking habits on the supplementary questionnaire. As shown in the Supplementary Materials (Table S3)
, these 585 smokers reported more smoking cessation being underway than the other smokers. Fifthly, some classification biases were plausible. All data were self-reported. It may be difficult to distinguish former and current smokers in a young population since students who smoke little or irregularly, as well as those who have stopped very recently, may find it difficult to define their smoking status. Similarly, the definition of former e-cigarette users in this cross-sectional analysis did not make it possible to distinguish those who stopped e-cigarette and tobacco consumption from those who stopped e-cigarette use but resumed tobacco-smoking. We collected age at start of tobacco smoking and of e-cigarette use. These questions were not mandatory and we had this information for only 550 students (59.3%) among 928 smokers/former smokers who had tried an e-cigarette at least once. It was impossible to specify the sequence for 40 students who started e-cigarette and tobacco the same year. Finally, former smokers reported more than current smokers that they had tried e-cigarettes to stop or decrease tobacco smoking in our study. However, a descriptive cross-sectional study did not allow for an evaluation of the direction of the association between e-cigarette use and smoking status. Former smokers could use the e-cigarette more than current smokers because they quit smoking after using e-cigarettes. They could also have quit smoking and then restarted using nicotine via the e-cigarettes to avoid smoking again. Therefore, caution is required when extrapolating our findings to other young populations.