Associations between e-cigarette access and smoking and drinking behaviours in teenagers

Rapid growth in the marketing and use of e-cigarettes is generating widespread public health debate across the globe. Designed to provide a comparable smoking experience to conventional cigarettes, e-cigarettes are battery-powered nicotine delivery devices that provide doses of nicotine through an aerosol, typically combined with flavouring and driven by a component such as propylene glycol [1,2]. Unlike many nicotine replacement therapies, e-cigarettes are often marketed to smokers as healthier alternatives to conventional tobacco products, rather than as aids to ending nicotine dependence [3]. Proponents stress their harm reduction benefits in moving smokers away from the damaging toxins in conventional tobacco products, and their potential in helping them quit smoking altogether [4,5]. However, opponents highlight their unknown quality, safety and efficacy; their deleterious impact on the ‘no smoking’ health message; and their potential to harm children, including as a possible gateway to cigarette smoking [2,4,6]. Thus, despite growing consensus among health professionals that e-cigarettes are a less damaging delivery mechanism for nicotine than conventional tobacco products [6], a lack of regulation governing them in many countries means that an addictive drug can be promoted to, and cheaply accessed by, children [7]. However, the extent of children’s access to e-cigarettes and the characteristics of those that access them remain poorly defined.

Research on e-cigarette use in children is only just emerging, yet studies in several countries are showing their rapid penetration into adolescent markets. In the USA, studies show that lifetime e-cigarette use in grades 6-12 school students (~age 11-18) more than doubled from 3.3% in 2011 to 6.8% in 2012 [1]. Whilst most e-cigarette users in the 2012 survey were current smokers, 9.3% had never smoked a conventional cigarette. In France, a 2012 study of Parisian school students aged 12-19 years found that 8.1% had tried e-cigarettes, with prevalence ranging from 4.4% of non-smokers up to 33.4% of regular smokers [8]. In Korea, prevalence of ever using e-cigarettes in 13-18 year olds was found to have increased almost 20-fold between 2008 and 2011, from 0.5% to 9.4%, with 1.4% of e-cigarette users in 2011 having never smoked cigarettes [9].

In the UK, the marketing and sale of e-cigarettes has increased substantially over recent years [3,10], with 1.3 million people estimated to use them in 2013 [11]. A survey in 2013 found that two thirds of 11-18 year olds had heard of e-cigarettes, and that 5% of those who had heard of them had tried them [12]. E-cigarettes are expected to be regulated in the UK through European legislation by 2016 and a ban on sales to under 18s is expected to precede this in England [13,14]. At present, however, they are governed only through general product safety regulations with no specific sale or marketing restrictions. Trading Standards is the body responsible for enforcing regulations on the sale of consumer goods in the UK, including sales of age-restricted products. To inform their work in North West England, Trading Standards run biennial school surveys on alcohol and tobacco-related behaviours among 14-17 year old students, with the legal purchase age for both these substances being 18 years. With growing concerns around the exposure of children to unregulated e-cigarettes, a question was added to the 2013 survey asking students if they had ever bought or tried e-cigarettes. This question offers the opportunity to identify levels of e-cigarettes access among children under 18 years and the characteristics of those that are most likely to be accessing them; with preventing child access to e-cigarettes being a key focus of legislative approaches to e-cigarettes not only in the UK but internationally. Thus, here we explore associations between e-cigarette access, student demographics, and their patterns of tobacco and alcohol use.

The study used data from the 5^th iteration of the Trading Standards North West Alcohol and Tobacco Survey conducted among 14-17 year olds in schools in North West England. The survey consists of closed, self-completed questions including: age, gender, alcohol consumption (drinking frequency, binge drinking frequency, drink types consumed, drinking location, drinking to get drunk) smoking behaviours (smoking status, age of first smoking), alcohol and tobacco access methods, parental smoking, and involvement in violence when drunk [15-17]. The question on e-cigarette access asked students “have you ever tried or purchased e-cigarettes”.

As with previous survey rounds, the questionnaire was made available to secondary schools across North West England through local authority Trading Standards departments [15-17]. Within schools, the questionnaire was delivered to students by teachers during normal school lessons between January and April 2013. Students self-completed the questionnaire voluntarily and anonymously. Compliance was not recorded by schools with the sample not intended to be representative of North West school children, but rather to provide a broad sample of students from a range of community types. A total of 114 schools participated, with representation from schools in all 23 upper tier local authorities in the region. The initial survey in 2005 was approved by the Trading Standards North West Executive Committee and schools participate on a voluntary basis under their own approval systems. Ethical approval to analyse the survey data for this study was obtained from Liverpool John Moores University Research Ethics Committee.

The total dataset included 18,233 participants aged 14-17 years. Cases were excluded if they were missing data on gender (n = 144), e-cigarette access (n = 602), or if questionnaires were spoiled (n = 23; e.g. unrealistic answers). The Index of Multiple Deprivation (IMD) 2010 [18] was used to assign participants to an ecological quintile of deprivation using a three stage process. Where full postcode was available (n = 8,676 cases), these were mapped directly to a Lower Super Output Area (English geographical areas with an average population size of 1,500 [19]) for which an IMD score is routinely calculated at a national level. Where only partial postcode was provided, those that spanned more than one LSOA were assigned to the LSOA that contained the majority of postcodes possible within the partial postcode (n = 2,926 cases). Where no postcode information was available, students were allocated to an LSOA based on school postcode; a proxy measure used in previous studies (n = 5,844 cases) [15-17]. Students were assigned to a national quintile of deprivation based on the IMD score of their allocated LSOA (1 = most affluent, 5 = most deprived). Students for whom no marker of deprivation was available were excluded from analyses (n = 1,253 cases). Thus, the final sample was 16,193.

The demographic breakdown of the sample is shown in Table 1. One in five respondents (19.2%) reported having accessed e-cigarettes, with access being higher in males than in females and increasing with age and deprivation (Table 1). Levels of e-cigarette access increased from 4.9% of those who had never smoked cigarettes, to half (50.7%) of ex-smokers and over two thirds of regular smokers (67.2% light regular smokers, 75.8% heavy regular smokers, P < 0.001; Table 1). Thus, across all students who had accessed e-cigarettes, 35.8% were regular smokers, 11.6% only smoked when drinking, 13.6% were ex-smokers, 23.3% had tried smoking but didn’t like it, and 15.8% had never smoked. Teenagers with a parent/guardian who smoked were more likely to have accessed e-cigarettes than those with non-smoking parents/guardians (Table 1). E-cigarette access also showed a strong relationship with alcohol use. Students who drank alcohol at any level were significantly more likely to have accessed e-cigarettes than non-drinkers (23.7% v 9.3%, P < 0.001), with prevalence rising from less than one in ten in non-drinkers or occasional moderate drinkers to 41.9% in regular drinkers that binge (Table 1).

The relationship between e-cigarette access and male gender remained after adjusting for confounders through logistic regression (Table 1). However, relationships between e-cigarettes and age and deprivation were no longer significant. Any smoking experience was associated with e-cigarette access, with adjusted odds ratios (AORs) rising to 50.28 in regular heavy smokers. An independent relationship also remained between e-cigarette access and having parents/guardians that smoke. Compared with non-drinkers, binge drinkers had increased odds of e-cigarette access, with AORs rising from 1.46 for occasional drinkers to 1.89 for those drinking weekly or more.

A total of 6,323 drinkers provided details on the drink types they consumed in a normal week. Spirits were the most common type of drink consumed (63.1%), followed by lager (52.7%), alcopops (51.6%), bottles/cans of cider (49.4%), wine (42.9%) and large value bottles of cider (31.5%). A logistic regression model using all variables in Table 4 plus drink type variables found that consumption of spirits (AOR 1.32 [1.10-1.60], P = 0.004) was independently associated with e-cigarette access.

As public health debate on the pros and cons of e-cigarettes continues, moves to regulate their promotion and sale are increasingly being proposed and adopted by governments across the world [6]. A key feature of such regulation is preventing child access to e-cigarettes. Thus, understanding the extent of e-cigarette access by young people and the characteristics of those that access them will be crucial in informing prevention and control strategies. Using a sample of over 16,000 14-17 year old school students in North West England, we found that almost one in five had either tried or purchased e-cigarettes. Such rapid penetration into teenage culture of what is essentially a new drug use option is without precedent. As with findings from studies elsewhere [7-9], e-cigarette access was most common in students who smoked conventional cigarettes, particularly those who smoked in greater quantities. Thus, 67.2% of light smokers (<5 per day) and 75.8% of heavier smokers reported having accessed e-cigarettes. These figures are far higher than those for e-cigarette use by adolescent smokers reported in previous studies (e.g. France, 33.4% [8], Korea, 36.6% [9]), and this will in part reflect the different age ranges and broader question asked in our study. However, findings are consistent with other recent studies reporting high levels of e-cigarette access among tobacco smokers (e.g. USA [20]), and this may indicate the rapid expansion in promotion [10] and reducing price [21] of e-cigarettes that means they are widely visible and easily available to teenagers with an interest in smoking.

A key public health concern with e-cigarettes is their potential to recruit children to nicotine dependence. In adults, e-cigarettes are typically used by smokers to help them reduce or quit tobacco use and uptake levels among non-smokers are thought to be very low (typically <1%) [22-25]. Here, however, almost one in twenty (4.9%) teenagers who had never smoked conventional cigarettes reported having accessed e-cigarettes. In fact, although ex-smokers had greater odds of e-cigarette access than never smokers, never smokers accounted for a much larger proportion of the sample (61.2% v 5.1% for ex-smokers) and therefore a larger proportion of those reporting e-cigarette access (15.8%) than ex-smokers (13.6%). Among never-smokers, odds of having accessed e-cigarettes were greater in 14 year olds than in those aged 15 or 16 years. This likely reflects the fact that many young people who are inclined towards smoking will have already tried cigarettes by the age of 15. For individuals who have never used a substance, however, accessing that substance is a first step to initiating use. Further, for conventional tobacco products, perceived ease of accessibility has been found to increase adolescents’ risk of smoking uptake [26].

In the UK, as in many other countries, today’s adolescents have grown up with a strong ‘smoking kills’ message and reducing social acceptance of smoking. Smoking prevention education is delivered in schools from an early age, tobacco advertising is banned, tobacco packaging carries strong visual health warnings and smoke-free legislation prohibits smoking in virtually all enclosed work and public spaces. As tobacco control and smoking prevention activity has increased, smoking prevalence in children has reduced, with 23% of 11-15 year olds in England in 2012 having ever tried smoking compared with 49% in 1996 [27]. However, adolescence is a period of experimentation and while anti-smoking efforts may have deterred many teenagers from trying tobacco, the marketing of e-cigarettes as a healthy alternative may proffer a viable new method for them to experience the nicotine ‘hit’ without the perceived harmful impacts of tobacco. For other products, health claims have been found to produce a ‘halo effect’ that not only encourages purchasing, but also reduces consumers’ information-seeking (e.g. food and nutritional information [28]). Specifically for teenagers, glamorisation of e-cigarettes in advertising, celebrity endorsement and the range of attractive designs and flavourings available are likely to be furthering their appeal [3]. Here, almost a quarter of teenagers that had accessed e-cigarettes had tried smoking conventional cigarettes but not liked them. Although we cannot determine whether this experience occurred before or after accessing e-cigarettes, it is likely that flavourings make e-cigarettes an attractive option to teenagers who would otherwise be put off conventional cigarettes by their taste.

A key finding from our study was the association between alcohol consumption and e-cigarette access. Even after controlling for smoking behaviours, teenagers who drink regularly and binge drink were significantly more likely to have accessed e-cigarettes (Table 1). This association was particularly strong in those that had never smoked. Thus, over one in ten never-smokers who drank regularly and binged had accessed e-cigarettes, with their odds of e-cigarette access more than four times those of never-smokers who did not drink. Among all drinkers, e-cigarette access was associated with binge drinking, drinking to get drunk, involvement in violence after drinking and consumption of spirits; a drink type that has been associated with alcohol-related harm in previous studies [16]. These findings suggest that teenagers that access e-cigarettes are those that are most vulnerable to other forms of substance use and risk-taking behaviours, and that e-cigarettes are rapidly being added to at-risk teenagers’ substance using repertoires.

The high prevalence of e-cigarette access amongst teenagers in our study, and particularly their use among those that have never smoked conventional cigarettes, highlights the urgent need for age restrictions on the promotion and sale of e-cigarettes. With such restrictions increasingly being introduced, understanding how teenagers that access e-cigarettes are able to access other age-restricted products (i.e. cigarettes and alcohol) will support enforcement work. Among smokers, obtaining cigarettes from friends or family over the age of 18 was independently associated with e-cigarette access. Friends and family are also key sources of e-cigarettes for teenagers [29]. Among drinkers, teenagers who bought alcohol themselves from off-licensed premises and those who asked adults outside shops to purchase alcohol for them (known as proxy purchasing) also had increased odds of e-cigarette access. Both these methods represent mechanisms by which minors access alcohol illicitly without the knowledge or supervision of parents or guardians, and both have been associated with risky alcohol consumption and alcohol-related harms in previous studies [15,16]. Their association with e-cigarettes suggests that teenagers who access e-cigarettes are already familiar with strategies to bypass age legislation on restricted products.

Across all analyses conducted in our study, having a parent or guardian who smokes was one of the strongest predictors of e-cigarette access. Associations between parental and child smoking are widely reported [30], and are thought to operate through both genetic and environmental influences. Similar relationships may exist for propensity to experiment with e-cigarettes. The Trading Standards survey did not ask students whether or not their parents used e-cigarettes, yet the availability of e-cigarettes in the home may be an important consideration for future interventions. Although the content of e-cigarettes varies [31], single cartridges typically contain several hundred ‘puffs’ and unguarded, could easily be ‘shared’ by children without the adult users’ knowledge. The involvement of parents along with schools in work to address e-cigarette use in children is likely to be particularly important as their lack of smoke and odour means that, unlike conventional cigarettes, they can easily be used in bedrooms or on school property without detection.

Like all cross-sectional surveys, this study had a number of limitations. Firstly the question on e-cigarettes included in the Trading Standards survey was specific to access, asking whether students had tried or bought e-cigarettes. However, preventing child access to e-cigarettes is the focus of current regulatory responses to e-cigarettes. While it is not possible to determine how many of those teenagers who accessed e-cigarettes had either bought or used them, it seems reasonable to assume that many teenagers who are motivated to purchase an e-cigarette would also be interested in trying it. The survey did not record any measure of e-cigarette access frequency nor of when e-cigarette access had occurred, and thus it was not possible to identify whether teenagers that reported both e-cigarette access and smoking had accessed e-cigarettes before or after conventional cigarettes. While school surveys can be limited by time availability in the classroom, results from this survey justify a greater focus on e-cigarettes in the next iteration of the survey, including questions on frequency of e-cigarette use and age of initiation. School participation was voluntary and compliance data were not recorded, thus selection bias cannot be ruled out. In the absence of residential postcode data, students were assigned to a quintile of deprivation on an ecological basis and while this may have meant some students were misallocated, the deprivation profile of the sample was generally consistent with that for the 14-17 year old population in the North West region (sample, quintile 1, 12.5%; 2, 15.1%; 3, 15.8%; 4, 18.7%; 5, 37.9%; North West, quintile 1, 15.6%; 2, 16.3%; 3, 15.6%; 4, 18.0%; 5 34.5%). Students that could not be assigned to a deprivation quintile were excluded from analysis and therefore represent additional potential bias in the final sample. As with all surveys of self-reported social behaviours, students’ may have under or over reported e-cigarette access, smoking and drinking behaviours due to factors including social desirability, poor recall or lack of knowledge. Finally, while little data are available on the regional distribution of e-cigarette use among either children or adults in England, prevalence of tobacco smoking tends to be higher in the North than in the South [32]. Thus findings should not be considered representative of all 14-17 year olds in England or the North West region.

Identifying which children are accessing e-cigarettes is crucial for targeting health information and understanding which young people may be most vulnerable as sales of e-cigarettes become age restricted. Our study suggests that a substantial number of teenagers are accessing e-cigarettes, including those who have never smoked conventional tobacco products. However, those most likely to access e-cigarettes are those who engage in other substance-related risk behaviours including regular smoking, binge drinking, drinking to get drunk and alcohol-related violence. Thus, findings appear more consistent with teenagers viewing e-cigarettes as a recreational substance rather than a smoking cessation tool. E-cigarette access is also associated with specific alcohol access patterns, including self-purchase of alcohol in off-licensed premises and the recruitment of proxy purchasers from outside such premises. Thus, high risk teenagers that access e-cigarettes are likely to already be familiar with mechanisms for avoiding age restrictions on substances. In particular, findings highlight the urgent need for controls on e-cigarette sales to children. The longer such controls are delayed, the greater the number of children likely to want to access e-cigarettes illicitly once a ban on sales to children is imposed.