According to the most recent Office for National Statistics data, around one in six adults in Great Britain smoke tobacco.1 Much has been achieved in reducing smoking rates, with almost half of adults reporting being smokers in 1974, when Office for National Statistics data began. Nevertheless, smoking remains a major driver of ill health and health inequalities, and continues to be a major national and global cause of preventable death.2 Many adult smokers first took up smoking in adolescence,3 and earlier initiation is associated with greater endurance into adulthood.4 Hence, as well as effective intervention to support smokers to quit, reducing the prevalence of smoking requires prevention of new generations from taking up smoking.
Young people’s smoking peaked in UK nations in the mid to late 1990s, coinciding with publication of the government’s Smoking Kills: A White Paper on Tobacco.5 Subsequent decades have seen increasing regulation of the marketing of tobacco,6 as well as regulation on age of sales and legislation prohibiting smoking in public places.7 Declines seen for smoking uptake are mirrored internationally across other outcomes, such alcohol use,8 perhaps indicating that broader social trends beyond tobacco policy have contributed to wider changes in adolescent risk behaviour. Nevertheless, as a whole, tobacco regulation has likely contributed significantly to the prolonged downwards trajectory in young people’s smoking since the 1990s. Regular smoking among secondary school-aged young people is now estimated at around 2–4%,9 while experimentation with tobacco has become a minority behaviour of which most young people disapprove.10 Nevertheless, a substantial number of young people, particularly young people from poorer backgrounds, still leave school as smokers.9 As UK nations move towards ambitions to become smoke free by 2030,11,12 further action is needed to support cessation among smokers, while preventing the next generation from taking up smoking.
Recent years have seen increasing attention to potential (positive and negative) effects of e-cigarettes on smoking and smoking-related morbidity and mortality. Many people argue that e-cigarettes have potential to improve public health by offering smokers a less harmful alternative, or even make smoking obsolete.13 However, concerns have centred around whether or not the appeal of e-cigarettes goes beyond smokers, creating new generations of young people addicted to nicotine.14 In particular, there are concerns that e-cigarettes may lead to a reversal of successes in reducing the appeal of tobacco to young people.
The emergence of e-cigarettes in the UK
E-cigarettes and smoking cessation
Invented in China in 2003, e-cigarettes entered European Union (EU) markets around 200715 and began to gain significant traction as smoking-cessation aids around 2011, with rapid growth over the next couple of years.16 The devices themselves are rapidly changing, with early products looking very much like cigarettes. More recent changes in product design have moved away from resemblance to cigarettes, with increased popularity of tank-based devices. Many people argue that public health communities should endorse e-cigarettes as a means of helping smokers to quit the more harmful behaviour of smoking.17,18 Although long-term health effects are unlikely to be known for some time, some studies show important acute health benefits of switching to e-cigarettes.19 Hence, e-cigarettes, where effective as cessation aids, may play an important role in reducing mortality and morbidity.20
UK trials have found that e-cigarettes, accompanied with behavioural support, can be more effective than nicotine replacement therapy (NRT) in supporting cessation.21 Nevertheless, in early 2020, the World Health Organization still described evidence for the effectiveness of e-cigarettes in cessation as inconclusive.22 More recently, a Cochrane review concluded that there was moderate-certainty evidence that e-cigarettes are effective in helping smokers to quit and, indeed, appear to be more effective than NRT.23 Public Health England’s recent evidence update argues that evidence is now increasingly strong that e-cigarettes make a significant positive contribution to public health via supporting cessation.24 Indeed, reflecting this rapidly advancing evidence base, in 2021, regulators set out new guidance on how e-cigarettes might be offered on prescription as a smoking-cessation aid in England.25
Although there is emerging consensus that e-cigarettes have a role to play in harm-reduction strategies,26 there is also consensus that e-cigarettes are not harmless.27 Many people in the public health community continue to argue for caution until more is known about the potential harms of e-cigarettes.28,29 However, although much has been written about a binary divide between ‘enthusiasts’ who wish to harness the potential of e-cigarettes for cessation and ‘opponents’ who see e-cigarettes as a threat to young people,30 in reality, positions are often more nuanced. Many people are both enthusiastic about the role of e-cigarettes in cessation, while taking seriously the need to prevent use by children and young people.31 For people accepting a position that e-cigarettes have a role in supporting cessation, it remains important that use is restricted to smokers or ex-smokers as a means of quitting and preventing relapse back to smoking.32
Young people’s use of e-cigarettes
Today’s young people have grown up in an environment of falling prevalence of tobacco use, among adults and peers. However, today’s young people are also the first generation to have grown up in a time when e-cigarettes have become a major presence in UK society.33 The growing presence of e-cigarettes has led to some concerns that e-cigarettes might lead to a new generation of tobacco-naive young people addicted to nicotine. Some concerns were expressed early in the emergence of e-cigarettes that marketing for e-cigarettes used approaches that mirror those used by the tobacco industry, targeting young people as potential new consumers.34 Indeed, as tobacco companies have bought e-cigarette companies or developed their own products, there is increasing overlap between actors responsible for these markets.35 E-cigarettes commonly have features that many people argue might appeal to young people, such as their colourful and often sweetly flavoured nature.36,37
Young people’s use of e-cigarettes did not begin to be measured in UK social surveys until around 2013. Early estimates from Wales indicated that ever use of e-cigarettes was approximately equal to ever use of tobacco cigarettes, as 12% of young people had tried vaping and 12% had also tried smoking.38 However, the proportion of young people reporting more regular use was small. Over the next 2–3 years, data accumulated from surveys across UK nations (see Figure 1), reinforcing this picture of sharp increase in experimentation with e-cigarette use, but with regular use remaining low and concentrated among smokers.24,39
FIGURE 1 Percentages of young people who have ever used an e-cigarette or who use an e-cigarette at least weekly, by smoking status. Reproduced with permission from Bauld et al.39 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The figure includes minor additions and formatting changes to the original text. (Figure not reproduced here; see the original article.)
Nevertheless, rapid growth of experimentation in the early years of the emergence of e-cigarettes in the UK signalled a need for careful ongoing monitoring. Concerns that e-cigarettes may become harmful for young people centred both on potential indirect harms, via their perceived potential role in reversing successes in reducing smoking uptake, but also direct harms of use by young people who would not otherwise be smoking.
Effects of e-cigarettes on young people’s smoking uptake
There are two major hypothesis surrounding potential mechanisms through which e-cigarettes may lead to increases in young people’s smoking uptake: (1) the gateway hypothesis and (2) the renormalisation hypothesis.
The gateway hypothesis
The gateway hypothesis has a long history in drug research. The gateway hypothesis has long been debated, for example whether or not cannabis acts as a gateway into harder drug use.40 Applying the gateway hypothesis to e-cigarette use, it is assumed that e-cigarettes might appeal to young people who would not have otherwise smoked. Therefore, these young people may then become addicted to nicotine via use of e-cigarettes, increasing vulnerability to subsequent tobacco use.41 There is, indeed, now a very large and consistent body of systematic review-level evidence, much of which is from the UK, which finds that young people who try an e-cigarette are more likely to go onto smoke than young people who do not.42,43 The evidence is limited by issues such as publication bias, high sample attrition and inadequate adjustment for potential confounders. Hence, it remains unclear if using e-cigarettes causally increases risk of future smoking or if this association is explained by common liability, or if both offer a partial explanation. Recent studies find that although unadjusted associations of ever e-cigarette use with subsequent smoking are large, the associations become smaller or non-significant the more common causes are adjusted for.44 One large study in France found that among 17- to 19-year-olds, having tried an e-cigarette was associated with reduced likelihood of daily smoking, but this was moderated by age of first use, and very early experimenters who had used an e-cigarette by age 9 were more likely to become smokers.45 Given that few non-smokers were becoming regular users of e-cigarettes while experimentation was becoming widespread, it is likely that any gateway effects which are occurring may have to date occurred on a small scale, unlikely to reverse population trends in tobacco use.
The renormalisation hypothesis
The renormalisation hypothesis, although often conflated with the gateway hypothesis, differs from the gateway hypothesis in its focuses on sociological processes around norms for tobacco use. Several decades ago, smoking was a highly normalised behaviour.1 Smoking was a social practice adopted widely across socioeconomic groups, with non-smokers tolerant and accepting of others’ smoking. Although controversial in creation of stigma for those who continue to smoke, with this stigma falling mostly on already marginalised groups,46 much success in reducing population-level uptake of tobacco has been achieved via denormalising smoking. This denormalising has increasingly been achieved through excluding smoking from the rhythms and activities of daily life, restricting when and where smoking can take place to make tobacco an unattractive and unappealing product for non-smokers. Actions have included mandating health warnings on cigarette packs, with international evidence suggesting that these warnings can have important impacts on both cessation and uptake.47 Smoking in enclosed workplaces was prohibited in Scotland in 2006, and elsewhere in the UK in 2007. Although implemented largely to protect hospitality workers,48 impacts on children received much attention. Plans were met with arguments that this would harm children by displacing smoking into the home, but UK and international evidence found that this was not the case, as second-hand smoke exposure and smoking in the home declined following legislation.49,50 This legislation, recently selected by the Royal Society for Public Health (London, UK) as the greatest achievement of the twenty-first century,51 played a major role in communicating that smoking in front of non-smokers, including children, was unacceptable. It was followed by other moves, including increases in age of sales to 18 years and point of sale restrictions. Laws prohibiting proxy purchasing of tobacco sought to further restrict young people’s access to tobacco, with current arguments for increasing age of sales further to 21 years centred partly around further inhibiting the ability of young people to obtain cigarettes via older friends.52 From 2015, bans on smoking in cars carrying children were introduced.11
The hypothesis that e-cigarettes renormalise smoking is based around assumptions that because e-cigarettes mimic the act of smoking, their growing presence will reverse successes in denormalising smoking.53 In support of this notion, one recent USA-based study found that greater exposure to second-hand e-cigarette aerosol in public places was associated with increased tolerance of, and susceptibility to, future smoking.54 However, an increasing number of studies focused on whether or not smoking rates among young people increase in parallel with growing population prevalence of e-cigarettes have offered limited support for this notion. Recent international studies from Taiwan55 and New Zealand56 find that smoking rates fell as fast, or faster, during the emergence of e-cigarettes than in the preceding years. Analysis of data from the National Youth Tobacco Survey (NYTS) in the USA also indicate continuing decline in tobacco use during this period, with ever smoking declining at a faster rate post 2011, although the rate of decline in current smoking began to plateau in 2014.57 Also using NYTS data, Foxon and Selya58 found that a counterfactual trend projected from data up to 2009 overestimated observed smoking rates from 2010 to 2018. Sokol and Feldman59 similarly found that projections of smoking rates for US 12th graders (i.e. aged 17/18 years) from Monitoring the Futures data prior to 2014 overestimated prevalence to 2018, concluding that many adolescent e-cigarette users would likely otherwise have become smokers.59
Notions that e-cigarettes renormalise the different, but related, behaviour of tobacco smoking centre around the assumption that the behaviours are viewed by young people as sufficiently related that growth in one will renormalise the other. The extent to which this assumption is upheld may vary across contexts and may be influenced, in part, by how products are positioned in public policy and discourse. In the USA, the Food and Drug Administration include e-cigarettes as ‘tobacco products’.60 However, given that e-cigarettes do not use tobacco, this labelling has been described as a particularly US phenomenon.61 There is some evidence from the UK that finds that children see tobacco and e-cigarettes as distinct products.62,63 In one study, for example, young people whose parents used e-cigarettes were most likely to perceive them as things adults use to stop smoking, with this perception of e-cigarettes as cessation aids associated with lower smoking susceptibility.64 Hence, where perceived as stop smoking aids, others’ use of e-cigarettes might be interpreted by young people as a social display of efforts to give up smoking, rather than as an endorsement of a smoking-like behaviour.
Use of e-cigarettes among young people as an emerging public health risk in its own right
Although links between e-cigarettes and smoking have dominated debates regarding potential harms of e-cigarettes to young people, e-cigarettes themselves are unlikely to be harmless. Hence, although switching from tobacco to e-cigarettes is likely to reduce harm, taking up e-cigarette use without prior smoking introduces new potential for, as yet, unknown harms. In one recent study,65 estimates of the extent to which risk-adjusted nicotine product days decreased over time as tobacco use fell but e-cigarette use grew were described as highly dependent on assumptions regarding risks of e-cigarettes relative to tobacco. Although commonly based on animal models and, therefore, with questionable transferability, common arguments for regulating e-cigarettes to prevent young people from using them have included potential impacts of nicotine on brain development among young people.66
As described, few young people in the UK have historically used e-cigarettes regularly, unless they are also smokers. However, in the USA, the 2018 NYTS indicated that in the space of a year the percentage of high school students reporting past 30-day use of e-cigarettes doubled to almost one in five, with more than one in four young people reporting past 30-day vaping in 2019.67 This growth was framed widely as an epidemic of young people’s vaping,68 triggering international calls for regulation. In 2019, Hammond et al. published an influential analysis of change in young people’s smoking and e-cigarette use from 2017 to 2018, finding increases in vaping in the USA and Canada, although no growth in England. In Canada, analyses indicated that this increase in vaping was accompanied by growth in smoking.69 In 2020, the authors published an update in which data were calibrated with external data sources, with the consequence that the previously reported increase in smoking in Canada was attenuated. Hence, although revised analysis weakened common interpretations that e-cigarettes were acting as a pathway into smoking, the analysis did signal rapid growth in young people’s use of e-cigarettes as a potential problem in its own right. Newer data from the 2020 and 2021 NYTSs indicate that rapid growth in vaping from 2017 to 2019 did quickly begin to subside.70,71
In the context of these rising international concerns about the growth of young people’s use of e-cigarettes, concerns regarding the safety of e-cigarettes, particularly for young people, were further raised by the EVALI (e-cigarette or vaping use-associated lung injury) outbreak.67 This was an outbreak of vaping-related lung injury, which led to a number of deaths, predominantly in the USA in 2019, attributed to the specific chemicals tetrahydrocannabinol and vitamin E acetate,72 in unregulated devices. The outbreak received intense media coverage, and one study73 found the outbreak to be associated with increased risk perceptions for e-cigarettes among English adult smokers. Following this outbreak, debate regarding potential for an epidemic of young people’s smoking in the UK intensified.74 One group of paediatric health-care professionals, for example, wrote that with 30% of high school children in America using e-cigarettes, most of whom had never smoked tobacco, similar rates were inevitable in the UK.75 The claim that most use was occurring among tobacco-naive young people is not borne out by reanalysis of the NYTS data, which finds that most use, particularly regular use, occurred among young people who had smoked.76 However, this reflected growing concerns that the US epidemic would cross the Atlantic and that e-cigarette use in itself might become a commonly adopted risk behaviour among young people who would not otherwise be smokers.
E-cigarette regulation in the UK
Divergence in positions on e-cigarettes among the public health community is reflected in international approaches to regulation,77 which have ranged from highly restrictive regulation in countries such as Australia78 to more liberal approaches in countries such as the UK, where no specific regulation existed prior to 2015. Age of sales regulations for e-cigarettes were implemented in England and Wales in late 2015 and in Scotland in early 2017. In Wales, the Welsh Government’s 2015 Public Health Bill attempted to introduce legislation that would have prohibited the use of e-cigarettes in any place where tobacco cigarettes are currently prohibited.79 This failed to pass into law after a last-minute loss of cross-party support led to a minority Labour government losing at the final stage of the legislative process. The bill returned in 2017, re-introduced by a minority Labour Government, with provisions relating to e-cigarettes removed.80 Key arguments against including e-cigarettes alongside bans on smoking in public places have included that this may undermine quit attempts by reducing relative advantage of e-cigarettes over tobacco.81
In the meantime, in May 2016, Tobacco Products Directive (TPD) regulations were introduced in EU member states. The TPD regulations included a broad suite of regulations for tobacco, many of which reflected tobacco control actions already implemented in the UK, with exceptions including a ban on sales of smaller packs of tobacco cigarettes and (later) bans on sale of menthol-flavoured cigarettes.82 Article 20 was specific to e-cigarettes, and represented the first major supra-national regulation of e-cigarettes. TPD regulations prohibited cross-border advertising of e-cigarettes, with immediate effect from May 2016. Recent market liberalisation in Canada has been associated with increased marketing exposure and use of e-cigarettes among young people, with comprehensive provincial restrictions associated with lower exposure and use.83 Hence, it is plausible that marketing restriction within TPD may act to reduce young people’s exposure and use. TPD also included a suite of regulations on the products themselves, which were to be more gradually introduced, with full implementation to be achieved by May 2017. These regulations included a mandatory warning across 30% of the packet indicating that the products contain nicotine, which is a highly addictive substance. Manufacturers are mandated to notify the Medicines and Healthcare products Regulatory Agency of intention to launch new products 6 months in advance, including notification of all product ingredients. e-cigarettes and liquids are to be sold in tamper-proof containers to include an information leaflet listing all ingredients, with nicotine strength limited to 20 mg/ml. Device refills were restricted in size. TPD regulations aligned e-cigarettes with tobacco in so far as they involve regulation under the banner of a ‘Tobacco Products Directive’, but e-cigarettes themselves are not referred to as tobacco products. There is some limited evidence that first use of flavoured e-cigarettes is associated with more persistent use among young people.84 Hence, some EU member states went beyond mandatory requirements of TPD regulations, through introduction of flavour regulations left within TPD for individual member states to determine,77 although the UK has not to date.
The regulations were a cause of concern for the vaping community and some scientists who argued, in particular, that a reduction in nicotine strength might inhibit the usefulness of e-cigarettes as cessation devices or cause users to relapse to smoking, as the lower strength may not satisfy cravings.85 The warning focused on the addictiveness of nicotine was found in one study86 to likely put smokers off using e-cigarettes as cessation devices. However, surveys of adult smokers, ex-smokers and vapers indicate that, after implementation, the regulations went largely unnoticed by many, with most using compliant devices while unaware that these had changed.87 E-cigarettes remain the most popular smoking-cessation devices used by smokers and recent ex-smokers in the UK.16 Part of the rationale for introduction of these regulations include assumptions that e-cigarettes can develop into a gateway to nicotine addiction and tobacco consumption, and that e-cigarettes mimic and normalise the act of smoking.82 To date, however, there has been limited evaluation in the UK of (1) the role of e-cigarettes in renormalising smoking and (2) the impacts of this regulation on children and young people’s use of e-cigarettes (and tobacco).
Objectives
This report describes findings from a mixed-method natural experimental evaluation of the impacts of TPD regulations in the UK. Theoretically, we view e-cigarette regulation as an ‘event’ within a complex system88,89 that has the potential to bring about (intended and unintended) change through altering the behaviours and interactions of a diverse range of actors, including regulators, retailers, consumers and, ultimately, young people. Hence, the research first explores system history and starting points prior to regulations through examining trends in young people’s tobacco use in the lead up to regulation, addressing questions of whether or not e-cigarettes were re-normalising smoking in the lead up to implementation of TPD and exploring normative perceptions for e-cigarettes as a product in their own right. The study does this through combining quantitative analyses of trends in secondary school-based survey data in the lead up to regulation with qualitative pupil interviews, exploring perceived norms in relation to tobacco and e-cigarettes. Subsequently, interviews with policy stakeholders, trading standards officers (TSOs) and retailers are used to understand implementation processes, whereas retailer audits examine implementation fidelity. Mechanisms through which TPD regulations might achieve impact on young people’s use of e-cigarettes, and smoking, are investigated using repeated qualitative pupil interviews prior to use of before-and-after survey data to evaluate change in trend for e-cigarette use, and smoking, following TPD implementation.