Trends in Long-Term Vaping Among Adults in England, 2013-2023 (PDF)

RESEARCH BMJ: first...

RESEARCH BMJ: first published as 10.1136/bmj-2023-079016 on 17 July 2024. Downloaded from http://www.bmj.com/ on 22 July 2024 at Univ of Alberta Library Bibliographic Service Unit-Serials. Trends in long term vaping among adults in England, 2013-23: population based study Sarah E Jackson,1,2 Harry Tattan-Birch,1,2 Lion Shahab,1,2 Jamie Brown1,2 1 Department of Behavioural ABSTRACT October 2013 and March 2021, most long term Science and Health, University OBJECTIVE vapers mainly or exclusively used refillable electronic College London, London, UK To examine trends in long term (>6 months) vaping cigarettes (2.5% to 3.3% of adults) and few (0.1% of 2 SPECTRUM Consortium, among adults in England. adults) used disposable devices. However, prevalence Edinburgh, UK DESIGN of long term vaping using disposable devices Correspondence to: S E Jackson [email protected] Population based study. subsequently rose rapidly, and by October 2023 (or @DrSarahEJackson on X; similar proportions of adults mainly or exclusively ORCID 0000-0001-5658-6168) SETTING used disposable and refillable devices (4.9% (4.2% to Additional material is published England. 5.7%) and 4.6% (4.0% to 5.3%), respectively). online only. To view please visit PARTICIPANTS the journal online. CONCLUSIONS 179 725 adults (≥18 years) surveyed between October Cite this as: BMJ 2024;386:e079016 The prevalence of long term vaping increased http://dx.doi.org/10.1136/ 2013 and October 2023. substantially among adults in England during bmj‑2023‑079016 MAIN OUTCOME MEASURES 2013-23. Much of this increase occurred from 2021, Accepted: 30 May 2024 Time trends in prevalence of long term vaping using coinciding with the rise in popularity of disposable logistic regression, overall and by vaping frequency e-cigarettes. Half of long term vapers now mainly or (daily or non-daily), and main type of device used exclusively use disposable devices. The growth was (disposable, refillable, or pod). concentrated among people with a history of regular RESULTS smoking, but an increase also occurred among people The proportion of adults reporting long term vaping who never regularly smoked, especially young adults. Protected by copyright. increased non-linearly, from 1.3% (95% confidence interval 1.1% to 1.5%) in October 2013 to 10.0% Introduction (9.2% to 10.9%) in October 2023, with a particularly Electronic cigarettes are an effective tool for smoking pronounced rise from 2021. This rise included an cessation1 2 and are likely much less harmful than increase in long term daily vaping, from 0.6% (0.5% conventional cigarettes.3 They are not, however, risk- to 0.8%) to 6.7% (6.0% to 7.4%). Absolute increases free, particularly for people who have never smoked.3 in long term vaping were larger among people with As such, a challenge many countries currently a history of regular smoking (current smokers: 4.8% face is how to regulate e-cigarettes so that they are (4.0% to 5.8%) to 23.1% (20.4% to 25.9%); recent available and appealing to smokers as a means to former smokers: 5.7% (3.4% to 9.2%) to 36.1% quit smoking while minimising uptake among those (27.6% to 45.4%); long term former smokers: 1.4% who would not have otherwise smoked. Until recently, (1.0% to 1.9%) to 16.2% (14.2% to 18.4%)), but an England appeared to have struck the right balance, increase also occurred among people who had never with e-cigarettes widely used by smokers in quit regularly smoked (0.1% (0.0% to 0.2%) to 3.0% attempts4 and rarely used by young people and never (2.3% to 3.8%)). Growth was also more pronounced in smokers.5 6 However, the prevalence of vaping has young adults (eg, reaching 22.7% (19.2% to 26.5%) risen in recent years, particularly among adolescents of 18 year olds v 4.3% (3.6% to 5.2%) of 65 year (11-17 years)5 and young adults (18-24 years).4 6 This olds), including among those who had never regularly increase in prevalence has largely been attributed smoked (reaching 16.1% (11.1% to 22.7%) of 18 year to the introduction of new disposable devices.7 8 It is olds v 0.3% (0.1% to 0.6%) of 65 year olds). Between not clear how far this rise reflects an increase in short term, experimental use versus sustained, regular use. WHAT IS ALREADY KNOWN ON THIS TOPIC In addition, little is known about how the types of Vaping prevalence has increased substantially in England since new disposable products used by long term vapers—and how often electronic cigarettes became popular in mid-2021, particularly among they use them—is changing over time. Understanding adolescents and young adults this is important for gauging the likely public health and environmental effects of this rise in vaping uptake It is not clear how far this reflects an increase in experimental use versus long and for informing policy decisions. term, regular use The use of e-cigarettes increased rapidly among In addition, little is known about how the types of products used by long term adolescents and young adults in the US between 2015 vapers are changing over time and 2019, resulting in widespread concern about WHAT THIS STUDY ADDS public health.9 10 Further analyses of the data showed that much of the use was experimental,11 and evidence Long term vaping has noticeably increased among young adults since 2021, of an overall increase in the population burden of including among those who have never regularly smoked nicotine dependence was limited.12 Consistent with Half of long term vapers now mainly or exclusively use disposable e-cigarettes, this pattern of use, and following new regulations and most are using them every day restricting the availability of e-cigarettes, the the bmj | BMJ 2024;386:e079016 | doi: 10.1136/bmj-2023-079016 1  RESEARCH BMJ: first published as 10.1136/bmj-2023-079016 on 17 July 2024. Downloaded from http://www.bmj.com/ on 22 July 2024 at Univ of Alberta Library Bibliographic Service Unit-Serials. prevalence of use in the same population subsequently Before the covid-19 pandemic these areas were declined.13 Recently in the UK, vaping has increased randomly allocated to interviewers who travelled to noticeably among adolescents and young adults.5 7 It their selected areas to engage with one household is important to understand the extent to which this member aged ≥16 years. Face-to-face computer represents experimental or dependent use, and how assisted interviews were conducted until quotas based this differs across subgroups of the population (eg, on factors influencing the probability of being at home by smoking status, age, gender, and socioeconomic (ie, working status, age, and gender) were fulfilled. position). One indicator of more dependent use is the Morning interviews were avoided to maximise the prevalence of long term vaping. availability of participants. Long term vaping could be harm reducing or In-person interviews were halted in March 2020 harmful to public health, depending on who is using owing to covid-19 social distancing restrictions and the products and what they would otherwise be doing. replaced by telephone interviews from April 2020 Transitioning to long term vaping from cigarette onwards with adults aged ≥18 years. The lower age limit smoking—the most harmful form of nicotine use— returned to 16 years from January 2022. Telephone would reduce harm for people who are unable or interviews are now conducted by landline or mobile unwilling to stop all nicotine use.3 Conversely, long using random digit dialling or by targeted mobile. For term vaping could lead to harm among those who the sample processed, each eligible landline telephone would have never started smoking cigarettes.3 Long number across England has a random probability of term vaping with disposable e-cigarettes specifically selection proportionate to population distribution (ie, may also have a substantial environmental impact14: stratification of the landline telephone database by these products are designed for single use, so they and within Government Office Region; see19 for further generate more waste than rechargeable vaping details). To maximise response rates, sampling takes products (ie, refillable and pod devices).15 16 place mostly by landline earlier in the day and mostly The Smoking Toolkit Study (a nationally by mobile phone later in the day. representative, cross sectional survey) collects detailed The two methods of data collection show good Protected by copyright. data on vaping among adults in England each month. comparability: when social distancing restrictions In the current study we used data collected from 2013 were lifted, we ran a parallel telephone and face-to- to 2023 to examine how the prevalence of long term face survey wave and obtained similar estimates for (>6 months) vaping changed overall and by vaping key sociodemographic, smoking, and nicotine product frequency (daily or non-daily) in adults during this use measures.20 period, and the main type of device used (disposable, For this study, we used data collected from refillable, or pod). We also investigated whether participants surveyed between October 2013 (the first changes in any long term vaping and in long term wave to assess vaping among all adults) and October vaping using a disposable device differed by smoking 2023 (the most recent data at the time of analysis). status, age, gender, and occupational social grade. Since April 2022, vaping duration and device characteristics have only been assessed quarterly as a Methods means of reducing the cost of data collection for the Pre-registration Smoking Toolkit Study owing to limited funding. We The study protocol and analysis plan were pre- excluded waves in which vaping duration were not registered on Open Science Framework (https://osf. assessed (May, June, August, September, November, io/n2785/). We also carried out additional unplanned and December 2022, and February, March, May, analyses before and after peer review (see statistical August, and September 2023); trends were modelled analysis section). based on data collected in all other waves. Data on the main device type used have only been collected since Design July 2016, so trends in this outcome were limited to Data were drawn from the ongoing Smoking the period from July 2016 to October 2023. Because Toolkit Study, a monthly cross sectional survey of a data were not collected from teenagers aged 16 and representative sample of adults in England.17 The study 17 years between April 2020 and December 2021, uses a hybrid of random probability and simple quota we restricted our sample to those aged ≥18 years for sampling to select a new sample of about 1700 teenagers consistency across the time series. and adults (≥16 years) each month. Comparisons with sales data and other national surveys indicate that key Measures variables, including sociodemographic characteristics, Long term vaping smoking prevalence, and cigarette consumption, are Long term vaping was defined as current vaping for nationally representative.17 18 a period of more than six months. Current vaping The methods have been described in detail was assessed within several questions asking about elsewhere.17 19 Briefly, England is split into 165 665 use of a range of nicotine products, depending on output areas, each comprising about 300 households. the participant’s smoking status. Current smokers These areas are then stratified according to established were asked: “Which, if any, of the following are you geodemographic characteristics and geographical currently using to help you cut down the amount region, then randomly selected into an interviewer’s list. you smoke?” and “Do you regularly use any of the 2 doi: 10.1136/bmj-2023-079016 | BMJ 2024;386:e079016 | the bmj  RESEARCH BMJ: first published as 10.1136/bmj-2023-079016 on 17 July 2024. Downloaded from http://www.bmj.com/ on 22 July 2024 at Univ of Alberta Library Bibliographic Service Unit-Serials. following in situations when you are not allowed to liquids (rechargeable)” or “A modular system that you smoke?”; current smokers and those who had quit refill with liquids (you use your own combination of in the past year were asked: “Can I check, are you separate devices: batteries, atomizers, etc.)”), and pod using any of the following either to help you stop (“An e-cigarette or vaping device that uses replaceable smoking, to help you cut down or for any other reason pre-filled cartridges (rechargeable).”) at all?”; and non-smokers were asked: “Can I check, are you using any of the following?” We considered Smoking status those who reported using an e-cigarette (ie, selected Smoking status was assessed by asking participants the response option “electronic cigarette” or “Juul”) which of the following best applied to them: I smoke in response to any of these questions to be current cigarettes (including hand-rolled) every day; I smoke vapers. Other response options included different cigarettes (including hand-rolled), but not every day; I types of nicotine replacement therapy (eg, nicotine do not smoke cigarettes at all, but I do smoke tobacco gum, nicotine patch), heated tobacco products (heat- of some kind (eg, pipe, cigar, or shisha); I have stopped not-burn cigarette (eg, iQOS, heatsticks)), and nicotine smoking completely in the last year; I stopped smoking pouches (tobacco-free nicotine pouch or pod or “white completely more than a year ago; and I have never been pouches” placed on the gum). The supplementary file a smoker (ie, smoked for a year or more). provides details of these items (including an additional Those who responded to the first three of these question on nicotine vaping assessed in a subset of options were considered current smokers; those who waves). These items capture all vaping rather than just responded they had stopped smoking in the past year nicotine vaping, although most vapers (87%) sampled were considered recent former smokers; and those in waves that assessed nicotine content said that their who responded they had stopped smoking more usual device contained nicotine. than a year ago were considered long term former Current vapers were asked: “How long have you smokers. Those who responded they had never been been using this nicotine replacement product or these a smoker were considered never-smokers, and given products for?” Response options were 6-12 weeks, >12-26 weeks, >26-52 weeks, and capture some people who have smoked at all in the >52 weeks. past. However, our question does ask people to choose Participants who reported vaping for more than which of the response options best applies to them; six months were considered long term vapers.21 This the “smoked for a year or more” is an elaboration definition was used to indicate possible dependent rather than a definition. As such, it is likely that use. Because this measure for duration of use was not people who have smoked cigarettes with any kind specific to vaping but applied to all non-combustible of recent frequency (regardless of whether it is more nicotine products the participant reported using, we than a year) are more likely to respond to having conducted a sensitivity analysis in which we restricted smoked cigarettes (including hand-rolled) every day the definition of long term vaping to those reporting no or not every day (if currently doing so) or to have current use of nicotine replacement therapy, heated stopped smoking completely in the past year (if they tobacco products, or nicotine pouches. have recently stopped). Other national surveys use alternative definitions, but the surveys have produced Vaping frequency consistently similar estimates of prevalence, and the Vaping frequency was assessed by asking vapers: Smoking Toolkit Study and sales data are also closely “How many times per day on average do you use your aligned.22 nicotine replacement product or products?” Response options were 1, 2, 3-4, 5-7, 8-11, ≥12, not every day Sociodemographic characteristics but at least once a week, not every day and less than We modelled age as a continuous variable using once a week, and don’t know. restricted cubic splines. Descriptive data were also We considered those who reported use at least once provided by age group (18-24, 25-34, 35-44, 45-54, a day to be vaping daily and those who reported use 55-64, and ≥65 years). less than once a day to be vaping non-daily. Because Gender was self-reported as man or woman. In more this measure was not specific to vaping, we conducted recent waves, participants have also had the option a sensitivity analysis restricting the definition of daily to describe their gender in another way; those who and non-daily vaping to those reporting no current identified in another way were excluded from analyses use of nicotine replacement therapy, heated tobacco by gender owing to low numbers. products, or nicotine pouches. Occupational social grade was categorised as ABC1 (includes managerial, professional, and Main device type upper supervisory occupations) and C2DE (includes We assessed the main type of device used from July manual routine, semi-routine, lower supervisory, 2016 onwards by asking vapers: “Which of the state pension, and long term unemployed). This following do you mainly use...?” Response options occupational measure of social grade is a valid index of were disposable (“A disposable e-cigarette or vaping socioeconomic position that is widely used in research device (non-rechargeable)”), refillable (“An e-cigarette in UK populations. It has been identified as particularly or vaping device with a tank that you refill with relevant in the context of tobacco use.23 the bmj | BMJ 2024;386:e079016 | doi: 10.1136/bmj-2023-079016 3 RESEARCH BMJ: first published as 10.1136/bmj-2023-079016 on 17 July 2024. Downloaded from http://www.bmj.com/ on 22 July 2024 at Univ of Alberta Library Bibliographic Service Unit-Serials. Statistical analysis Unplanned analyses Data were analysed in R version 4.2.1. The Smoking Before peer review, we modelled age specific trends Toolkit Study uses raking to weight the sample to in long term vaping (ie, repeated the model testing match the population in England. This profile is the interaction between age and time) among determined each month by combining data from the participants who had never regularly smoked—those UK Census, the Office for National Statistics mid-year who responded that they had never been a smoker (ie, estimates, and the National Readership Survey.17 The smoked for a year or more). This allowed us to examine following analyses used weighted data. Missing cases the extent to which the increase in long term vaping we were excluded on a per analysis basis. observed among never smokers differed by age. After peer review, we added five further unplanned Trends in long term vaping among adults analyses. The first was a segmented regression analysis Among all adults, we reported the prevalence and that tested the association of the rise in popularity of corresponding 95% confidence interval (CI) of six vaping using disposable devices among young adults long term vaping categories by survey year: vaping in England with changes in the trend in long term (>6 months), daily vaping (>6 months, and currently vaping. Previous data from the Smoking Toolkit Study vaping daily), non-daily vaping (>6 months, and indicated that disposable e-cigarette use in Great currently vaping non-daily), vaping using a disposable Britain was relatively rare (6 months, and currently mainly or exclusively rose gradually to 2% (11% among 18 year olds) by using disposable e-cigarettes), vaping using a refillable April 2022.7 We therefore split the time series into two device (>6 months, and currently mainly or exclusively periods: October 2013 to May 2021 (pre-disposables using refillable e-cigarettes), and vaping using a pod period) and June 2021 to October 2023 (disposables device (>6 months, and currently mainly or exclusively period). We used logistic regression to model the trend using pod e-cigarettes). in long term vaping before the interruption (underlying We used logistic regression to analyse trends in secular trend; coded 1... n, where n was the total these long term vaping outcomes over the study number of waves) and the change in the trend (slope) Protected by copyright. period, with time (survey month) modelled using post-disposables relative to pre-disposables (coded 0 restricted cubic splines with five knots (decided before the rise in disposable e-cigarette use, and 1... m a priori and pre-registered, on the basis that this from June 2021 onwards, where m was the number number would be sufficient to accurately model of waves after June 2021). We assumed linear trends trends across years without overfitting; in cubic pre-interruption and post-interruption. The model was spline models, the results are robust to the specific adjusted for seasonality, which was modelled using a position and number of knots (at least when choosing smoothing term with cyclic cubic splines specified. We between four or five knots)24). This approach allowed also adjusted for the onset of the covid-19 pandemic for flexible and non-linear changes over time, while (coded 0 to February 2020 and 1 from March 2020), to avoiding categorisation. account for any influence of the pandemic on vaping patterns. For ease of interpretation, we multiplied Differences by smoking status, age, gender, and coefficients by 12 to convert results from monthly to occupational social grade annual trends. To explore moderation of trends in any long term The second unplanned analysis added after the peer vaping and in long term vaping using a disposable review repeated our (spline and segmented regression) device by smoking status, age, gender, and trend analyses restricting the definition of long term occupational social grade, we repeated these models vaping to those who did not also use other non- including the interaction between the moderator of combustible nicotine products, to explore the possible interest and time—thus allowing for time trends to influence of the measures assessing duration and differ across subgroups. Each of the interactions was frequency of use not referring exclusively to vaping. tested in a separate model. Age was modelled using The third repeated our primary (spline) models for restricted cubic splines with three knots (placed at long term daily vaping and long term non-daily vaping the 5%, 50%, and 95% quantiles), to allow for a among participants who had never regularly smoked, non-linear association between age and long term to explore whether the growth in vaping among this vaping. subgroup was predominantly daily or non-daily. The We used predicted estimates from these models fourth repeated our primary models for long term to plot the prevalence of long term vaping over the vaping, among all adults and by age, with adjustment study period among all adults and within each for psychological distress, to explore the extent to subgroup of interest. As age was modelled which the rise in long term vaping we observed may continuously using splines, we displayed estimates have been driven by rising levels of distress in the for six specific years of age (those aged exactly population over this period, particularly among young 18, 25, 35, 45, 55, and 65 years) to illustrate how adults.25 Because data on psychological distress were trends differed across the age spectrum. The model only collected between April 2020 and June 2023, used to derive these estimates included data from these analyses were restricted to participants surveyed participants of all ages, not only those with the during this period, and we reduced the number of previous exact specified ages. knots used to model time from five to three to avoid 4 doi: 10.1136/bmj-2023-079016 | BMJ 2024;386:e079016 | the bmj  RESEARCH BMJ: first published as 10.1136/bmj-2023-079016 on 17 July 2024. Downloaded from http://www.bmj.com/ on 22 July 2024 at Univ of Alberta Library Bibliographic Service Unit-Serials. overfitting the data (given the much shorter time vaping increased from 1.3% to 10.0% (table 1). period). The fifth provided detailed descriptive data The increase over time was non-linear: prevalence on vaping frequency among long term vapers (ie, the increased from 1.3% to 3.3% between October 2013 proportion selecting each response option for the and July 2017, was stable at 3.3% between July item assessing vaping frequency), to provide more 2017 and August 2019, then increased again—with information beyond daily versus non-daily use. a particularly sharp rise from late 2021—reaching 10.0% by October 2023 (fig 1). Patient and public involvement The unplanned segmented regression analysis was The wider Smoking Toolkit Study is discussed consistent with this rise being associated with an several times a year with a diverse patient and increase in popularity of disposable e-cigarettes (fig 1). public involvement group, and the authors regularly Before June 2021, the prevalence of long term vaping attend and present at meetings at which patients increased by 11.3% per year (relative risk for trend and members of the public are included. Interaction 1.113 (95% CI 1.092 to 1.133)). However, this trend and discussion at these events help shape the broad increased sharply from when disposable e-cigarettes research priorities and questions. A mechanism also became popular (relative risk for trend change 1.245 exists for generalised input from the wider public: (1.181 to 1.312)), with prevalence rising by 38.6% each month, interviewers seek feedback on the per year since (relative risk for trend×relative risk for questions from all respondents, who are representative trend change 1.113×1.245=1.386). These percentages of the English population. This feedback is limited represent the yearly relative rather than absolute and usually relates to understanding of questions and percentage point increase. item options. No patients or members of the public A greater increase occurred in long term daily were involved in setting the research questions or vaping than long term non-daily vaping over time the outcome measures, nor were they involved in the (table 1 and supplementary table S2). In October design and implementation of this specific study. 2013, equal proportions of long term vapers reported vaping daily and non-daily (0.6% and 0.6% of adults, Protected by copyright. Results respectively; table 1; estimates do not sum to the A total of 197 266 (unweighted) adults aged ≥18 years total prevalence of long term vaping owing to some in England were surveyed between October 2013 and missing data on vaping frequency). The trend in long October 2023. We excluded 17 541 surveyed in months term daily vaping mirrored the trend in any long term in which vaping duration was not assessed, resulting daily vaping, increasing to 6.8% of adults by October in a final analytical sample of 179 725 participants. Of 2023 (fig 2). Meanwhile, the prevalence of long term these, 125 751 were surveyed between July 2016 and non-daily vaping remained relatively stable (between October 2023 and provided data for analyses by the 0.6% and 0.7%) up to May 2021, then increased to main type of vaping device used. Supplementary table 1.6% by October 2023 (fig 2). The difference between S1 shows the characteristics of the whole analysed the prevalence of long term daily vaping compared sample and the subsample analysed by device type. with long term non-daily vaping was less pronounced among never smokers (1.5% (1.1% to 2.1%) v 0.6% Trends in long term vaping among adults (0.3% to 0.9%) in October 2023; supplementary Our primary model indicated that across the study figure S1). When we investigated changes in vaping period, the proportion of adults reporting long term frequency among long term vapers in detail (using Table 1 | Modelled estimates of prevalence of long term vaping among adults in England at start and end of study period, overall and by vaping frequency and main type of device used Prevalence, % (95% CI)* Outcomes for long term vaping October 2013 July 2016 October 2023 Overall 1.3 (1.1 to 1.5) 3.0 (2.8 to 3.2) 10.0 (9.2 to 10.9) By vaping frequency†: Daily vaping 0.6 (0.5 to 0.8) 2.1 (1.9 to 2.3) 6.7 (6.0 to 7.4) Non-daily vaping 0.6 (0.5 to 0.8) 0.7 (0.6 to 0.8) 1.6 (1.3 to 2.0) By main device type used†: Disposable – 0.1 (0.0 to 0.2) 4.9 (4.2 to 5.7) Refillable – 2.5 (2.1 to 2.8) 4.6 (4.0 to 5.3) Pod – 0.3 (0.2 to 0.5) 1.0 (0.8 to 1.3) CI=confidence interval. *Data for October 2013, July 2016, and October 2023 are weighted estimates of prevalence in these months from logistic regression, with survey month modelled non-linearly using restricted cubic splines (five knots). October 2013 and October 2023 were the first and last months in the time series, and July 2016 was the first month in which the main device type used by vapers was assessed. †Prevalence estimates do not sum to total prevalence of long term vaping owing to missing data (participants did not respond or responded don’t know) and each trend being modelled separately. Modelled estimates by device type sum to slightly more than the total modelled estimate, but in the unmodelled data (supplementary table S2) estimates by device type consistently sum to slightly less than the total estimate (consistent with a small deficit owing to missing data). Supplementary table S2 provides unmodelled estimates of prevalence within each survey year. the bmj | BMJ 2024;386:e079016 | doi: 10.1136/bmj-2023-079016 5 RESEARCH BMJ: first published as 10.1136/bmj-2023-079016 on 17 July 2024. Downloaded from http://www.bmj.com/ on 22 July 2024 at Univ of Alberta Library Bibliographic Service Unit-Serials. Primary analysis and September 2017, was stable at 2.7% to September Prevalence of long term vaping (%) 12 2019, then increased to 4.6% by October 2023 (fig 2). 10 The prevalence of long term vaping using a pod device was roughly stable (between 0.3% and 0.5%) from 8 October 2013 to March 2021, then increased steadily 6 to 1.0% by October 2023 (fig 2). Across the study period, 577/6173 long term vapers 4 (9.3%) also used other non-combustible nicotine products (of whom 516 used nicotine replacement 2 therapy, 48 heated tobacco products, and 36 nicotine 0 pouches). Results of sensitivity analyses in which these participants were not counted as long term Unplanned sensitivity analysis vapers showed a similar pattern (supplementary table Prevalence of long term vaping (%) 12 S4 and figures S2 and S3), although absolute estimates 10 of prevalence were slightly lower—for example, the overall estimate for prevalence of long term vaping 8 in October 2023 was 9.1% (95% CI 8.3% to 9.9%) 6 compared with 10.0% (9.2% to 10.9%) in the primary analysis. 4 2 Differences by smoking status, age, gender, and occupational social grade 0 The increase in long term vaping occurred 3 4 5 6 7 8 9 0 1 2 3 predominantly among current and former smokers, t1 t1 t1 t1 t1 t1 t1 t2 t2 t2 t2 Oc Oc Oc Oc Oc Oc Oc Oc Oc Oc Oc but a rise also occurred among never smokers in more Month Protected by copyright. recent years (from 6 months) increased from 2.5% to 2.7% between October 2013 vaping increased substantially from 2013 to 2023. In 6 doi: 10.1136/bmj-2023-079016 | BMJ 2024;386:e079016 | the bmj  RESEARCH BMJ: first published as 10.1136/bmj-2023-079016 on 17 July 2024. Downloaded from http://www.bmj.com/ on 22 July 2024 at Univ of Alberta Library Bibliographic Service Unit-Serials. Vaping frequency devices. The prevalence of long term vaping using Prevalence of long term vaping (%) 8 Daily a pod device increased between March 2021 and Non-daily October 2023 but remained relatively rare. 6 Comparison with other studies 4 The timing of the sharp rise in long term vaping coincided with the rising popularity of new disposable e-cigarettes from Spring 2021.7 8 This suggests that the 2 recent increase in vaping among adults—particularly young adults—in England does not just reflect an 0 increase in experimental use, but rather that a substantial number of those taking up vaping are going Main device type used* on to vape long term. Our data also suggest that most Prevalence of long term vaping (%) 8 Disposable of those vaping long term are vaping daily, including Pod more than two thirds of those who have never regularly 6 Refillable smoked (ie, it is not just infrequent use over a long period). In addition, although in recent years small 4 increases in long term vaping using refillable and pod devices have occurred, the increase in long term vaping 2 using disposable devices was substantially larger. This indicates that many vapers are now continuing to use disposable devices over the long term, rather than 0 transitioning to rechargeable devices. 3 4 5 6 7 8 9 0 1 2 3 We observed notable differences in trends in long t1 t1 t1 t1 t1 t1 t1 t2 t2 t2 t2 Oc Oc Oc Oc Oc Oc Oc Oc Oc Oc Oc term vaping by age and smoking status. The recent Month Protected by copyright. and rapid rise in long term vaping followed a clear Fig 2 | Trends in long term vaping by frequency and main device type among adults in inverse age gradient that mirrored the pattern we England, October 2013 to October 2023. Lines represent modelled weighted prevalence have seen for the prevalence of current vaping using by monthly survey wave, modelled non-linearly using restricted cubic splines disposable devices over this period.7 8 The rise occurred (five knots). Shaded bands represent 95% confidence intervals. Points represent predominantly among current and former smokers, unmodelled weighted prevalence by month. Vertical dashed line indicates start of the rise in popularity of vaping using disposable devices in June 2021. *Not assessed but an increase in long term vaping also occurred before July 2016. Prevalence estimates by vaping frequency and main device type used among those who had never regularly smoked. This do not sum to the total prevalence of long term vaping owing to some missing data is not necessarily a cause for concern if vaping is (participants either did not respond or responded that they did not know) and each diverting people who would have otherwise smoked trend being modelled separately towards a less harmful nicotine product. Simulation modelling of trends in smoking and vaping among adolescents in the US suggests a substantial diversion October 2013, when e-cigarettes were still relatively effect is likely.26 However, if these people would not new and less effective in delivering nicotine than have otherwise taken up smoking, then taking up current devices, around one in 80 adults was a long vaping as a regular habit will expose them to greater term vaper. This number increased to one in 30 harm than if they had neither vaped nor smoked.3 With adults by mid-2017 and was stable for several years. a growing proportion of people using e-cigarettes, Then a rapid rise started in 2021, and by October it becomes increasingly unlikely that all would have 2023 one in 10 adults in England reported having used cigarettes instead, given the declining trends in been vaping for more than six months. The rise in prevalence of cigarette smoking.27 Across our study long term vaping was largely driven by an increase period, national estimates of smoking prevalence in long term daily vaping. The absolute increases among adults in England fell from 18.8% in 2013 to in long term vaping were most pronounced among 12.9% in 2022.27 Recent data indicate a rise in the people with a history of regular smoking, but an proportion of young adults using inhaled nicotine increase also occurred among people who had never (ie, vaping or smoking) since disposable e-cigarettes regularly smoked. Growth was also most pronounced became popular in the absence of an acceleration in in young adults, including among those who had the decline in smoking, although declines in smoking never regularly smoked. This pattern persisted after appear to have been most pronounced in age groups adjustment for rises in psychological distress over with the largest increase in vaping.27 28 Our current this period.25 Between October 2013 and March study shows that the proportion of young adults 2021, most long term vapers mainly or exclusively vaping long term is now higher than smoking rates used refillable e-cigarettes, and few used disposable have been in this age group since 2017.27 A similar devices. However, the prevalence of long term vaping pattern has been documented among adolescents, using disposable devices subsequently rose rapidly, with rates of current vaping among 11-17 year olds and by October 2023 similar proportions of adults in Great Britain in 2023 higher than smoking rates in mainly or exclusively used disposable and refillable this age group have been for more than a decade.5 the bmj | BMJ 2024;386:e079016 | doi: 10.1136/bmj-2023-079016 7 RESEARCH BMJ: first published as 10.1136/bmj-2023-079016 on 17 July 2024. Downloaded from http://www.bmj.com/ on 22 July 2024 at Univ of Alberta Library Bibliographic Service Unit-Serials. Table 2 | Modelled estimates of changes in prevalence of long term vaping and long term vaping using a disposable device within subgroups of adults in England Any long term vaping, % (95% CI)* Long term vaping using disposable device, % (95% CI)* October 2013 October 2023 July 2016 October 2023 Smoking status Never smoker 0.1 (0.0 to 0.2) 3.0 (2.3 to 3.8) 0.0 (0.0 to 0.3) 1.8 (1.2 to 2.5) Former smoker: Long term (≥1 year) 1.4 (1.0 to 1.9) 16.2 (14.2 to 18.4) 0.1 (0.0 to 0.5) 4.4 (3.1 to 6.0) Recent (6 months) vaping smoking cessation. Cochrane Database Syst Rev 2024;1:CD010216. increased substantially among adults in England from doi:10.1002/14651858.CD010216.pub8 the bmj | BMJ 2024;386:e079016 | doi: 10.1136/bmj-2023-079016 11 RESEARCH BMJ: first published as 10.1136/bmj-2023-079016 on 17 July 2024. Downloaded from http://www.bmj.com/ on 22 July 2024 at Univ of Alberta Library Bibliographic Service Unit-Serials. 3 McNeill A, Simonavicius E, Brose LS, et al. Nicotine vaping in 25 Jackson SE, Brown J, Shahab L, McNeill A, Munafò MR, Brose L. England: an evidence update including health risks and perceptions, Trends in Psychological Distress Among Adults in England, September 2022. A report commissioned by the Office for Health 2020-2022. JAMA Netw Open 2023;6:e2321959. doi:10.1001/ Improvement and Disparities. London: Office for Health Improvement jamanetworkopen.2023.21959 and Disparities 2022. https://www.gov.uk/government/publications/ 26 Selya AS, Foxon F. Trends in electronic cigarette use and conventional nicotine-vaping-in-england-2022-evidence-update (accessed 3 smoking: quantifying a possible ‘diversion’ effect among US October 2022). adolescents. Addiction 2021;116:1848-58. doi:10.1111/ 4 Buss V, Kock L, West R, et al. Trends in electronic cigarette use in add.15385 England. 2024. https://smokinginengland.info/graphs/e-cigarettes- 27 Office for National Statistics. Adult smoking habits in the UK: 2022. latest-trends. 2023. https://www.ons.gov.uk/peoplepopulationandcommunity/ 5 Action on Smoking and Health. Use of e-cigarettes (vapes) among healthandsocialcare/healthandlifeexpectancies/bulletins/ young people in Great Britain. 2023. https://ash.org.uk/resources/ adultsmokinghabitsingreatbritain/2022 (accessed 7 September view/use-of-e-cigarettes-among-young-people-in-great-britain 2023). (accessed 17 August 2023). 28 Tattan-Birch H, Brown J, Shahab L, et al. Trends in vaping and 6 Action on Smoking and Health. Use of e-cigarettes among adults in smoking following the rise of disposable e-cigarettes: a repeat cross- Great Britain. 2023. https://ash.org.uk/resources/view/use-of-e- sectional study in England between 2016 and 2023. Lancet Reg cigarettes-among-adults-in-great-britain-2021 (accessed 30 October Health Eur 2024; doi:10.1016/j.lanepe.2024.100924 2023). 29 Action on Smoking and Health. Policy options to tackle the issue of 7 Tattan-Birch H, Jackson SE, Kock L, Dockrell M, Brown J. Rapid disposable (singleuse) vapes. 2023. https://ash.org.uk/resources/ growth in disposable e-cigarette vaping among young adults in view/policy-options-to-tackle-the-issue-of-disposable-single-use- Great Britain from 2021 to 2022: a repeat cross-sectional survey. vapes. Addiction 2023;118:382-6. doi:10.1111/add.16044 30 Smith MJ, MacKintosh AM, Ford A, Hilton S. Youth’s engagement and 8 Jackson SE, Tattan-Birch H, Shahab L, et al. Who would be perceptions of disposable e-cigarettes: a UK focus group study. BMJ affected by a ban on disposable vapes? A population study in Open 2023;13:e068466. doi:10.1136/bmjopen-2022-068466 Great Britain.Public Health 2024;227:291-98. doi:10.1016/j. 31 Department of Health. Neil O’Brien MP. Crackdown on illegal sale of puhe.2023.12.024 vapes. 2023. https://www.gov.uk/government/news/crackdown-on- 9 Centers for Disease Control and Prevention. Youth Tobacco Use illegal-sale-of-vapes (accessed 14 September 2023). Surged From 2017-2018. CDC, 2019. https://www.cdc.gov/ 32 Hartmann-Boyce J, Lindson N, Butler AR, et al. Electronic cigarettes vitalsigns/youth-tobacco-use/index.html (accessed 18 December for smoking cessation. Cochrane Database Syst Rev2022, 2019). doi:10.1002/14651858.CD010216.pub7 10 US Food and Drug Administration. Results from 2018 National Youth 33 Jackson SE, Kotz D, West R, Brown J. Moderators of real-world Tobacco Survey show dramatic increase in e-cigarette use among effectiveness of smoking cessation aids: a population study. youth over past year. FDA, 2019. https://www.fda.gov/news-events/ Addiction 2019;114:1627-38. doi:10.1111/add.14656 press-announcements/results-2018-national-youth-tobacco-survey- 34 Department of Health and Social Care. Prime Minister’s Office, 10 show-dramatic-increase-e-cigarette-use-among-youth-over (accessed Downing Street, Department for Environment, Food & Rural Affairs, Protected by copyright. 6 February 2020). et al. Disposable vapes banned to protect children’s health. GOV.UK. 11 Amato MS, Boyle RG, Levy D. How to define e-cigarette 2024. https://www.gov.uk/government/news/disposable-vapes- prevalence? Finding clues in the use frequency distribution. banned-to-protect-childrens-health (accessed 26 February 2024). Tob Control 2016;25(e1):e24-9. doi:10.1136/ 35 Goldenson NI, Leventhal AM, Simpson KA, Barrington-Trimis JL. A tobaccocontrol-2015-052236 Review of the Use and Appeal of Flavored Electronic Cigarettes. Curr 12 Jackson SE, Brown J, Jarvis MJ. Dependence on nicotine in US high Addict Rep 2019;6:98-113. doi:10.1007/s40429-019-00244-4 school students in the context of changing patterns of tobacco 36 Gades MS, Alcheva A, Riegelman AL, Hatsukami DK. The Role product use. Addiction 2021;116:1859-70. doi:10.1111/ of Nicotine and Flavor in the Abuse Potential and Appeal of add.15403 Electronic Cigarettes for Adult Current and Former Cigarette and 13 Birdsey J, Cornelius M, Jamal A, et al. Tobacco Product Use Among Electronic Cigarette Users: A Systematic Review. Nicotine Tob U.S. Middle and High School Students - National Youth Tobacco Res 2022;24:1332-43. doi:10.1093/ntr/ntac073 Survey, 2023. MMWR Morb Mortal Wkly Rep 2023;72:1173-82. 37 Gravely S, Cummings KM, Hammond D, et al. The Association doi:10.15585/mmwr.mm7244a1 of E-cigarette Flavors With Satisfaction, Enjoyment, and Trying 14 Smith L, Sutherland N. The environmental impact of disposable to Quit or Stay Abstinent From Smoking Among Regular Adult vapes. 15 Dec 2022. London, House of Commons Library. Vapers From Canada and the United States: Findings From the 15 Pourchez J, Mercier C, Forest V. From smoking to vaping: a new 2018 ITC Four Country Smoking and Vaping Survey. Nicotine Tob environmental threat?Lancet Respir Med 2022;10:e63-4. Res 2020;22:1831-41. doi:10.1093/ntr/ntaa095 doi:10.1016/S2213-2600(22)00187-4 38 Friedman A, Liber AC, Crippen A, et al. E-cigarette Flavor Restrictions’ 16 Reid HT, Fordham A, Rasha L, et al. Up in smoke: Considerations for Effects on Tobacco Product Sales. SSRN 2024. doi:10.2139/ lithium-ion batteries in disposable e-cigarettes. Joule 2023;7:2749- ssrn.4586701 59. doi:10.1016/j.joule.2023.11.008 39 Zeffman H, Seddon P. Ministers consider new vaping tax at Budget 17 Fidler JA, Shahab L, West O, et al. ‘The smoking toolkit study’: a next week. BBC News. 2024 Feb 27. https://www.bbc.com/news/uk- national study of smoking and smoking cessation in England. BMC politics-68411690 (accessed 27 February 2024). Public Health 2011;11:479. doi:10.1186/1471-2458-11-479 40 Jackson SE, Tattan-Birch H, Shahab L, Brown J. How has Expenditure 18 Jackson SE, Beard E, Kujawski B, et al. Comparison of Trends in on Nicotine Products Changed in a Fast-Evolving Marketplace? A Self-reported Cigarette Consumption and Sales in England, 2011 Representative Population Survey in England, 2018-2022. Nicotine to 2018. JAMA Netw Open 2019;2:e1910161. doi:10.1001/ Tob Res 2023;25:1585-93. doi:10.1093/ntr/ntad074 jamanetworkopen.2019.10161 41 Jackson SE, Shahab L, Kock L, West R, Brown J. Expenditure on 19 Kock L, Shahab L, Moore G, et al. Protocol for expansion of an smoking and alternative nicotine delivery products: a population existing national monthly survey of smoking behaviour and alcohol survey in England. Addiction 2019;114:2026-36. doi:10.1111/ use in England to Scotland and Wales: The Smoking and Alcohol add.14709 Toolkit Study. Wellcome Open Res 2021;6:67. doi:10.12688/ 42 Friedman AS, Pesko MF. Young adult responses to taxes on cigarettes wellcomeopenres.16700.1 and electronic nicotine delivery systems. Addiction 2022;117:3121- 20 Kock L, Tattan-Birch H, Jackson S, et al. Socio-demographic, smoking 8. doi:10.1111/add.16002 and drinking characteristics in GB: A comparison of independent 43 Local Government Association. Councils call for ban of disposable telephone and face-to-face Smoking and Alcohol Toolkit surveys vapes. 2023. https://www.local.gov.uk/about/news/councils-call- conducted in March 2022. Qeios 2022;16. doi:10.32388/CLXK4D ban-disposable-vapes (accessed 14 September 2023). 21 Shahab L, Goniewicz ML, Blount BC, et al. Nicotine, Carcinogen, 44 Ferguson D. Ministers set to ban single-use vapes in UK over child and Toxin Exposure in Long-Term E-Cigarette and Nicotine addiction fears. Guardian. 2023 Sept 11. https://www.theguardian. Replacement Therapy Users: A Cross-sectional Study. Ann Intern com/society/2023/sep/11/ban-on-single-use-vapes-in-uk-could- Med 2017;166:390-400. doi:10.7326/M16-1107 be-imminent (accessed 14 September 2023). 22 Jackson SE, Beard E, Kujawski B, et al. Comparison of Trends in 45 Mahase E. Paediatricians call for ban on disposable e-cigarettes as Self-reported Cigarette Consumption and Sales in England, 2011 child vaping rises. BMJ 2023;381:1266. doi:10.1136/bmj.p1266 to 2018. JAMA Netw Open 2019;2:e1910161. doi:10.1001/ 46 Pratt SI, Ferron JC, Brunette MF, Santos M, Sargent J, Xie H. E-Cigarette jamanetworkopen.2019.10161 Provision to Promote Switching in Cigarette Smokers With Serious 23 Kotz D, West R. Explaining the social gradient in smoking cessation: Mental Illness-A Randomized Trial. Nicotine Tob Res 2022;24:1405- it’s not in the trying, but in the succeeding. Tob Control 2009;18:43- 12. doi:10.1093/ntr/ntac082 6. doi:10.1136/tc.2008.025981 47 Pratt SI, Sargent J, Daniels L, Santos MM, Brunette M. Appeal of 24 Stone CJ, Koo C-Y. Additive splines in statistics. Proc Am Stat electronic cigarettes in smokers with serious mental illness. Addict Assoc 1985;45:48. Behav 2016;59:30-4. doi:10.1016/j.addbeh.2016.03.009 12 doi: 10.1136/bmj-2023-079016 | BMJ 2024;386:e079016 | the bmj  RESEARCH BMJ: first published as 10.1136/bmj-2023-079016 on 17 July 2024. Downloaded from http://www.bmj.com/ on 22 July 2024 at Univ of Alberta Library Bibliographic Service Unit-Serials. 48 Hammond D, Reid JL. Trends in vaping and nicotine product use 52 Beard E, West R, Michie S, Brown J. Association of prevalence of among youth in Canada, England and the USA between 2017 and electronic cigarette use with smoking cessation and cigarette 2022: evidence to inform policy. Tob Control 2023;8:tc-2023- consumption in England: a time-series analysis between 2006 and 058241. doi:10.1136/tc-2023-058241 2017. Addiction 2020;115:961-74. doi:10.1111/add.14851 49 Jackson SE, Beard E, Angus C, Field M, Brown J. Moderators of 53 Jackson SE, Brown J, Beard E. Associations of prevalence of changes in smoking, drinking and quitting behaviour associated with e-cigarette use with quit attempts, quit success, use of smoking the first COVID-19 lockdown in England. Addiction 2022;117:772- cessation medication, and the overall quit rate among smokers in 83. doi:10.1111/add.15656 England: a time-series analysis of population trends 2007-2022. 50 Jackson SE, Garnett C, Shahab L, Oldham M, Brown J. Association of Nicotine Tob Res 2024; doi:10.1093/ntr/ntae007 the COVID-19 lockdown with smoking, drinking and attempts to quit 54 Jackson SE, Beard E, Brown J. Smokers’ Use of E-Cigarettes in in England: an analysis of 2019-20 data. Addiction 2021;116:1233- Situations Where Smoking Is not Permitted in England: Quarterly 44. doi:10.1111/add.15295 Trends 2011-2020 and Associations With Sociodemographic 51 Beard E, West R, Michie S, Brown J. Association between and Smoking Characteristics. Nicotine Tob Res 2021;23:1831-8. electronic cigarette use and changes in quit attempts, success of doi:10.1093/ntr/ntab119 quit attempts, use of smoking cessation pharmacotherapy, and use of stop smoking services in England: time series analysis of population trends. BMJ 2016;354:i4645. doi:10.1136/ Supplementary information: Items assessing vaping bmj.i4645 status, additional tables S1-S5 and figures S1-S5 Protected by copyright. the bmj | BMJ 2024;386:e079016 | doi: 10.1136/bmj-2023-079016 13