Rapid growth in disposable e‐cigarette vaping among young adults in Great Britain from 2021 to 2022: a repeat cross‐sectional survey

Abstract Aims To estimate recent trends in the prevalence of disposable e‐cigarette vaping in Great Britain, overall and across ages, and to measure these trends in the context of changes in smoking and vaping prevalence. Design The Smoking Toolkit Study, a monthly representative cross‐sectional survey. Setting Great Britain. Participants A total of 36 876 adults (≥ 18 years) completed telephone interviews between January 2021 and April 2022. Measurements Current e‐cigarette vapers were asked which type of device they mainly use. We estimated age‐specific monthly time trends in the prevalence of current disposable e‐cigarette use among vapers and inhaled nicotine use (vaping/smoking), smoking and vaping among adults. Findings From January 2021 to April 2022, there was an 18‐fold increase in the percentage of vapers who used disposables, rising from 1.2 to 22.2% [prevalence ratio (PR) = 18.0; 95% compatibility interval (CI) = 9.18–49.0]. Growth in disposable e‐cigarette vaping was most pronounced in younger adults (interaction P‐value = 0.013): for example, the percentage of 18‐year‐old vapers using disposables rose from 0.4 to 54.8% (PR = 129; 95% CI = 28.5–4520), while it rose from 2.1 to 10.0% (PR = 4.73; 95% CI = 2.06–23.6) among 45‐year‐old vapers. However, the overall percentage of people currently using any inhaled nicotine remained stable over time both among all adults (20.0 versus 21.2%; PR = 1.06; 95% CI = 0.92–1.22) and among 18‐year‐olds (30.2 versus 29.7%; PR = 0.99; 95% CI = 0.80–1.22). In 18‐year‐olds, vaping prevalence grew (11.3 versus 17.7%; PR = 1.57; 95% CI = 1.12–2.29), and there was imprecise evidence for a decline in smoking (24.5 versus 19.5%; PR = 0.80; 95% CI = 0.63–1.04). In 45‐year‐olds, there was relatively little change in vaping (PR = 1.08; 95% CI = 0.88–1.33) or smoking prevalence (PR = 1.01; 95% CI = 0.88–1.16). Conclusions Use of disposable e‐cigarettes in Great Britain grew rapidly between 2021 and 2022, especially among younger adults, but the overall prevalence of inhaled nicotine use was stable over time. Most young adult vapers in Great Britain now use disposable products.


INTRODUCTION
Early electronic cigarettes ('e-cigarettes') were disposable products that were poor at delivering nicotine. Over time, new e-cigarette types were developed to deliver nicotine contained in e-liquid more effectively through rechargeable devices with refillable tanks or replaceable pods (e.g. Juul) [1]. These devices came to dominate the global e-cigarette market and, by 2019, fewer than one in 10 vapers used disposables in England or the United States [1][2][3]. Recently, a new form of disposable e-cigarette has started being sold under brand names such as 'Puff bar', 'Elf bar' or 'Geek bar' [4]. Unlike earlier disposables, these products deliver nicotine effectively using a similar technology to pod devices, including high-strength (20 mg/ml in UK/EU) nicotine salts e-liquid [5]. They retail for approximately £5-7 (US$7-9) in the United Kingdom-about half the price of a pack of 20 cigarettes. US data show that, in 2021, disposables surpassed pods as the most commonly used type of e-cigarette among adolescents [2]. Little is known about the popularity of disposables in other countries and older age groups. It is also unclear whether these products attract people who would otherwise smoke cigarettes, vape other types of e-cigarettes or who would remain abstinent from nicotine entirely. This study aims to estimate recent trends in the prevalence of disposable e-cigarette vaping in Great Britain, overall and across ages, and to explore these trends in the context of other changes in smoking and vaping prevalence.

Design
The Smoking Toolkit Study (STS) is a monthly cross-sectional survey that recruits a nationally representative sample of adults (≥ 18 years) in Great Britain. It uses a hybrid of population and quota sampling.
Great Britain is divided into areas of approximately 300 households, which are stratified by region and demographic profile before being selected at random to be included on the interview list. In selected areas, interviews are performed with one individual per household until age, employment status and gender quotas are met. Raking is used to construct survey weights, adjusting data so that the demographic profile of the weighted sample matches that of Great Britain.
This demographic profile is ascertained monthly using data from three sources: the 2011 UK Census, the Office for National Statistics midyear estimates and the annual National Readership Survey. Methods are described in detail elsewhere [6]. Participants were told that this question referred to cigarettes and other kinds of tobacco, not e-cigarettes or heat-not-burn products.
Participants selecting (i) to (iii) were classified current smokers, (iv) and (v) former smokers and (vi) never smokers.
Vaping status was assessed by asking participants whether they were currently using e-cigarettes to cut down on the amount they smoke, in situations when they are not allowed to smoke, to help them stop smoking or for any other reason. Those who responded positively to any of these questions were considered current vapers.
Current vapers were asked which type of device they mainly use.
Those who responded, 'a disposable e-cigarette or vaping device (non-rechargeable)' were considered disposable e-cigarette vapers.
They could only choose one device type (the one they 'mainly' use).
Participants were asked to provide their exact age in years. Those who refused to give their exact age were asked to select their age group from a list. For participants who only responded to the latter question (2% of respondents), exact age was imputed as the mean age within the age group they selected. Participants were also asked for their gender.

Analysis
Weighted logistic regression was used to estimate monthly time trends in the proportion of (i) adults and (ii) current vapers who use disposable e-cigarettes, overall and for specific ages (using survey weights described earlier). For the overall analysis, models only included predictors for time. For the age-specific analysis, models included time, age and their interaction as predictors-thus allowing for time trends to differ across ages. Both age and time were modelled continuously using restricted cubic splines with three knots (placed at earliest, middle and latest month for time and 5, 50 and 95% quantiles for age among vapers). This allowed the relationship of prevalence with age and time to be flexible and non-linear, while avoiding categorization [7]. Age was modelled continuously, so we displayed estimates for four specific ages (18-, 25-, 35-and 45-yearolds) to illustrate how trends differed across ages. Note that the model used to derive these estimates included data from participants of all ages, not only those who were aged exactly 18, 25, 35 or 45 years.
Prevalence ratios (PR) for the change in prevalence across the whole time-series (April 2022 versus January 2021) were presented, alongside 95% compatibility intervals (95% CIs) calculated using bootstrapping [8][9][10][11]. We ran analogous analyses to estimate time trends in the proportion of adults who currently (i) vape, (ii) smoke or (iii) use any form of inhaled nicotine-whether smoked or vaped.
Note that prevalence of disposable e-cigarette use was very low in older age groups, which meant that we were unable to estimate time trends in these groups. Finally, we reported the percentage of disposable e-cigarette vapers who reported being current, former or never smokers. Participants with missing data for their smoking or vaping status (< 1%) were excluded from analyses that required this information. R version 4.1.0 was used for analyses (code: https:// osf.io/km3x6/). Despite this, the overall percentage of adults currently using any inhaled nicotine (smoked or vaped) was relatively stable during the study period ( More detailed monthly trends in the prevalence of inhaled nicotine use, vaping and smoking among adults of different ages are shown in Supporting information, Figs S1-S3.

RESULTS
Most disposable e-cigarette vapers were current (71.6%) or former smokers (18.8%), with few reporting never having smoked regularly (9.6%). The proportion of disposable vapers who also smoked was similar across ages, but it may have declined slightly over time (Supporting information, Figs S4 and S5).
T A B L E 1 Age-specific trends in current vaping, smoking and disposable e-cigarette vaping prevalence in Great Britain.