Joint smoking–vaping prevalence rates among American youth and young adults 2011–2022

Abstract

The Center for Disease Control and Prevention (CDC) annually tracks American youth and adult smoking prevalence using data from the National Youth Tobacco Survey (NYTS) and the National Health Interview Survey (NHIS). The NYTS and the NHIS began collecting information on vaping in 2011 and 2014 respectively. However, since those years the CDC has only reported smoking and vaping rates separately, which presents a long-term and important information gap, given the decade-long debate about whether e-cigarettes help people who smoke reduce or quit, and whether they are a gateway to youth smoking. This short report provides joint smoking and vaping prevalence rates for American high school students from 2011 to 2022 from the NYTS, and rates for American adults 18–44 years old over the same period from the NHIS. The results show that cigarette smoking declined sharply, especially in high schoolers and emerging young adults (18–20 years) while vaping increased substantially. In addition, the prevalence of vaping among people who formerly smoked also increased. The importance of these trends is highlighted by sharp increases in quit ratios among emerging and junior (21–24 years) young adults, who had historically low levels of quitting.

Introduction

Cigarette smoking has been the leading cause of preventable morbidity and mortality in the US for decades [1], which is why the Center for Disease Control and Prevention (CDC) annually tracks American youth and adult smoking prevalence using data from the National Youth Tobacco Survey (NYTS) [2] and the National Health Interview Survey (NHIS) [3] respectively. The NYTS and the NHIS began collecting information on vaping in 2011 and 2014 respectively. However, since those years the CDC has only reported smoking and vaping rates separately, which presents a long-term and important information gap, given the decade-long debate about whether e-cigarettes help people who smoke reduce or quit, and whether they are a gateway to youth smoking. The practice also ignores contemporaneous dual use, resulting in erroneous overestimates of nicotine use in these populations [4, 5].

This short report provides joint smoking and vaping prevalence rates for American high school students from 2011 to 2022 from the NYTS, and rates for American adults 18–44 years old over the same period from the NHIS.

Methods

NYTS

The NYTS was designed to provide national information on middle and high school youth tobacco use and on participants’ tobacco-related beliefs, attitudes, behaviors and exposure to tobacco influences. Public use data are available for selected years from 1999 to 2009, and for all years from 2011. Public use files were downloaded directly from the CDC website for years 2011–2022 [6], when both smoking and vaping information was collected.

In all years, participants were asked on how many days of the past thirty that they smoked cigarettes (answers 2011–2018: 0, 1–2, 3–5, 6–9, 10–19, 20–29, 30; 2019–2022: 0–30). A separate question in 2011 asked if they used e-cigarettes as one of nine other tobacco products on at least one or more of the past thirty days. In 2014 a vaping section was established, which expanded to six questions matching other products; this included a question about days used in the past 30 (same answers and years as cigarette smoking, 2014–2022). The number of vaping questions doubled in 2015 and remained the same for the next five years. In 2020 the section increased to sixteen questions, and it replaced cigarettes as the questionnaire’s lead tobacco product. Before 2021 the NYTS was administered in person at participating schools. However, that year, due to the COVID-19 epidemic, the survey was administered entirely online; 51% of students completed it in school and 49% at home or some other place [7].

Following the CDC, we defined current smoking or vaping as using the respective products at least one day in the past 30 [2]. We limited our analysis to high school participants who had information on smoking and vaping, and we applied weighting variables for each year.

NHIS

The NHIS is a multipurpose health survey of the non-institutionalized adult (18 + years) population of the US, conducted since 1957 and administered by the CDC. The design oversamples Black and Hispanic populations to allow for more precise estimates in these groups, and it employs a complex design involving stratification, clustering, multistage sampling and weighting. The survey has used computer-assisted personal interviewing since 1997.

NHIS adult sample public use data files for each year from 2011 to 2022 were obtained from the CDC [8].

We used CDC criteria to establish the smoking status of participants [3]. NHIS asked ‘Have you smoked at least 100 cigarettes in your entire life?’ Subjects answering ‘no’ were classified as people who never smoked. Those who answered ‘yes’ were classified as people who ever smoked and were asked: ‘Do you now smoke cigarettes every day, some days or not at all?’ Respondents answering ‘every day’ or ‘some days’ were classified as people who currently smoke; those responding ‘not at all’ were classified as people who formerly smoked.

Starting in 2014, participants were asked if they ever used e-cigarettes. Those answering yes were asked if they now use e-cigarettes every day, some days or not at all. Respondents who answered ‘every day’ or ‘some days’ were classified as people who currently use e-cigarettes. Because no question documented established use, the definition of “former” vaping cannot be established in this survey.

We present smoking prevalence, vaping prevalence and combined smoking–vaping prevalence in the following groups: high school students in NYTS, adult subjects in NHIS age 18–20 years (hereafter emerging young adults), 21–24 years (hereafter junior young adults), 25–34 years (senior young adults) and 35–44 years (mid-age adults).

Quit ratios, the prevalence of ex-smoking divided by prevalence of ever smoking X 100 [9], were calculated for all adult age groups from 2000 to 2022.

Statistical Package for the Social Sciences Version 29.0 (SPSS Inc., Chicago, IL, USA) with complex samples was used to provide prevalence estimates and their 95% confidence intervals (CI), after adjustment for non-response. This procedure was validated by matching smoking prevalence rates reported by the CDC over the study period.

Results

Figure 1 shows that smoking prevalence among high school students using NYTS data was near 16% in 2011, and thereafter declined substantially. Prevalence of dual use was three percent or less before 2014, then it remained around four to five percent until 2020, afterwards dropping to 1.5% by 2022. Prevalence of exclusive vaping in 2011 was initially less than one percent, increased substantially to 7–10% between 2014 and 2017, then surged to 22% by 2019, followed by a reduction to 10% and 12.6% in 2021 and 2022. All patterns in this group showed substantial instability from year to year.

Fig. 1

Prevalence of Smoking and Vaping in the Past 30 Days Among American High School students, NYTS 2011–2022. *See notes on survey methodology changes in the discussion

Using NHIS data, smoking among emerging young adults is shown in Fig. 2. It started at the same level as high school students in 2011, but it did not begin to steadily decline until 2015. By 2022 prevalence of smoking in this group was the same as that in high school students. Prevalence of dual use in 2014 was 3.7%, but this was the highest of the observed period, falling to 0.8% by 2022. The prevalence of vaping/former smoking was 2% or less for all years, but exclusive vaping gradually increased in the single digits, peaking at 9.4% in 2022. In addition, the prevalence of vaping was lower than among high school students and more stable.

Fig. 2

Prevalence of Smoking and Vaping Among American Emerging Young Adults (18–20 years), NHIS 2011–2022

Smoking among junior young adults, shown in Fig. 3, was 21–22% until 2014, when it started to slowly decline compared to high schoolers and emerging young adults. Smoking prevalence remained near 10% until 2021 and fell to about 7% during the last two years. Dual use was relatively stable around 2–3.5%. Prevalence of vaping/former smoking was less than 1% until 2018, since then it had increased every year to 3.8% in 2022. Similarly, exclusive vaping remained less than 2% until 2018, but then increased sharply to 10% in 2022.

Fig. 3

Prevalence of Smoking and Vaping Among American Junior Young Adults (21–24 years), NHIS 2011–2022

Smoking and vaping among senior s and mid-age adults is seen in Figs. 4 and 5. Smoking prevalence declined substantially in both age groups with slightly higher prevalence among mid-age adults in 2022 (14% vs. 11%). For most years, dual use was relatively stable in both groups at 2–3%. Vaping/former smoking and exclusive vaping increased in both groups but were lower than in high schoolers and young adults.

Fig. 4

Prevalence of Smoking and Vaping Among American Senior Young Adults (25–34 years), NHIS 2011–2022

Fig. 5

Prevalence of Smoking and Vaping Among American Mid-Age Adults (35–44 years), NHIS 2011–2022

Quit ratios for all adult age groups from 2011 to 2022 are seen in Fig. 6. Emerging young adults had the lowest quit ratio (15%) in 2011, followed by each older age group (23%, 38% and 44% respectively). The quit ratio of emerging young adults was unstable but rose sharply after 2015, ending at 52% in 2022. Junior young adults also had sharp increases after 2016, reaching 50% in 2022. The two older groups, which had higher quit ratios in 2011, had more modest but consistent increases during the period, reaching 60%.

Fig. 6

Quit Ratios (%) Among Americans Who Smoked (Former/Ever) According to Age Group, NHIS 2011–2022

Discussion

There has been a significant information void in the annual CDC reports of youth and adult smoking and vaping since 2011 and 2014 respectively. Because the CDC only reports the behaviors separately, as many as 5.7% (in 2018) of high schoolers and 3.7% (in 2014) of emerging young adults were in effect double-counted by the agency because they were using both products. Furthermore, CDC reports do not classify current vaping according to smoking status, which is especially important for understanding the proportion of current vaping concurrent with former smoking. In this paper we extend the CDC prevalence rates by reporting youth and adult dual use and adult vaping former smoking, thus resolving the information void. We limit our interpretation to general insights that are apparent from the figures.

In 2011 smoking prevalence was 15.7% among high schoolers and 14.7% among emerging young adults, and over the next 11 years declined the most (87 and 86% respectively), compared with the other age groups. In the three older age groups smoking declined less with increasing age (junior young adults, 68%; senior young adults, 54%; mid-age adults, 30%).

Smoking declines were coexistent with increases in vaping, although prevalence of the latter was particularly unstable in high schoolers, with sharp increases in three years (2015, 2018, 2019) but steep declines in three other years (2016, 2020, 2021). Another important point is that, compared with the 2011 baseline, combined smoking–vaping prevalence was only higher in five of the eleven years (2014–15, 2018–20). The public health implications of our findings have been discussed by Delnevo and Villanti, who studied smoking prevalence among high schoolers from 1991 to 2022 using three U.S. national youth surveys [10]. They concluded that “the emergence of e-cigarettes has coincided with the most rapid declines in cigarette use over the past thirty years.”

The CDC has acknowledged that the establishment of a separate vaping section in the 2014 NYTS might have played a role in the vaping rate increase among high school students that year [11]. Less is known about the effect of the increase in the doubling vaping questions from six to twelve in 2015, or the increase to sixteen, along with becoming the lead tobacco product, in the 2020 survey. However, the effect of one event, the COVID-19 epidemic, is clear. It closed many schools, prompting the NYTS in 2021 to mix at-home surveys with those administered entirely in schools in previous years [12]. This change in survey setting was associated with lower vaping prevalence among at-home participants [13]. Fortunately, the NYTS returned to school-administered in 2022.

The declining smoking trend in high schoolers was also seen in emerging young adults, who were post high school and partially or fully independent with respect to purchase and use of tobacco products, at least until 2019. Vaping wasn’t as high or as volatile in this group, but the highest prevalence was in 2022. Again, with 2011 as a baseline, nicotine prevalence was only higher in one year (2014) of the eleven. E-cigarettes are technology products, which may partially explain why the youngest age groups in this study, high schoolers and emerging young adults, have had the highest prevalence of their use in this study.

Junior young adults had a much higher smoking rate in 2011 (22%), and it declined much less than high schoolers and emerging young adults. However, by 2022 smoking prevalence was 7%. Vaping also consistently increased, especially from 2020, with increases in dual use, former smoking and exclusive vaping. Smoking and vaping trends followed similar patterns in the next two ten-year age groups, with smoking declines accompanied by vaping increases. With 2011 as a baseline, it is remarkable that all three of these age groups saw lower nicotine prevalence rates for all subsequent years.

In 2011 emerging young adults had the lowest quit ratio (15%) with higher rates with consecutively older age groups, which is consistent with other studies [14, 15]. Although quit ratios in all groups increased steadily, there was a sharp increase starting in 2015, especially in emerging young adults; the quit ratio more than doubled from 21% that year to 52% in 2022. It is important to note that the high quit ratio in 2022 among emerging young adults is contemporary with 2% smoking prevalence in both that group and in high schoolers. If quit ratios continue to increase and current smoking continues to decline, it may result in little to no smoking-attributable mortality in these birth cohorts. The quit ratios of other age groups in 2022, 50% in junior young adults, 59% in senior young adults and 61% in mid-age adults, portends a substantial reduction in smoking-attributable consequences.

There are several limitations. Responses from NYTS and NHIS surveys were self-reported and were not verified by bio testing. Large changes in the NYTS questionnaires related to vaping, discussed previously, as well as the COVID-19 pandemic, may have had effects on vaping rates. Similarly, according to the CDC, the 2020 and 2021 NHIS surveys were telephone-based rather than in-person, which might have affected prevalence rates [16]. Finally, vaping products changed rapidly during the study period, and questionnaire updates may not have captured all new styles, which may have led to underreporting, especially among youths.

Two additional matters deserve comments. As noted in the methods, the NHIS lacks questions for established vaping. Absent this threshold for regular use, vaping prevalence in NHIS surveys may be overestimated, although there is no practical way to provide any quantitative estimate. In addition, it is important to emphasize that our designation of current vaping and former smoking does not imply that the former caused the latter. It is, however, a subcategory of vaping that contains important information and does not warrant omission.

In conclusion, we report for the first time joint smoking–vaping prevalence from the two most important CDC-sponsored national survey series. The results show that cigarette smoking declined sharply, especially in high schoolers and emerging young adults while vaping increased substantially. In addition, the prevalence of vaping among former smoking also increased. The importance of these trends is highlighted by sharp increases in quit ratios among emerging and junior young adults, who had historically low levels of quitting.