Abstract
Background
Evidence of US adult flavored e-cigarette use prevalence stratified by age, smoking status, and purpose for vaping (ie, quitting smoking, to use when or where smoking is not allowed) can inform policies that reduce the tobacco-related cancer burden.
Methods
Current flavored e-cigarette use (use 1 or more nontobacco flavors) prevalence estimates were compared across subpopulation groups using 2-sided statistical significance tests in the July 2018 Current Population Survey Tobacco Use Supplement, a nationally representative cross-sectional adult survey (n = 46 759).
Results
Current flavored e-cigarette use was reported by 1.6% (95% confidence interval [CI] = 1.47% to 1.69%) of all respondents. Among current vapers, the percentage of those who used flavored e-cigarettes was higher for adults aged 18-24 years (89.6%), 25-34 years (86.7%), and 35-44 years (76.0%) than for adults aged 45 years and older (60.4%, Ps < .001); was higher in never smokers (89.8%) than current (72.9%), long-term former (73.9%), and recent former (80.4%) smokers (Ps ≤ .009); was higher in smokers who reportedly did (78.9%) vs did not (71.1%) use e-cigarettes to vape where or when smoking is not allowed (P = .005); and did not differ between smokers who reportedly did (75.0%) vs did not (73.9%) vape to quit smoking (P = .71). Individuals who vaped to quit smoking and currently used flavored e-cigarettes constituted 0.9% (95% CI = 0.82% to 0.99%) of all adults (weighted N = 2 251 000, 95% CI = 2 046 000 to 2 476 000) and 57.2% of current flavored e-cigarette users.
Conclusions
Flavored e-cigarette use prevalence was low among US adults overall but common for current vapers. Flavored e-cigarette use was disproportionately prevalent among never smokers and other subpopulations that might experience harm from vaping.
E-cigarette use should be avoided by tobacco nonusers and other populations liable to be harmed by vaping, including youth (1,2). By contrast, adult combustible cigarette smokers could potentially reduce harm by quitting smoking and using e-cigarettes instead (1). E-cigarette regulatory policies should consider implications across distinct subpopulations separated by their potential for vaping-related harm or harm reduction.
Flavors are principal targets of tobacco product regulation. In February 2020, the US Food and Drug Administration (FDA) announced enforcement actions would be taken against unauthorized cartridge-based e-cigarette products in flavors other than tobacco and menthol (3). The FDA has not announced intentions to implement comprehensive sales restrictions on all flavored e-cigarette products. Instead, manufacturers’ marketing applications of individual products will be considered on a case-by-case basis (3). Premarket review decisions and other future regulatory actions that limit the availability of e-cigarettes in nontobacco flavors could reduce adolescent and young adult vaping (4). However, the net population health impact of such regulations could be unfavorable in 2 circumstances. First, many US adults use flavored e-cigarettes to quit smoking. Second, there are few US adult users of flavored e-cigarettes who are never smokers or are vulnerable to vaping’s adverse effects for other reasons.
Previous research demonstrates that flavored e-cigarette use is more common in never vs ever smokers and younger vs older adults (5-8), yet evidence gaps remain. Age and smoking status alone do not entirely characterize all subpopulations with different potentials to experience vaping-related harm or harm reduction. Although young adults are generally considered a population vulnerable to vaping’s adverse effects (2), some young adults use e-cigarettes to quit smoking (9). Furthermore, some smokers do not use e-cigarettes to quit smoking but rather as means to self-administer nicotine when or where smoking combustible cigarettes is not allowed (10). This type of dual product use might cause harm by increasing exposure to toxins in both cigarette smoke and e-cigarette aerosol (11,12). Additionally, existing flavored e-cigarette use prevalence studies only report percentage estimates within the current vaper population (5-8). Extrapolating the public health importance of such estimates to the entire population is difficult because current vapers constituted 3.7% or less of US adults from 2014 to 2018 (13).
This study estimated the prevalence of current flavored e-cigarette use stratified by subpopulation variables that differentiate potential for vaping-related harm or harm reduction. We report estimates stratified by age, smoking status, and 2 reported reasons for vaping: to quit smoking and to vape where or when smoking is not allowed.
Methods
Participants and Procedures
The US Census Bureau’s Current Population Survey (CPS) interviews US residents every month to obtain a nationally representative probability sample of households using a stratified 2-stage sampling scheme (14). Every 3-4 years, civilian noninstitutionalized CPS respondents aged 18 years and older are randomly selected to receive the Tobacco Use Supplement (TUS-CPS) questions. In July 2018, 84.5% of CPS-eligible households participated in the CPS, and 57.7% of TUS-eligible CPS participants provided self-response data (n = 46 579) (14). To reduce reporting error, proxy surveys were excluded. The Supplementary Methods provide further methodological details about the TUS-CPS. Depicted in Supplementary Table 1 (available online), female sex, older age, higher education, and lower income respondents constituted a higher percentage of CPS-TUS self-responders than proxy respondents or nonrespondents. Given use of deidentified data, this study was exempt by the University of Nebraska institutional review board.
Measures
E-cigarette use and flavor used were assessed in separate questions. Participants were asked, “Do you use e-cigarettes now, on some days, every day, or not at all?” Those who reported using e-cigarettes now, on some days, or every day were classified as current e-cigarette users. Current e-cigarette users were asked whether they usually used flavored e-cigarettes and to indicate which of 4 flavor categories they used (select all that apply: “Tobacco,” “Menthol or mint,” “Fruit, candy, sweets, chocolate, clove, spice, herb, or alcohol,” “Other”). Respondents indicating not using flavored e-cigarettes were further queried whether they usually use tobacco-flavored e-cigarettes. Responses were coded into a flavored e-cigarette use binary variable (yes [use ≥1 nontobacco flavors] vs no or unflavored [use only tobacco-flavored or flavorless e-cigarettes]) because US local and state e-cigarette flavor sales restrictions predominately exempt only tobacco-flavored or flavorless e-cigarettes (15).
Four subpopulation stratification variables were assessed. Self-reported age was a 4-level variable (18-24, 25-34, 35-44, and ≥ 45 years). Smoking status was a 4-level variable (never smoked [<100 cigarettes lifetime], recent former [≥100 cigarettes; now smokes “not at all” and last smoked ≤1 year ago], long-term former [≥100 cigarettes; now smokes “not at all” and last smoked >1 year ago], and current [≥100 cigarettes; now smokes]) (16). Ever smokers were asked whether they used e-cigarettes “because you can use them at times when or in places where smoking cigarettes isn’t allowed” (yes or no) and “to help you quit smoking cigarettes” (yes or no).
Sociodemographics and tobacco product use characteristics were assessed to describe the sample and model as covariates. These measures included assessments of self-reported sex, race or ethnicity, education, income, vaping duration, daily vs nondaily vaping, and current use of 1 or more other (noncigarette or non–e-cigarette) tobacco products. Table 1 lists response options.
Table 1.
Sample characteristics stratified by e-cigarette use status and flavor useda
| Variable | All TUS-CPS self-respondents |
Current e-cigarette users |
||||
|---|---|---|---|---|---|---|
| Not current e-cigarette users, % (95% CI) (n = 45 884) |
Current e-cigarette users, % (95% CI)b (n = 875) |
P c | Use unflavored e-cigarettes only, % (95% CI)d (n = 233) |
Use flavored e-cigarettes, % (95% CI)e (n = 642) |
P c | |
| Sociodemographics | ||||||
| Female | 52.0 (51.5 to 52.4) | 42.3 (39.0 to 45.6) | <.001 | 37.1 (31.2 to 43.5) | 43.8 (40.0 to 47.6) | .04 |
| Age group, y | ||||||
| >45 | 54.4 (53.9 to 54.9) | 32.8 (29.8 to 36.0) | <.001 | 57.2 (50.8 to 63.4) | 25.6 (22.4 to 29.1) | <.001 |
| 35-44 | 16.3 (16.0 to 16.6) | 14.6 (12.4 to 17.1) | 15.4 (11.4 to 20.6) | 14.4 (11.9 to 17.3) | ||
| 25-34 | 17.7 (17.4 to 18.1) | 25.4 (22.6 to 28.4) | 14.9 (10.9 to 20.0) | 28.5 (25.1 to 32.1) | ||
| 18-24 | 11.6 (11.3 to 11.9) | 27.2 (24.3 to 30.3) | 12.5 (8.4 to 18.3) | 31.5 (27.9 to 35.3) | ||
| Race or ethnicity | ||||||
| Non-Hispanic White | 62.9 (62.5 to 63.4) | 73.5 (70.4 to 76.3) | <.001 | 76.7 (70.9 to 81.7) | 72.5 (68.9 to 75.9) | .52 |
| Non-Hispanic Black | 12.0 (11.7 to 12.3) | 6.2 (4.8 to 8.0) | 6.1 (3.7 to 9.9) | 6.3 (4.6 to 8.4) | ||
| Hispanic | 16.5 (16.2 to 16.9) | 12.1 (10.1 to 14.5) | 9.6 (6.5 to 14.1) | 12.9 (10.4 to 15.8) | ||
| Otherf | 8.5 (8.3 to 8.8) | 8.2 (6.5 to 10.2) | 7.6 (4.9 to 11.8) | 8.3 (6.4 to 10.7) | ||
| Education | ||||||
| Less than high school | 9.0 (8.7 to 9.2) | 8.4 (6.7 to 10.4) | <.001 | 10.1 (6.8 to 14.6) | 7.9 (6.1 to 10.3) | .45 |
| High school graduate | 27.0 (26.6 to 27.4) | 35.6 (32.5 to 38.8) | 34.6 (28.8 to 40.9) | 35.8 (32.2 to 39.6) | ||
| Some college | 28.8 (28.4 to 29.2) | 38.6 (35.5 to 41.9) | 36.3 (30.4 to 42.6) | 39.3 (35.6 to 43.2) | ||
| College graduate | 35.2 (34.8 to 35.7) | 17.4 (15.0 to 20.0) | 19.0 (14.5 to 24.6) | 16.9 (14.2 to 20.0) | ||
| Income, $ | ||||||
| <20 000 | 13.4 (13.1 to 13.7) | 17.5 (15.1 to 20.1) | <.001 | 19.4 (14.8 to 24.9) | 16.9 (14.2 to 20.0) | .81 |
| 20 000-49 999 | 28.6 (28.2 to 29.1) | 30.3 (27.4 to 33.5) | 29.3 (23.8 to 35.4) | 30.6 (27.2 to 34.3) | ||
| 50 000-74 999 | 18.9 (18.5 to 19.3) | 21.7 (19.1 to 24.6) | 21.0 (16.3 to 26.7) | 21.9 (18.9 to 25.3) | ||
| 75 000-99 999 | 12.4 (12.1 to 12.7) | 12.9 (10.8 to 15.3) | 13.9 (10.1 to 19.0) | 12.6 (10.2 to 15.4) | ||
| 100 000+ | 26.6 (26.2 to 27.1) | 17.6 (15.2 to 20.3) | 16.4 (12.2 to 21.7) | 18.0 (15.2 to 21.1) | ||
| Tobacco product use | ||||||
| Cigarette smoking | ||||||
| Current | 11.4 (11.2 to 11.7) | 42.7 (39.3 to 46.2) | <.001 | 51.1 (44.7 to 57.5) | 40.3 (36.3 to 44.3) | <.001 |
| Former (>1 y)g | 16.4 (16.1 to 16.8) | 26.8 (24.0 to 29.9) | 30.9 (25.3 to 37.2) | 25.6 (22.4 to 29.1) | ||
| Former (≤1 y)g | 1.1 (1.0 to 1.2) | 10.1 (8.3 to 12.3) | 8.8 (5.8 to 13.1) | 10.5 (8.4 to 13.1) | ||
| Never | 71.0 (70.6 to 71.5) | 20.4 (17.6 to 23.5) | 9.2 (6.1 to 13.6) | 23.6 (20.2 to 27.4) | ||
| Vaped for >1 y | 0.6 (0.5 to 0.6) | 55.5 (52.2 to 58.8) | <.001 | 52.6 (46.2 to 58.9) | 56.4 (52.5 to 60.2) | .59 |
| Daily vaping | NA | 46.2 (42.9 to 49.5) | NA | 33.5 (27.8 to 39.8) | 49.9 (46.0 to 53.7) | <.001 |
| Currenty use other tobacco producth | 3.8 (3.7 to 4.0) | 15.6 (13.4 to 18.2) | <.001 | 14.3 (10.3 to 19.3) | 16.0 (13.4 to 19.0) | .47 |
Weighted percentage (95% CI) within column unless otherwise specified. CI = confidence interval; NA = not applicable; TUS-CPS = Tobacco Use Supplement of the Current Population Survey.
Excludes 13 respondents without e-cigarette flavors used data.
From χ2 test.
Used only tobacco-flavored or flavorless e-cigarettes (see Table 2 for prevalence of tobacco and flavorless product use within group).
Used 1 or more nontobacco flavors (see Table 2 for prevalence of use of each flavor).
Includes Asian American, American Indian or Alaskan Native, Hawaiian or Pacific Islander, multiracial or multiethnic, or other race or ethnicity.
One year or less: quit smoking within the year preceding the survey. More than 1 year: quit smoking more than 1 year before the survey.
Use of regular cigar, cigarillo, little filtered cigar, regular pipe filled with tobacco, water pipe or hookah pipe filled with tobacco, smokeless tobacco, or dissolvable tobacco.
Statistical Analysis
After descriptive analyses, primary analyses involved calculating flavored e-cigarette use weighted percentages within the entire sample and among the subsample of current e-cigarette users. We also estimated US population counts (weighted Ns) of current flavored and unflavored e-cigarette use. Estimates were reported both nonstratified and stratified by age, smoking status, use of e-cigarettes to quit smoking, and e-cigarette use when or where smoking is not allowed. Then, a multivariable logistic regression model was tested that included age, smoking status, and all sociodemographics and tobacco product use covariates listed in Table 1 as simultaneous regressors. For the 2 reasons for vaping variables, associations with flavored e-cigarette use were tested in separate covariate-adjusted regression models. Unadjusted logistic regressions were also reported. To explore interactive associations, we examined age × smoking status and reason for vaping × smoking status omnibus interaction terms and report stratified estimates of flavored e-cigarette use within each combination of the interacting variables. For variables with 3 groups or more, regression models were reassessed with pairwise χ2 comparisons. Estimated weighted Ns of adult flavored e-cigarette users with potential for harm reduction included ever smokers that reportedly vaped to quit smoking. Estimated weighted Ns of adult flavored e-cigarette users without potential for vaping-related harm reduction were calculated as the sum of 2 groups. This estimate included the number of never smokers plus the number of smokers who reportedly did not vape to quit smoking. With the exception of vaping duration (N missing = 32) for which missingness was addressed using a missing indicator approach, listwise deletion was applied to missing data for other regressors (N missing range = 1-5). Results were weighted to the US adult population of 249 531 000. Analyses were performed with SAS 9.4 using replicate weights and the balanced repeated replication method to address the complex survey design (14). Statistical significance was .05 (2-sided tests) with a Benjamini-Hochberg multiple test correction to control study-wise false discovery rate to .05 (17).
Results
Study Sample
Among the total sample of TUS-CPS self-respondents (n = 46 759), 888 (2.1%, 95% confidence interval [CI] = 1.9% to 2.2%) reported current e-cigarette use, of whom 875 provided data on e-cigarette flavors used (see Table 1 demographics). Among all TUS-CPS self-respondents, current unflavored and flavored e-cigarette use prevalence estimates were 0.5% (95% CI = 0.41% to 0.53%) and 1.6% (95% CI = 1.47% to 1.69%), respectively, reflecting an estimated 1 156 000 (95% CI = 1 013 000 to 1 321 000) and 3 933 000 (95% CI = 3 661 000 to 4 225 000) US adults. Within current vapers, 77.3% (95% CI = 74.4% to 79.9%) reportedly used flavored e-cigarettes. “Fruit, candy, sweets, chocolate, clove, spice, herb, or alcohol” was the most common nontobacco flavor class used, followed by “Menthol or mint” and “Other” (Table 2).
Table 2.
Prevalence of specific e-cigarette flavors used in current e-cigarette usersa
| Flavor | % (95% CI) |
|---|---|
| All current e-cigarette users (n = 875) | |
| Flavorless only | 5.2 (3.9 to 6.9) |
| Tobaccob | 29.9 (26.9 to 33.0) |
| Menthol or mintb | 22.7 (20.0 to 25.6) |
| Fruit, candy, sweets, chocolate, clove, spice, herb, or alcoholb | 61.3 (58.0 to 64.4) |
| Other nontobacco flavorb | 10.0 (8.2 to 12.3) |
| Flavored e-cigarette use status | |
| Tobacco-flavored or flavorless e-cigarettes only (n = 233) | 22.7 (20.1 to 25.6) |
| Flavorless onlyc | 5.2 (3.9 to 6.9) |
| Tobacco onlyc | 17.5 (15.1 to 20.2) |
| >1 nontobacco flavors (n = 642) | 77.3 (74.4 to 79.9) |
| Menthol or mint onlyc | 11.3 (9.3 to 13.5) |
| Fruit, candy, sweets, chocolate, clove, spice, herb, or alcohol onlyc | 46.4 (43.1 to 49.7) |
| Other nontobacco flavor onlyc | 3.4 (2.3 to 4.9) |
| ≥2 flavorsc | 16.2 (13.9 to 18.8) |
Weighted percentage (95% CI) of all current e-cigarette users in study sample. CI = confidence interval.
Not mutually exclusive from one another; sums total to more than 100%.
Mutually exclusive from one another; sums total to 100%.
Flavored e-Cigarette Use Prevalence Stratified by Subpopulation Groups Among the Subsample of Current e-Cigarette Users
The unadjusted percentage and adjusted odds of using flavored e-cigarettes was statistically significantly higher among those aged 18-24 years (89.6%, adjusted odds ratio [AOR] = 6.42, 95% CI = 3.59 to 11.50), 25-34 years (86.7%, AOR = 4.80, 95% CI = 3.21 to 7.18), and 35-44 years (76.0%, AOR = 2.43, 95% CI = 1.58 to 3.72) than those aged 45 years and older (60.4%) (Table 3). In pairwise comparisons, the flavored e-cigarette use percentage was higher in adult current vapers aged 35-44 years vs 45 years and older, and lower among respondents aged 35-44 years or 45 years and older, than those aged 18-24 years or 25-34 years. Flavored e-cigarette use was statistically significantly higher among never vs current smokers (89.8% vs 72.9%, difference = 16.8%, 95% CI = 11.3% to 22.4%; AOR = 1.84, 95% CI = 1.05 to 3.23), never vs recent former smokers (89.8% vs 80.4%, difference = 9.4%, 95% CI = 2.1% to 16.6%), and never vs long-term former smokers (89.8% vs 73.9%, difference = 15.9%, 95% CI = 9.3% to 22.4%) and did not differ in other pairwise smoking status contrasts. The covariate-adjusted age × smoking status interaction was not statistically significant (P = .78), indicating no evidence that the association of smoking status with flavored e-cigarette use varied by age (stratified prevalences in Supplementary Table 2, available online).
Table 3.
Flavored e-cigarette use, by age, smoking status, and reasons for vaping in current e-cigarette usersa
| Variable | Prevalence |
Unadjusted association estimated |
Adjusted association estimated |
|||
|---|---|---|---|---|---|---|
| No.b (%)c | Difference, % (95% CI)c | OR (95% CI) | P | AOR (95% CI) | P | |
| All current e-cigarette users | ||||||
| Age, y | ||||||
| ≥45+ | 216 (60.4) | Referent | Referent | Referent | — | |
| 35-44 | 118 (76.0) | 15.7 (8.1 to 23.2)e | 2.08 (1.42 to 3.05) | <.001h | 2.43 (1.58 to 3.72)f | <.001h |
| 25-34 | 181 (86.7) | 26.3 (20.9 to 31.8)e | 4.28 (2.98 to 6.15) | <.001h | 4.80 (3.21 to 7.18)f | <.001h |
| 18-24 | 127 (89.6) | 29.2 (22.8 to 35.6)e | 5.63 (3.30 to 9.59) | <.001h | 6.42 (3.59 to 11.50)f | <.001h |
| Cigarette smoking | ||||||
| Current | 279 (72.9) | Referent | Referent | Referent | — | |
| Former (last smoked >1 y ago) | 187 (73.9) | 1.0 (−5.5 to 7.5)g | 1.05 (0.75 to 1.47) | .77 | 0.81 (0.54 to 1.21)f | .30 |
| Former (last smoked ≤1 y ago) | 66 (80.4) | 7.5 (0.5 to 14.5)g | 1.52 (0.99 to 2.34) | .06 | 0.94 (0.54 to 1.65)f | .83 |
| Never | 106 (89.8) | 16.8 (11.3 to 22.4)g | 3.25 (1.94 to 5.44) | <.001h | 1.84 (1.05 to 3.23)f | .03h |
| Ever smokers only | ||||||
| Vape when or where smoking not allowed | ||||||
| No | 312 (71.1) | Referent | Referent | Referent | — | |
| Yes | 223 (78.9) | 7.7 (2.4 to 13.0) | 1.51 (1.14 to 2.02) | .005h | 1.40 (1.01 to 1.95)f | .046h |
| Vape to help quit smoking | ||||||
| No | 131 (73.9) | Referent | Referent | Referent | — | |
| Yes | 405 (75.0) | 1.1 (−4.9 to 7.1) | 1.06 (0.77 to 1.46) | .71 | 0.87 (0.61 to 1.25)f | .44 |
Used 1 or more nontobacco flavors (see Table 2 for prevalence of use of each flavor within group). AOR = adjusted odds ratio; CI = confidence interval; OR = odds ratio.
Unweighted sample size.
Weighted estimates within row unless otherwise specified.
Estimate of association of respective regressor with flavored e-cigarette use (yes vs no) from logistic regression models using sampling weights.
Pairwise comparisons: 25-34 vs 35-44 years (P = .002), 18-24 vs 35-44 years (P < .001), 18-24 vs 25-34 years (P = .38).
Model includes all variables in Table 1 as simultaneous regressors.
Pairwise comparisons: former (>1 y) vs never smokers (P < .001), former (< 1 y) vs never smokers (P = .009), and former (>1 y) vs former (<1 y) (P = .10) product(s).
Statistically significant after multiple test correction to control study-wise false discovery rate to .05.
Among ever smoking current e-cigarette users, the unadjusted percentage and adjusted odds of using flavored e-cigarettes was statistically significantly higher in respondents who did vs did not report using e-cigarettes because they could vape when or where smoking is disallowed (78.9% vs 71.1%, difference = 7.7%, 95% CI = 2.4% to 13.0%, P = .005; AOR = 1.52, 95% CI = 1.10 to 2.10) (Table 3). Flavored e-cigarette use percentage did not differ between smokers who did vs did not report using e-cigarettes to quit smoking (75.0% vs 73.9%, difference = 1.1%, P = .71, 95% CI = −4.9% to 7.1%) (Table 3). The covariate-adjusted reason for vaping × smoking status interactions were not statistically significant for use of e-cigarettes to quit smoking (P = .32) or to vape when or where smoking is not allowed (P = .12) (see stratified prevalences in Supplementary Table 3, available online).
Flavored e-Cigarette Use Prevalence Stratified by Subpopulation Groups Among All Adults
Figure 1 depicts the weighted percentages and Ns of flavored or unflavored e-cigarette use across each subpopulation in the entire sample. An estimated 0.5% (95% CI = 0.43% to 0.58%) of all respondents were 18- to 24-year-old current flavored e-cigarette users, 0.5% (95% CI = 0.39% to 0.51%) were 25- to 34-year-old flavored e-cigarette users, 0.2% (95% CI = 0.19% to 0.27%) were 35- to 44-year-old flavored e-cigarette users, and 0.4% (95% CI = 0.35% to 0.47%) were flavored e-cigarette users aged 45 years and older. Among all adults, 0.4% (95% CI = 0.31% to 0.44%) were never smokers, 0.2% (95% CI = 0.13% to 0.21%) were recent former smokers, 0.4% (95% CI = 0.35% to 0.46%) were long-term former smokers, and 0.6% (95% CI = 0.56% to 0.71%) were current smokers that used flavored e-cigarettes. Individuals who vaped for the purpose of quitting smoking and currently used flavored e-cigarettes constituted 0.9% (95% CI = 0.82% to 0.99%) of all adults. An estimated 0.6% (95% CI = 0.49% to 0.62%) of adults were current flavored e-cigarette users who reportedly used e-cigarettes to vape when or where smoking was not allowed.
Figure 1.
Estimated prevalence of current flavored and unflavored e-cigarette use by age, smoking status, and reason for vaping among all US adults. TUS-CPS = Tobacco Use Supplement of the Current Population Survey. aWeighted percent (error bars are 95% confidence intervals [CIs]). bWeighted N (error bars are 95% CIs) based on weighted total population estimate of 249 531 000 US adults. cEstimates include only ever smokers.
Estimated Population Counts (Weighted Ns) Stratified by Potential for Vaping-Related Harm Reduction
The total adult US noninstitutionalized civilian population of current flavored e-cigarette users with potential for vaping-related harm reduction included an estimated 2 251 000 (95% CI = 2 046 000 to 2 476 000) ever smokers who reported vaping for the purpose of quitting smoking. An estimated 1 682 000 (95% CI = 1 493 000 to 1 894 000) adults were current flavored e-cigarette users without potential for vaping-related harm reduction, which is the sum of 922 000 (95% CI = 775 000 to 1 097 000) never smokers and 760 000 (95% CI = 645 000 to 895 000) smokers who reportedly did not vape to quit smoking. Among all adult current flavored e-cigarette users, 57.2% used e-cigarettes to quit smoking and 42.8% did not use e-cigarettes to quit smoking.
Discussion
This study found that current flavored e-cigarette use was reported by a low percentage of US adults overall (1.6%, 95% CI = 1.47% to 1.69%) but constituted most adult current e-cigarette users (77.3%, 95% CI = 74.4% to 79.9%) in July 2018. The Population Assessment of Tobacco and Health (PATH) study found that flavored e-cigarette use was reported by 83.5% of US adult current e-cigarette users in 2014-2015. Like the current study, the PATH study found that flavored e-cigarette use was more common among younger adults and never smokers (6,8). This study’s multivariable analysis provides new evidence that age and smoking status are each independently associated with flavored e-cigarette use, with no evidence of interactive associations.
The current findings that flavored e-cigarette use prevalence varies by age and smoking status is important because the health consequences of vaping also vary by age and smoking status. A National Academies of Science, Engineering, and Medicine report concluded that e-cigarette aerosol contains numerous potentially toxic compounds but at levels substantially lower than cigarette smoke (1). That report deduced that completely switching from smoking to vaping reduces short-term adverse health outcomes in several organ systems (1). Along with the National Academies of Science, Engineering, and Medicine report, a US Surgeon General’s report concluded that e-cigarette use delivers substantial levels of nicotine, which may adversely affect adolescent and young adult neurodevelopment, increases risk of nicotine dependence, increases the likelihood that never smokers will start smoking combustible cigarettes, and is associated with respiratory symptoms (1,2).
The current investigation addresses previously understudied heterogeneity within dual users of combustible and e-cigarettes. Interpreting the health implications of vaping in dual users—the modal smoking status of US adult e-cigarette users (16)—is difficult. Some dual users may be in a temporary transition state that precedes smoking cessation (18). For others, dual use may be prolonged (18). Some dual users do not use e-cigarettes to quit smoking but instead to self-administer nicotine indoors or other circumstances where or when smoking is not allowed (19). Long-term dual use may be problematic. Dual users report greater cardiopulmonary symptoms and have higher biomarker concentrations of carcinogens and respiratory and cardiovascular toxicants than cigarette-only users (11,12). Although flavored e-cigarette use was not associated with reported use of e-cigarettes to quit smoking in this study, it was more common among vapers who did vs did not report vaping because e-cigarettes could be used where or when smoking is disallowed.
The US Tobacco Control Act stipulates that federal regulatory decision making must consider the health impact for the entire population, balancing the number of people who may benefit with the number placed at risk from a policy change (20). The FDA has expressed concern that e-cigarette flavor regulations could deter adults who might prefer flavored vaping products from using e-cigarettes for the purpose of quitting smoking (21). Consequently, flavored e-cigarette sales restrictions were not issued when the FDA initially assumed regulatory jurisdiction over e-cigarettes (21). The population health importance of this concern has been unclear, in part because the number of US adult flavored e-cigarette users who vape to quit smoking was unknown.
This study’s estimated US total population of adult current flavored e-cigarette users who vape to quit smoking (2 251 000, 95% CI = 2 046 000 to 2 476 000) addresses an evidence gap. This estimate can be considered alongside the 5 380 000 estimated US youth past-month e-cigarette users in 2019 reported elsewhere (22). Most (8,23), but not all (22), studies find that the vast majority of US youth vapers use flavored products. An additional 1 682 000 (95% CI = 1 493 000 to 1 894 000) US adult flavored e-cigarette users in this study vaped for reasons other than to help with smoking cessation. Head-to-head comparisons between TUS-CPS and youth estimates cannot be made because of different survey methodologies. This limitation notwithstanding, the collective evidence suggests that adults who vape e-cigarettes to quit smoking constitute a minority of the total US youth and adult population who vape flavored e-cigarettes. Accordingly, future e-cigarette flavor regulations might reduce the availability of e-cigarette products that are perhaps used by more US residents potentially harmed by vaping than by people who might benefit from vaping.
There are caveats to this study’s policy implications. If a future policy change were to make flavored e-cigarettes less accessible, the extent to which current flavored e-cigarette users would switch to unflavored e-cigarettes, use combustible tobacco, or quit using all tobacco products is unknown and merits inquiry. Additionally, the population health impact of potential regulations is based on not only the prevalence but also the magnitude of harm or harm reduction by vaping. The magnitude of harm experienced from vaping for never smokers or long-term dual users is not entirely clear, nor is the magnitude of harm reduction experienced by completely switching from cigarettes to e-cigarettes. Each of these magnitude estimates should be factored into the projected population health impact of any regulation.
First, self-report data are subject to measurement error. The CPS takes multiple steps to minimize measurement error (see Supplementary Methods, available online). This study used only self-respondents and a simple binary outcome that amalgamated multiple specific flavors into 2 clear, conceptually discrete categories (use of nontobacco flavor; yes vs no). This approach reduces confusion or recall errors compared with nuanced multinomial classifications (eg, those distinguishing fruit vs dessert flavors). Yet, measurement error cannot be ruled out and is particularly salient with low base rates regarding proportional effects on estimates. The percentages of flavored e-cigarette use in the overall CPS-TUS sample had low base rates and are subject to this limitation. A second related limitation of the CPS-TUS assessment is that parsing differences between use of specific flavors (eg, menthol vs mint) is not possible and requires other data sources. Third, although the use of sampling weights aims to maximize generalizability to the US population and corrects for under- or oversampling of certain groups, sampling biases cannot be ruled out because the demographic makeup differed between self-responders included in this study and proxy- and nonresponders excluded from the study. Fourth, the cross-sectional design and availability of flavored e-cigarette use data in current vapers only preclude inferences regarding the order of vaping and smoking. Additionally, this limitation excludes former vapers from analyses. Fifth, population count estimates are subject to error, exclude military and institutionalized groups, and are deflated by missing data, although missingness was rare. Sixth, the use of e-cigarettes to quit smoking measure may not necessarily correspond with future quitting behavior.
Although most adult vapers used flavored e-cigarettes in July 2018, current e-cigarette users who vaped nontobacco flavors constituted a small percentage of all US adults. Flavored e-cigarette use was disproportionately common in groups for whom vaping-related harms may exceed vaping-related harm reduction. Flavored e-cigarette users were constituted by a mix of smokers who vape to quit smoking, smokers who vape for other reasons, and never smokers, with none of these subgroups constituting an overwhelming majority. The current findings, alongside other research on the prevalence and health effects of flavored e-cigarette use, will be critical to informing regulatory strategies that optimize impact on the tobacco product–related cancer burden for the entire population.
Funding
Research reported in this publication was supported by the National Cancer Institute and the Food and Drug Administration Center for Tobacco Products (CTP) under Award Number R03CA228909 to Dr Dai and Award Number U54CA180905 to Dr Leventhal and by the National Institute on Drug Abuse Award Number K24DA048160 to Dr Leventhal.
Notes
Role of funder: The funding agency had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclosures: The authors report no potential conflicts of interests.
Prior presentations: None.
Disclaimers: The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or the Food and Drug Administration.
Data availability
HD had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. The CPS-TUS data are publicly available and can be downloaded here: https://cancercontrol.cancer.gov/brp/tcrb/tus-cps/. Statistical software code used for this article can be provided upon request to HD at daisy.dai@unmc.edu.
Supplementary Material
References
- 1.National Academies of Sciences Engineering Medicine. Public Health Consequences of E-Cigarettes. Washington, DC: The National Academies Press; 2018. [PubMed] [Google Scholar]
- 2.U.S. Department of Health and Human Services. E-Cigarette Use Among Youth and Young Adults: A Report of the Surgeon General. Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2016. [Google Scholar]
- 3.The Food and Drug Administration. FDA finalizes enforcement policy on unauthorized flavored cartridge-based e-cigarettes that appeal to children, including fruit and mint. https://www.fda.gov/news-events/press-announcements/fda-finalizes-enforcement-policy-unauthorized-flavored-cartridge-based-e-cigarettes-appeal-children. 2020. Accessed February 28, 2020.
- 4. Tackett AP, Hebert ET, Stevens EM, et al. E-cigarette regulation: a delicate balance for public health. Addiction. 2020. doi:10.1111/add.15092. Online ahead of print. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Bonhomme MG, Holder-Hayes E, Ambrose BK, et al. Flavoured non-cigarette tobacco product use among US adults: 2013-2014. Tob Control. 2016;25(Suppl 2):ii4-ii13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Schneller LM, Bansal-Travers M, Goniewicz ML, et al. Use of flavored electronic cigarette refill liquids among adults and youth in the US-Results from Wave 2 of the Population Assessment of Tobacco and Health Study (2014-2015). PLoS One. 2018;13(8):e0202744. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Soneji SS, Knutzen KE, Villanti AC.. Use of flavored e-cigarettes among adolescents, young adults, and older adults: findings from the population assessment for tobacco and health study. Public Health Rep. 2019;134(3):282-292. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Rostron BL, Cheng YC, Gardner LD, et al. Prevalence and reasons for use of flavored cigars and ENDS among US youth and adults: estimates from wave 4 of the PATH study, 2016-2017. Am J Health Behav. 2020;44(1):76-81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Temple JR, Shorey RC, Lu Y, et al. E-cigarette use of young adults motivations and associations with combustible cigarette alcohol, marijuana, and other illicit drugs. Am J Addict. 2017;26(4):343-348. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Coleman BN, Rostron B, Johnson SE, et al. Electronic cigarette use among US adults in the Population Assessment of Tobacco and Health (PATH) Study. Tob Control. 2017;26(e2):e117. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Wang JB, Olgin JE, Nah G, et al. Cigarette and e-cigarette dual use and risk of cardiopulmonary symptoms in the Health eHeart Study. PLoS One. 2018;13(7):e0198681. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Goniewicz ML, Smith DM, Edwards KC, et al. Comparison of nicotine and toxicant exposure in users of electronic cigarettes and combustible cigarettes. JAMA Netw Open. 2018;1(8):e185937. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Dai H, Leventhal AM.. Prevalence of e-cigarette use among adults in the United States, 2014-2018. JAMA. 2019;322(18):1824. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.National Cancer Institute. Current Population Survey, July 2018. Tobacco Use Supplement File, Technical Documentation. https://cancercontrol.cancer.gov/brp/tcrb/tus-cps/results/data1819/cpsjul18-TechnicalDocumentation.pdf. Accessed December 27, 2019.
- 15.Public Health Law Center. U.S. Sales Restrictions on Flavored Tobacco Products. 2020. https://publichealthlawcenter.org/sites/default/files/resources/US-sales-restrictions-flavored-tobacco-products-2020.pdf.
- 16. Jamal A, Phillips E, Gentzke AS, et al. Current cigarette smoking among adults - United States, 2016. MMWR Morb Mortal Wkly Rep. 2018;67(2):53-59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Thissen D, Steinberg L, Kuang D.. Quick and easy implementation of the Benjamini-Hochberg procedure for controlling the false positive rate in multiple comparisons. J Educ Behav Stat. 2002;27(1):77-83. [Google Scholar]
- 18. Coleman B, Rostron B, Johnson SE, et al. Transitions in electronic cigarette use among adults in the Population Assessment of Tobacco and Health (PATH) Study, Waves 1 and 2 (2013-2015). Tob Control. 2019;28(1):50-59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Robertson L, Hoek J, Blank ML, et al. Dual use of electronic nicotine delivery systems (ENDS) and smoked tobacco: a qualitative analysis. Tob Control. 2019;28(1):13-19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.The Food and Drug Administration. Tobacco Control Act. 2009. http://www.fda.gov/TobaccoProducts/GuidanceComplianceRegulatoryInformation/ucm246129.htm. Accessed January 18, 2020.
- 21.The Food Drug Administration HHS. Deeming tobacco products to be subject to the Federal Food, Drug, and Cosmetic Act, as amended by the Family Smoking Prevention and Tobacco Control Act; restrictions on the sale and distribution of tobacco products and required warning statements for tobacco products. Final rule. Fed Regist. 2016;81(90):28973. [PubMed] [Google Scholar]
- 22. Wang TW, Gentzke AS, Creamer MR, et al. Tobacco product use and associated factors among middle and high school students - United States, 2019. MMWR Surveill Summ. 2019;68(12):1-22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Leventhal AM, Miech R, Barrington-Trimis J, et al. Flavors of e-cigarettes used by youths in the United States. JAMA. 2019;322(21):2132. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
HD had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. The CPS-TUS data are publicly available and can be downloaded here: https://cancercontrol.cancer.gov/brp/tcrb/tus-cps/. Statistical software code used for this article can be provided upon request to HD at daisy.dai@unmc.edu.