Abstract
Dual-users of cigarettes and e-cigarettes are commonly treated as a single group. Our study applied a more nuanced classification of this complex behavior to examine its associations with future tobacco use behaviors using data from Waves 1 and 3 of the Population Assessment of Tobacco and Health. Dual-users at Wave 1 (n=1,665) were categorized into 4 groups based on the frequency with which they used each product (i.e., some days, daily). Analyses identified sociodemographic correlates of group membership and the prevalence of (1) completely switching to e-cigarettes and (2) quitting both products by Wave 3. Dual-users who smoked cigarettes every day and used e-cigarettes some days (69.6%) were the majority and more likely to have lower education (p<.001). Although some day smoking and daily e-cigarette use was the least common category (5.9%), these individuals were most likely to have completely switched to e-cigarettes by Wave 3 (aOR=6.19, 95% CI=3.91, 9.79). Dual-users who smoked and used e-cigarettes some days were most likely to have completely quit tobacco by Wave 3 (aOR=3.98, 95% CI=2.93, 5.40). In general, dual-users who had higher education or income were more likely to have completely switched to e-cigarettes or quit tobacco use by Wave 3. Adults who concurrently use cigarettes and e-cigarettes exhibit considerable heterogeneity in their use of these tobacco products. Dual-users that are higher on the socioeconomic gradient are more likely to engage in plausibly less harmful dual-use behaviors, which are more strongly associated with harm reduction and cessation behaviors. Future research should consider this variation to more accurately characterize the public health impact of dual-use.
Keywords: Cigarette smoking, e-cigarette vaping, dual-use, harm reduction, health disparities
Introduction
Cigarette and e-cigarette dual-use (hereafter referred to as “dual-use”) is a common tobacco use behavior. Among U.S. adults, nearly 70% of current e-cigarette users also currently smoke cigarettes [5]. E-cigarettes have been lauded as a safer means of nicotine delivery than cigarettes [26] because they eliminate the combustion byproducts from cigarettes that are responsible for most of the health consequences of smoking [10]. In order for measurable harm reduction to occur, smokers must completely switch from cigarettes to e-cigarettes [32]. However, there is inconclusive evidence that dual-use meaningfully reduces exposure to the toxicants present in cigarettes. Many cigarette smokers who engage in dual-use report using e-cigarettes to transition off of cigarettes or as a nicotine quitting aid, [22] but population patterns of complete switching and cessation among dual-users are understudied. Dual-use may operate as a transition phase en route to harm reduction or complete cessation. Conversely, this behavior may reduce the likelihood of quitting cigarettes if it is associated with continued use of both products or relapse back to smoking. Thus, there is a crucial need to study dual-use and its association with tobacco use trajectories to better characterize public health implications of dual-use.
Nicotine and tobacco researchers have generally treated cigarette and e-cigarette dual-use as a homogeneous behavior [19, 20, 25, 30, 24]. Recognizing that dual-use may actually comprise a number of conceptually distinct tobacco use behaviors, scientists have recently begun to consider nuanced classification schemes for describing individuals who concurrently use both products [28, 2]. These classification schemes often specify four distinct dual-use behaviors that vary by individual frequencies of cigarette and e-cigarette use. Epidemiological studies that apply these classification schemes often examine cross-sectional and longitudinal associations of these four dual-use behaviors. For example, Borland et al. [3] found that predominant smoking was the most common dual-use behavior among adult tobacco users in the U.S., Canada, England, and Australia while light dual-use was the least common. The authors also found differences in beliefs and attitudes towards smoking and vaping among the four dual-use behaviors. For example, predominant smokers and heavy dual-users had the most positive attitudes toward smoking and the least positive attitudes toward vaping [3]. At least one qualitative study also found differences in psychosocial factors for dual-use among e-cigarette users with varying frequencies of smoking [27]. Therefore, the fourfold classification schemes of dual-use reflect the heterogeneity inherent specifically in dual-use and in tobacco use more generally.
Less understood is the longitudinal relationship between dual-use typologies and associated smoking trajectories. One recent study highlighted the variability in e-cigarette use transitions, noting that a substantial number of dual-users may discontinue e-cigarette use in the future while continuing to smoke cigarettes, but the authors did not account for behavioral differences among dual-users [4]. Heterogeneity in dual-use is particularly important to consider in the context of tobacco harm reduction efforts because some dual-use behaviors may undermine efforts to reduce harm from tobacco use or completely quit, while others may facilitate cessation. Moreover, these differences could impact smoking disparities if dual-use categories are associated with sociodemographic characteristics known to be linked with smoking behaviors. It is crucial to consider how these sources of heterogeneity affect tobacco harm reduction and cessation efforts, particularly to minimize health disparities and ensure equitable impact.
Using longitudinal data from Waves 1 and 3 of the Population Assessment of Tobacco and Health (PATH) Study, we sought to examine heterogeneity in dual-use with respect to variations in frequencies of cigarette and e-cigarette use and sociodemographic characteristics. Our study also sought to examine prospective associations between dual-use behaviors and two tobacco use outcomes, (1) completely switching to e-cigarettes and (2) completely quitting tobacco use. These outcomes are important to examine because they are the stated end goals of harm reduction and tobacco control efforts, respectively. The findings of our study will improve estimation of the public health impact of dual-use as well as improve its measurement.
Methods
Participants and procedures
Data for our study came from Waves 1 and 3 of the PATH Study [15], a nationally representative, longitudinal cohort study of U.S. adults (ages ≥18 years) and adolescents (ages 12 to 17). The PATH Study collects information on tobacco use behaviors such as cigarette smoking, psychosocial determinants of these behaviors, and related health outcomes using Audio Computer-Assisted Self-Interviewing. Waves 1 and 3 of PATH were conducted during 2013–2014 and 2015–2016, respectively. Detailed information on the design and methods of the PATH Study are available elsewhere [15]. Our analytic sample included 1,665 U.S. adults who currently smoked every day or some days and currently used e-cigarettes (i.e., “vaped”) every day or some days at Wave 1 and had complete data on tobacco use and sociodemographic variables at Waves 1 and 3. Columbia University Medical Center’s Institutional Review Board approved all study procedures.
Measures
Sociodemographic variables included age, sex, race and ethnicity (white, non-Hispanic; or other), sexual orientation (gay, lesbian, or bisexual; or straight), education (< high school, high school degree or equivalent, some college or associate degree, or college degree or higher), and poverty (below federal poverty level, at or near poverty level, or at or above twice poverty level).
Cigarette and e-cigarette dual-use
Current cigarette smoking was defined as having ever smoked at least 100 cigarettes and now smoking every day or some days [16]. Current e-cigarette users were adults who reported ever vaping and now vape every day or some day [7]. We classified participants’ dual-use behavior based on their smoking and vaping frequencies. Adults who were every day smokers and vapers were classified as heavy dual-users. Those who were every day smokers and some day vapers were classified as predominant smokers. Adults who were some day smokers and every day vapers were classified as predominant vapers. Finally, adults who were some day smokers and vapers were classified as light dual-users.
Harm reduction and tobacco cessation
The primary outcomes were harm reduction and tobacco cessation at Wave 3. Harm reduction was defined as exclusively vaping every day or some days, and tobacco cessation was defined as neither smoking nor vaping.
Statistical analysis
We calculated weighted sample characteristics and prevalence rates of the four dual-use behaviors. To characterize sociodemographic heterogeneity among adult dual-users with respect to dual-use behavior, we used the Rao-Scott χ2 test to examine differences in the prevalence of dual-use behaviors by sociodemographic characteristics. We present weighted prevalence of dual-use behaviors within a sociodemographic subgroup (e.g., gay, lesbian, or bisexual adults).
We used multivariate logistic regression to examine how the adjusted odds of harm reduction and tobacco cessation at Wave 3 varied by dual-use behaviors at Wave 1 while controlling for sociodemographics characteristics. We modeled the odds of the outcome behavior of interest against all other outcomes. In these analyses, predominant smokers were the referent group because they were the largest group of dual-users. We initially used separate models to examine the unadjusted prospective associations of harm reduction and tobacco cessation with dual-use behaviors and each of the sociodemographic characteristics. We then estimated adjusted associations with a single model that included the main independent variable as well as all sociodemographic characteristics.
We used R (ver. 3.5.3) [31] with the survey package (ver. 3.36) [18] to account for complex survey design and weights when estimating point estimates, confidence intervals, tests of association, and multivariate logistic regression models. Analyses used a critical alpha of .05 (or equivalent 95% confidence intervals).
Results
One-fifth (20.2%) of dual-users were young adults between the ages of 18 and 25 years (Table 1). Approximately half (49.2%) were female and three-quarters were white, non-Hispanic. Over a third had a household income that was below the federal poverty line (35.5%) while 10.2% were gay, lesbian, or bisexual.
Table 1:
Dual-users of cigarettes and e-cigarettes, Wave 1 of the Population Assessment of Tobacco and Health (n = 1,665)
| % | 95% CI | |
|---|---|---|
| Age (years) | ||
| 18–25 | 20.2 | 19.0, 21.5 |
| 26–34 | 24.6 | 22.9, 26.2 |
| 35–44 | 20.6 | 18.9, 22.4 |
| 45–54 | 19.0 | 17.6, 20.4 |
| 55+ | 15.6 | 14.2, 17.0 |
| Sex | ||
| Male | 50.8 | 48.8, 52.8 |
| Female | 49.2 | 47.2, 51.2 |
| Race | ||
| White | 75.0 | 73.3, 76.8 |
| Other | 25.0 | 23.2, 26.7 |
| Sexual orientation | ||
| Gay, lesbian, or bisexual | 10.2 | 9.1, 11.3 |
| Straight | 89.8 | 88.7, 90.9 |
| Education | ||
| < high school | 12.7 | 11.5, 14.0 |
| High school degree or equivalent | 35.2 | 33.3, 37.1 |
| Some college or Associate’s degree | 38.9 | 36.7, 41.0 |
| College degree or higher | 13.2 | 12.0, 14.4 |
| Poverty | ||
| Below poverty level | 35.5 | 33.7, 37.3 |
| At or near poverty level | 27.6 | 25.7, 29.5 |
| At or above twice poverty level | 36.9 | 34.7, 39.1 |
| Dual-use behavior | ||
| Heavy dual-user | 9.9 | 8.7, 11.1 |
| Predominant smoker | 69.6 | 67.8, 71.4 |
| Predominant vaper | 5.9 | 4.9, 6.9 |
| Light dual-user | 14.6 | 13.3, 15.8 |
| Tobacco use outcomes | ||
| Harm reduction | 6.8 | 5.6, 8.0 |
| Tobacco cessation | 9.3 | 8.3, 10.4 |
Notes. Includes respondents with data from Waves 1 and 3. Heavy dual-users were every day smokers and vapers. Predominant smokers were every day smokers and some day vapers. Predominant vapers some day smokers and every day vapers. Light dual-users were some day smokers and vapers. Harm reduction was defined as exclusively vaping every day or some days at Wave 3. Tobacco cessation was defined as neither smoking nor vaping at Wave 3.
At Wave 1, the vast majority of dual-users were predominant smokers (69.6%, 95% CI: 67.8%, 71.4%). Light dual-use was the second most prevalent dual-use behavior (14.6%, 95% CI: 13.3%, 15.8%), followed by heavy dual-use (9.9%, 95% CI: 8.7%, 11.1%) and predominant vaping (5.9%, 95% CI: 4.9%, 6.9%).
Sociodemographic heterogeneity in dual-use behaviors
The prevalence of dual-use behaviors differed by age (p < .001), the differences being driven largely by young adults (between the ages of 18 and 25) and older adults (ages 55 and above). More specifically, light dual-use was slightly more common among young adults (22.8%), and predominant smoking was slightly less common (62.6%). In contrast, heavy dual-use was more common (13.3%) and light dual-use (7.3%) were less common among older adults. The patterns of dual-use prevalence among other age groups generally mirrored patterns in the overall sample.
The prevalence of dual-use behaviors also differed by sex (p = .02), race/ethnicity (p < .001), and education (p < .001). A larger percentage of males were heavy dual-users (11.3%) than females (8.5%) while a larger percentage of females (71.4%) were predominant smokers than males (67.9%). Among adults from non-White racial/ethnic backgrounds, light dual-use (23.3%) was more common while heavy dual-use (5.9%) was less common. Additionally, light dual-use (25.9%) was more common among adults with a college degree or higher while predominant smoking (59.0%) was less common.
Harm reduction
Overall, 6.8% (95% CI = 5.6, 8.0) of dual-users engaged in harm reduction by switching to exclusive vaping at Wave 3. Heavy dual-users were more likely to engage in subsequent harm reduction than predominant smokers (8.9% v. 4.5%, aOR = 2.00, 95% CI = 1.21, 3.30) and predominant vapers were much more likely to do so (22.7% v. 4.5%, aOR = 6.19, 95% CI = 3.91, 9.79). Light dual-users were also more likely to completely switch to vaping (10.2% v. 4.5%, aOR = 2.28, 95% CI = 1.44, 3.61). Females were less likely to engage in harm reduction than males (5.4% v. 8.1%, aOR = 0.70, 95% CI = 0.52, 0.93). Dual-users who had a household income at or above twice the federal poverty level were more likely to engage in harm reduction than dual-users who had a household income below the poverty level (10.0% v. 4.5%, aOR = 2.22, 95% CI = 1.51, 3.28).
Tobacco cessation
Overall, 9.3% (95% CI = 8.3, 10.4) of dual-users stopped using both cigarettes e-cigarettes at Wave 3. Light dual-users were much more likely to engage in subsequent tobacco cessation than predominant smokers (23.1% v. 6.6%, aOR = 3.98, 95% CI = 2.93, 5.40). Dual-users between the ages of 26 and 34 or 35 and 44 were less likely to engage in subsequent tobacco cessation than young dual-users (9.5 v. 13.3, aOR1 = 0.53, 95% CI1 = 0.35 – 0.80; 6.7, 13.3, aOR2 = 0.65, 95% CI2 = 0.44 – 0.96). Females were somewhat less likely to engage in subsequent tobacco cessation than males (8.0 v. 10.5, aOR = 0.77, 95% CI = 0.61, 0.98). Dual-users who had at least a high school degree or equivalent were much more likely to engage in subsequent tobacco cessation than dual-users with less than a high school education (8.5 v. 3.6, aOR1 = 2.53, 95% CI1 = 1.23, 5.21; 11.1 v. 3.6, aOR2 = 3.38, 95% CI2 = 1.67, 6.83; 11.5 v. 3.6, aOR3 = 3.12, 95% CI3 = 1.39, 6.99).
Discussion
Among U.S. adults who concurrently used cigarettes and e-cigarettes, we observed heterogeneity in dual-use owing to two sources, smoking and vaping frequencies and sociodemographic characteristics. The overwhelming majority of dual-users were predominant smokers; heavy and light dual-use were much less prevalent while predominant vaping was the least prevalent dual-use behavior. Plausibly less harmful dual-use behaviors (e.g., light dual-use) tended to be more common among more educated and wealthier dual-users, patterns that follow from previous research on e-cigarettes [23, 13]. Although harm reduction by completely switching to e-cigarettes and tobacco cessation were not very common between waves, dual-use behaviors and some sociodemographic characteristics (e.g., education, income) were associated with both. Our results suggest that cigarette and e-cigarette dual-use should not be treated as a single risk behavior, but instead refers to a group of distinct behaviors with different epidemiological profiles.
While completely switching to e-cigarettes eliminates the health harms of cigarette smoking originating from the combustion byproducts [8], the observed heterogeneity in dual-use behaviors suggests that these benefits may not be realized by all dual-users in the context of harm reduction efforts. Most notable in this regard is our finding that predominant smokers, the largest group of dual-users, were less likely to engage in harm reduction by Wave 3. Furthermore, predominant smokers and heavy dual-users were similarly unlikely to engage in subsequent tobacco cessation. It is possible that these dual-users, who were all daily smokers, had greater nicotine dependence so that vaping did not sufficiently aid them in transitioning to exclusive e-cigarette use or quitting tobacco use altogether [17, 29]. Another plausible explanation is that these individuals largely used e-cigarette devices with product characteristics that did not facilitate tobacco use transitions. To this end, a number of studies have already noted great diversity in available e-cigarette devices and how they are marketed and used [33, 12]. Future studies should explore these possibilities in greater detail given that they may account for why the largest proportion of dual-users may be less likely to engage in harm reduction.
Also notable is that the likelihood of harm reduction or tobacco cessation essentially followed the socioeconomic gradient. Generally speaking, dual-users with higher levels of education or income were more likely to engage in harm reduction or tobacco cessation by Wave 3. This is problematic given that plausibly less harmful dual-use behaviors were more prevalent among dual-users with higher levels of education or income. Therefore, observed patterns of harm reduction have the potential to reinforce long-standing smoking-related health disparities along the socioeconomic gradient through inequitable impact [9] a common unintended consequence of many tobacco control policies [11]. This is particularly true given that people with higher levels of education or income have generally reported greater awareness of e-cigarettes and ever use [14]. A health equity perspective should be adopted when designing, implementing, and evaluating harm reduction policies [21].
Our study, like other recently published studies [3], essentially treated dual-use as a manifest variable with directly observable levels (e.g., heavy dual-use). Such an approach assumes that the frequencies of cigarette and e-cigarette use are perfectly measured (i.e., do not have measurement error) and are sufficient variables for classifying dual-use. That predominant smoking was overwhelmingly prevalent while the other three dual-use behaviors were much less prevalent raises questions about whether predominant smokers can be further classified into distinct groups or whether one or more of the other dual-use behaviors can be considered a single behavior. Future research should consider dual-use as a latent variable and employ latent class analysis [6] to refine dual-use behavior categories. Such an approach permits the inclusion of psychosocial characteristics such as risk perceptions of tobacco products that may further discriminate between dual-users beyond just self-reported cigarette and e-cigarette use.
Strengths of our study include the use of longitudinal data from PATH and complex survey procedures to calculate nationally representative prevalence rates and odds ratios. Our general findings also have face validity given that the associations of predominant vaping with harm reduction and light dual-use with complete quitting both at a later time follow from prior expectations. A limitation is that, although there is variability in defining e-cigarette use [1] potentially leading to alternate dual-use classification schemes, we did not explore other classification schemes due to the possibility of exceedingly small cell sizes. Future studies should explore alternate dual-use classification schemes based on varying definitions of e-cigarette use (e.g., those involving puffs per day or time to first puff).
Conclusions
Predominant smokers account for the overwhelming majority of U.S. adults who concurrently use cigarettes and e-cigarettes. Other dual-use behaviors are not negligible even if they are less prevalent and especially so given their differential associations with subsequent harm reduction and complete quitting. Researchers should continue to refine the measurement of dual-use behaviors to better understand the mechanisms through which they may lead to optimal smoking cessation outcomes, and to assess the impact on priority populations, including vulnerable groups.
Table 2:
Prevalence of dual-use behaviors by sociodemographic groups, Wave 1 of the Population Assessment of Tobacco and Health (n = 1,665)
| Predominant smoker | Heavy dual-user | Predominant vaper | Light dual-user | p | |
|---|---|---|---|---|---|
| Age (years) | |||||
| 18–25 | 62.6 | 8.7 | 6.0 | 22.8 | < .001 |
| 26–34 | 70.4 | 9.5 | 5.6 | 14.4 | |
| 35–44 | 70.6 | 10.1 | 5.9 | 13.4 | |
| 45–54 | 73.7 | 8.7 | 4.3 | 13.3 | |
| 55+ | 71.1 | 13.3 | 8.2 | 7.3 | |
| Sex | |||||
| Male | 67.9 | 11.3 | 6.7 | 14.1 | .02 |
| Female | 71.4 | 8.5 | 5.0 | 15.0 | |
| Race | |||||
| White | 70.9 | 11.2 | 6.2 | 11.7 | < .001 |
| Other | 65.8 | 5.9 | 4.9 | 23.3 | |
| Sexual orientation | |||||
| Gay, lesbian, or bisexual | 70.5 | 6.4 | 7.8 | 15.3 | .10 |
| Straight | 69.5 | 10.3 | 5.7 | 14.5 | |
| Education | |||||
| < high school | 72.2 | 9.2 | 5.9 | 12.6 | < .001 |
| High school degree or equivalent | 72.8 | 11.5 | 5.5 | 10.1 | |
| Some college or Associate’s degree | 69.4 | 8.9 | 6.3 | 15.4 | |
| College degree or higher | 59.0 | 9.3 | 5.8 | 25.9 | |
| Poverty | |||||
| Below poverty level | 71.5 | 8.9 | 6.5 | 13.1 | .23 |
| At or near poverty level | 70.4 | 10.5 | 5.3 | 13.8 | |
| At or above twice poverty level | 67.2 | 10.5 | 5.8 | 16.5 |
Notes. Includes respondents with data from Waves 1 and 3. Heavy dual-users were every day smokers and vapers. Predominant smokers were every day smokers and some day vapers. Predominant vapers some day smokers and every day vapers. Light dual-users were some day smokers and vapers were.
Table 3:
Associations between Wave 1 dual-use behaviors and Wave 3 tobacco use outcomes, Population Assessment of Tobacco and Health (n = 1,665)
| Harm reduction | Tobacco cessation | |||||
|---|---|---|---|---|---|---|
| Weighted % | OR [95 % CI] | aOR [95% CI] | Weighted % | OR [95 % CI] | aOR [95% CI] | |
| Dual-use behavior | ||||||
| Predominant smoker | 4.5 | 1 | 1 | 6.6 | 1 | 1 |
| Heavy dual-user | 8.9 | 2.09 [1.27, 3.44] | 2.00 [1.21, 3.30] | 8.7 | 1.36 [0.87, 2.12] | 1.31 [0.83, 2.07] |
| Predominant vaper | 22.7 | 6.30 [4.17, 9.52] | 6.19 [3.91, 9.79] | 8.2 | 1.27 [0.73, 2.20] | 1.20 [0.69, 2.08] |
| Light dual-user | 10.2 | 2.42 [1.61, 3.63] | 2.28 [1.44, 3.61] | 23.1 | 4.25 [3.21, 5.64] | 3.98 [2.93, 5.40] |
| Age (years) | ||||||
| 18–25 | 6.6 | 1 | 1 | 13.3 | 1 | 1 |
| 26–34 | 8.2 | 1.26 [0.81, 1.94] | 1.31 [0.82, 2.10] | 9.5 | 0.68 [0.48, 0.97] | 0.78 [0.54, 1.12] |
| 35–44 | 6.4 | 0.96 [0.60, 1.53] | 0.95 [0.58, 1.56] | 6.7 | 0.47 [0.32, 0.68] | 0.53 [0.35, 0.80] |
| 45–54 | 5.0 | 0.75 [0.44, 1.27] | 0.81 [0.45, 1.46] | 7.5 | 0.53 [0.35, 0.79] | 0.65 [0.44, 0.96] |
| 55+ | 7.5 | 1.15 [0.72, 1.84] | 1.14 [0.67, 1.95] | 7.3 | 0.67 [0.44, 1.02] | 0.97 [0.62, 1.50] |
| Sex | ||||||
| Male | 8.1 | 1 | 1 | 10.5 | 1 | 1 |
| Female | 5.4 | 0.65 [0.49, 0.87] | 0.70 [0.52, 0.93] | 8.0 | 0.75 [0.59, 0.94] | 0.77 [0.61, 0.98] |
| Race | ||||||
| White | 7.0 | 1 | 1 | 8.7 | 1 | 1 |
| Other | 6.1 | 0.86 [0.57, 1.30] | 0.90 [0.57, 1.40] | 11.3 | 1.34 [1.06, 1.68] | 1.14 [0.89, 1.47] |
| Sexual orientation | ||||||
| Gay, lesbian, or bisexual | 6.9 | 1 | 1 | 9.6 | 1 | 1 |
| Straight | 6.8 | 1.02 [0.55, 1.89] | 1.12 [0.61, 2.06] | 7.2 | 0.73 [0.44, 1.22] | 0.71 [0.40, 1.24] |
| Education | ||||||
| < high school | 6.0 | 1 | 1 | 3.6 | 1 | 1 |
| High school degree or equivalent | 5.5 | 0.92 [0.56, 1.50] | 0.78 [0.47, 1.28] | 8.5 | 2.47 [1.16, 5.24] | 2.53 [1.23, 5.21] |
| Some college or Associate’s degree | 7.0 | 1.19 [0.72, 1.94] | 0.90 [0.51, 1.57] | 11.1 | 3.34 [1.65, 6.75] | 3.38 [1.67, 6.83] |
| College degree or higher | 10.4 | 1.82 [0.96, 3.44] | 1.1 [0.50, 2.05] | 11.5 | 3.46 [1.59, 7.40] | 3.12 [1.39, 6.99] |
| Poverty | ||||||
| Below poverty level | 4.5 | 1 | 1 | 8.2 | 1 | 1 |
| At or near poverty level | 5.3 | 1.19 [0.77, 1.82] | 1.19 [0.76, 1.85] | 10.7 | 1.34 [0.95, 1.90] | 1.21 [0.85, 1.72] |
| At or above twice poverty level | 10.0 | 2.33 [1.68, 3.24] | 2.22 [1.51, 3.28] | 9.4 | 1.16 [0.86, 1.56] | 0.92 [0.67, 1.26] |
Notes. Includes respondents with data from Waves 1 and 3. Heavy dual-users were every day smokers and vapers. Predominant smokers were every day smokers and some day vapers. Predominant vapers some day smokers and every day vapers. Light dual-users were some day smokers and vapers were. Harm reduction was defined as exclusively vaping every day or some days at Wave 3. Tobacco cessation was defined as neither smoking nor vaping at Wave 3.
Acknowledgments
Funding: This work was supported by the Office of the Director, National Institutes of Health (award number DP5OD023064).
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- [1].Amato Michael S, Boyle Raymond G, and Levy David. “How To Define E-Cigarette Prevalence? Finding Clues in the Use Frequency Distribution”. In: Tobacco Control 25e1 (2015), e24–e29. ISSN: 1468–3318 DOI: 10.1136/tobaccocontrol-2015-052236 URL: 10.1136/tobaccocontrol-2015-052236. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Azagba Sunday, Shan Lingpeng, and Latham Keely. “Adolescent Dual Use Classification and Its Association With Nicotine Dependence and Quit Intentions”. In: Journal of Adolescent Health 652 (2019), pp. 195–201. ISSN: 1054–139X DOI: 10.1016/j.jadohealth.2019.04.009 URL: 10.1016/j.jadohealth.2019.04.009. [DOI] [PubMed] [Google Scholar]
- [3].Borland Ron et al. “A New Classification System for Describing Concurrent Use of Nicotine Vaping Products Alongside Cigarettes (so-called “Dual use”): Findings From the Itc-4 Country Smoking and Vaping Wave 1 Survey”. In: Addiction (2019). ISSN: 1360–0443 DOI: 10.1111/add.14570 URL: 10.1111/add.14570. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Coleman Blair 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).” In: Tobacco control 281 (2019), pp. 50–59. DOI: tobaccocontrol-2017–054174[pii] URL: https://doi.org/tobaccocontrol-2017-054174[pii]. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [5].Coleman Blair N et al. “Electronic Cigarette Use Among Us Adults in the Population Assessment of Tobacco and Health (PATH) Study, 2013–2014”. In: Tobacco Control 26e2 (2017), e117–e126. ISSN: 1468–3318 DOI: 10.1136/tobaccocontrol-2016-053462 URL: 10.1136/tobaccocontrol-2016-053462. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [6].Collins Linda M and Lanza Stephanie T. Latent class and latent transition analysis: With applications in the social, behavioral, and health sciences. Vol. 718 John Wiley & Sons, 2009. [Google Scholar]
- [7].Cullen Karen A. et al. “Notes From the Field: Use of Electronic Cigarettes and Any Tobacco Product Among Middle and High School Students - United States, 2011–2018”. In: MMWR. Morbidity and Mortality Weekly Report 6745 (2018), pp. 1276–1277. ISSN: 1545–861X DOI: 10.15585/mmwr.mm6745a5 URL: 10.15585/mmwr.mm6745a5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Douglas Clifford E. et al. “The American Cancer Society Public Health Statement on Eliminating Combustible Tobacco Use in the United States”. In: CA: A Cancer Journal for Clinicians 684 (2018), pp. 240–245. ISSN: 0007–9235 DOI: 10.3322/caac.21455 URL: 10.3322/caac.21455. [DOI] [PubMed] [Google Scholar]
- [9].Drope Jeffrey et al. “Who’s Still Smoking? Disparities in Adult Cigarette Smoking Prevalence in the United States”. In: CA: A Cancer Journal for Clinicians 682 (2018), pp. 106–115. ISSN: 0007–9235 DOI: 10.3322/caac.21444 URL: 10.3322/caac.21444. [DOI] [PubMed] [Google Scholar]
- [10].Farsalinos Konstantinos E. and Polosa Riccardo. “Safety Evaluation and Risk Assessment of Electronic Cigarettes As Tobacco Cigarette Substitutes: a Systematic Review”. In: Therapeutic Advances in Drug Safety 52 (2014), pp. 67–86. ISSN: 2042–0994 DOI: 10.1177/2042098614524430 URL: 10.1177/2042098614524430. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].Graham Hilary. “Why Social Disparities Matter for Tobacco-Control Policy”. In: American Journal of Preventive Medicine 372 (2009), S183–S184. ISSN: 0749–3797 DOI: 10.1016/j.amepre.2009.05.007 URL: 10.1016/j.amepre.2009.05.007. [DOI] [PubMed] [Google Scholar]
- [12].Han Sifei and Kavuluru Ramakanth. “Exploratory Analysis of Marketing and Non-Marketing E-Cigarette Themes on Twitter”. In: Social Informatics (2016), pp. 307–322. ISSN: 1611–3349 DOI: 10.1007/978-3-319-47874-6_22 URL: 10.1007/978-3-319-47874-6_22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [13].Harlow Alyssa F Stokes Andrew, and Brooks Daniel R. “Socioeconomic and Racial/ethnic Differences in E-Cigarette Uptake Among Cigarette Smokers: Longitudinal Analysis of the Population Assessment of Tobacco and Health (PATH) Study”. In: Nicotine & Tobacco Research (2018). ISSN: 1469–994X DOI: 10.1093/ntr/nty141 URL: 10.1093/ntr/nty141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [14].Hartwell Greg et al. “E-Cigarettes and Equity: a Systematic Review of Differences in Awareness and Use Between Sociodemographic Groups”. In: Tobacco Control 26e2 (2016), e85–e91. ISSN: 1468–3318 DOI: 10.1136/tobaccocontrol-2016-053222 URL: 10.1136/tobaccocontrol-2016-053222. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [15].Hyland Andrew et al. “Design and Methods of the Population Assessment of Tobacco and Health (PATH) Study”. In: Tobacco Control 264 (2016), pp. 371–378. ISSN: 1468–3318 DOI: 10.1136/tobaccocontrol-2016-052934 URL: 10.1136/tobaccocontrol-2016-052934. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [16].Jamal Ahmed et al. “Current Cigarette Smoking Among Adults - United States, 2016”. In: MMWR. Morbidity and Mortality Weekly Report 672 (2018), pp. 53–59. ISSN: 1545–861X DOI: 10.15585/mmwr.mm6702a1 URL: 10.15585/mmwr.mm6702a1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [17].Liu Guodong et al. “A Comparison of Nicotine Dependence Among Exclusive E-Cigarette and Cigarette Users in the Path Study”. In: Preventive Medicine 104 (2017), pp. 86–91. ISSN: 0091–7435 DOI: 10.1016/j.ypmed.2017.04.001 URL: 10.1016/j.ypmed.2017.04.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [18].Lumley Thomas. “Analysis of Complex Survey Samples”. In: Journal of Statistical Software 98 (2004). ISSN: 1548–7660 DOI: 10.18637/jss.v009.i08 URL: 10.18637/jss.v009.i08. [DOI] [Google Scholar]
- [19].McCabe Sean Esteban et al. “E-Cigarette Use, Cigarette Smoking, Dual Use, and Problem Behaviors Among U.S. Adolescents: Results From a National Survey.” In: The Journal of Adolescent Health 612 (2017), pp. 155–162. DOI: S1054–139X(17)30070–8[pii] URL: https://doi.org/S1054-139X(17)30070-8[pii]. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [20].Morean Meghan, Suchitra Krishnan-Sarin, and Stephanie S O’Malley. “Comparing Cigarette and E-Cigarette Dependence and Predicting Frequency of Smoking and E-Cigarette Use in Dual-Users of Cigarettes and E-Cigarettes.” In: Addictive Behaviors 87 (2018), pp. 92–96. DOI: S0306–4603(18)30188–6[pii] URL: https://doi.org/S0306-4603(18)30188-6[pii]. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [21].Passey Megan and Bonevski Billie. “The Importance of Tobacco Research Focusing on Marginalized Groups”. In: Addiction 1097 (2014), pp. 1049–1051. ISSN: 0965–2140 DOI: 10.1111/add.12548 URL: 10.1111/add.12548. [DOI] [PubMed] [Google Scholar]
- [22].Pepper Jessica et al. “Reasons for Starting and Stopping Electronic Cigarette Use”. In: International Journal of Environmental Research and Public Health 1110 (2014), pp. 10345–10361. ISSN: 1660–4601 DOI: 10.3390/ijerph111010345 URL: 10.3390/ijerph111010345. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [23].Phillips Elyse et al. “Tobacco Product Use Among Adults - United States, 2015”. In: MMWR. Morbidity and Mortality Weekly Report 6644 (2017), pp. 1209–1215. ISSN: 1545–861X DOI: 10.15585/mmwr.mm6644a2 URL: 10.15585/mmwr.mm6644a2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [24].Piper Megan E et al. “Changes in Use Patterns Over 1 Year Among Smokers and Dual Users of Combustible and Electronic Cigarettes”. In: Nicotine & Tobacco Research (2019). ISSN: 1469–994X DOI: 10.1093/ntr/ntz065 URL: 10.1093/ntr/ntz065. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [25].Megan E Piper et al. “Dual Users Compared To Smokers: Demographics, Dependence, and Biomarkers.” In: Nicotine & Tobacco Research (2018). DOI: 5144984[pii] URL: https://doi.org/5144984[pii]. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [26].Polosa Riccardo et al. “A Fresh Look At Tobacco Harm Reduction: the Case for the Electronic Cigarette”. In: Harm Reduction Journal 101 (2013), p. 19 ISSN: 1477–7517 DOI: 10.1186/1477-7517-10-19 URL: 10.1186/1477-7517-10-19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [27].Robertson Lindsay et al. “Dual Use of Electronic Nicotine Delivery Systems (ENDS) and Smoked Tobacco: a Qualitative Analysis”. In: Tobacco Control (2018), tobaccocontrol–2017–054070. ISSN: 1468–3318 DOI: 10.1136/tobaccocontrol-2017-054070 URL: 10.1136/tobaccocontrol-2017-054070. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [28].Ronayne David and Sgroi Daniel. “On the Motivations for the Dual-Use of Electronic and Traditional Cigarettes”. In: Applied Economics Letters 2512 (2017), pp. 830–834. ISSN: 1466–4291 DOI: 10.1080/13504851.2017.1368985 URL: 10.1080/13504851.2017.1368985. [DOI] [Google Scholar]
- [29].Sung Hai-Yen et al. “Polytobacco Use and Nicotine Dependence Symptoms Among Us Adults, 2012–2014”. In: Nicotine & Tobacco Research 20suppl 1 (2018), S88–S98. ISSN: 1469–994X DOI: 10.1093/ntr/nty050 URL: 10.1093/ntr/nty050. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [30].Symes Yael R et al. “Dual Cigarette and E-Cigarette Use in Cancer Survivors: an Analysis Using Population Assessment of Tobacco Health (PATH) Data.” In: Journal of Cancer Survivorship: Research and Practice 132 (2019), pp. 161–170. DOI: 10.1007/s11764-019-0735-y URL: 10.1007/s11764-019-0735-y. [DOI] [PubMed] [Google Scholar]
- [31].R Core Team. “R: a Language and Environment for Statistical Computing”. In: (2013). [Google Scholar]
- [32].Warner Kenneth E. “How To Think-Not Feel-About Tobacco Harm Reduction”. In: Nicotine & Tobacco Research (2018). ISSN: 1469–994X DOI: 10.1093/ntr/nty084 URL: 10.1093/ntr/nty084. [DOI] [PubMed] [Google Scholar]
- [33].Zhu Shu-Hong et al. “Four Hundred and Sixty Brands of E-Cigarettes and Counting: Implications for Product Regulation”. In: Tobacco Control 23suppl 3 (2014), pp. iii3–iii9 ISSN: 1468–3318 DOI: 10.1136/tobaccocontrol-2014-051670 URL: 10.1136/tobaccocontrol-2014-051670. [DOI] [PMC free article] [PubMed] [Google Scholar]