Abstract
Introduction:
In observational studies, vaping daily is positively associated with cigarette smoking abstinence, while non-daily vaping is associated with less smoking abstinence (versus no e-cigarette use). It remains unknown whether cigarette smokers who vape daily have different motivations for using e-cigarettes than those who vape non-daily.
Methods:
Using latent class analysis and 10 self-reported reasons for vaping, we identified sub-groups of participants based on vaping motivations among 1544 adult (≥18 y) dual users of e-cigarettes and cigarettes at wave 4 of the Population Assessment of Tobacco and Health (2016–2018). We examined the association of motivation sub-groups with vaping frequency at wave 4, and subsequent cigarette smoking abstinence at wave 5 (2018–2019). Additional analyses examined the association of vaping frequency with smoking abstinence before and after adjustment for motivation sub-groups.
Results:
Four distinct sub-groups of e-cigarette users emerged, including 54 % of participants who were “Vaping Enthusiasts”, 20 % who vaped for “Convenience and Social Acceptability”, 10 % classified as “Experimenters” and 16 % who vaped for “Quitting Smoking and Harm Reduction.” The Convenience and Social Acceptability sub-group were less likely than “Vaping Enthusiasts” (AOR=0.29, 95 %CI[0.20–0.42]) and “Quitting Smoking and Harm Reduction” (AOR=0.41, 95 %CI[0.24–0.71]) classes to vape daily (versus non-daily). Sub-groups were not associated with smoking abstinence after one year. Adjustment for motivation sub-groups did not attenuate a positive association of daily vaping with smoking abstinence.
Conclusions:
Cigarette smokers who vaped for convenience and social acceptability were less frequent e-cigarette users than those with other vaping motivations. Vaping motivations were not associated with subsequent smoking abstinence.
Keywords: E-cigarettes, Cigarettes
1. Introduction
Electronic cigarettes (e-cigarettes) are commonly used as a cessation aid by adults who are attempting to quit smoking combustible cigarettes (Benmarhnia et al., 2018; Pepper and Brewer, 2014; Spears et al., 2018; Tackett et al., 2015). Growing research from observational studies indicates that daily e-cigarette use may be effective for smoking cessation, while non-daily e-cigarette use may inhibit smoking cessation relative to no e-cigarette use (Berry et al., 2019; Biener and Hargraves, 2015; Brose et al., 2015; Verplaetse et al., 2019). Daily vaping may increase likelihood of smoking abstinence due to sufficient doses of nicotine to reduce cigarette cravings. Non-daily vaping may not deliver adequate nicotine doses to curb cigarette cravings, leading to continued dual use of both products, rather than full substitution of e-cigarettes for cigarettes (Martínez et al., 2020; Morgan Snell et al., 2020).
The divergent effects of e-cigarette use frequency on smoking abstinence may also be explained by different motivations for using e-cigarettes by individuals who vape daily versus non-daily. Daily users may be more motivated to quit smoking, and thus use e-cigarettes as a substitution product. On the other hand, non-daily users may simply reflect individuals who vape out of convenience, when smoking is not allowed or tolerated, or reflect recreational use. If frequency of e-cigarette use is associated with motivations for e-cigarette use, then differences in smoking abstinence by vaping frequency may be at least partially the result of uncontrolled confounding rather than a reflection of the effectiveness of the vaping behavior for smoking abstinence.
There are a number of potential motivations for vaping among adults who smoke cigarettes, including cigarette smoking cessation or reduction, flavors, convenience, curiosity, positive sensory experiences, and influence from peers (Kinouani et al., 2020). There is some evidence that daily (versus nondaily) vaping is associated with more positive vaping-related attitudes (Borland et al., 2019). However, few studies have explored whether cigarette smokers who vape daily have different motivations for using e-cigarettes than those who vape non-daily. Additionally, studies on motivations for using e-cigarettes have largely examined individual reasons for vaping among adults who smoke cigarettes, rather than examine whether certain motivations for using e-cigarettes cluster together to represent underlying sub-groups of adult e-cigarette users. Evans-Polce et al. identified three distinct sub-groups of e-cigarette users among US 12-graders, including those who use e-cigarettes to substitute smoking, for recreation, or for experimentation (Evans-Polce et al., 2018). Additionally, distinct sub-groups of motivations for adult cannabis use (i.e., therapeutic vs. non-therapeutic) (Lake et al., 2020) and prescription opioid misuse (i.e., pain relief vs. recreation)(Votaw et al., 2019) have been identified in the literature. However, few studies have examined whether there are distinct sub-groups of e-cigarette users among adults who use both e-cigarettes and cigarettes, and whether sub-groups are associated with vaping frequency and cigarette smoking outcomes (Borland et al., 2019; Sutton et al., 2022).
In this study, we use nationally representative data and latent class analysis to identify sub-groups of participants with similar motivations for e-cigarette use among adults who are dual users of combustible cigarettes and e-cigarettes. We classify sub-groups based on self-reported reasons for using e-cigarettes, and then test the association between e-cigarette use sub-groups, frequency of e-cigarette use, and subsequent cigarette smoking abstinence. In additional analyses, we examine the association of vaping frequency with smoking abstinence, before and after adjustment for e-cigarette motivation sub-groups.
2. Materials and methods
2.1. Data source and sample
This study uses publicly available data from the adult cohort at wave 4 (2016–2018) and wave 5 (2018–2019) of the Population Assessment of Tobacco and Health (PATH) study, a large national cohort study on tobacco product use (Hyland et al., 2017). At Wave 1 (2013–2014), 32, 320 adult participants were initially recruited, and participants completed computer-assisted personal interviews approximately 12-months apart. At wave 4, PATH included a replenishment sample to address study attrition among the wave 1 cohort and formed the new wave 4 cohort (33,822 adult participants).
Adult participants (≥18 y) were eligible for inclusion in the analytic sample if they were established combustible cigarette smokers at wave 4 and reported using an electronic nicotine product every day (i.e., daily e-cigarette use) or some days (i.e., non-daily e-cigarette use) at the time of data collection. Established combustible cigarette smoking was defined as having smoked at least 100 cigarettes in one’s lifetime and currently smoking every day or some days at wave 4. The final analytic sample for determining e-cigarette user sub-groups included 1544 adults who were dual users of e-cigarettes and combustible cigarettes at wave 4 and had non-missing data on e-cigarette use frequency, covariates, and at least one reason for using e-cigarettes (n = 47 removed because of missing data). In analyses of wave 5 cigarette smoking outcomes, the sample was further restricted to 1244 participants with wave 5 follow-up data.
2.2. Measures
2.2.1. Independent variable: motivations for E-cigarette use
The primary items used to establish e-cigarette user motivations included 10 questions on reasons for using e-cigarettes measured at wave 4. Participants were prompted to select from the following reasons for using e-cigarettes (select all that apply): “They are affordable,” “I can use electronic nicotine products at times when or in places where smoking cigarettes isn’t allowed,” “They might be less harmful to me than smoking cigarettes,” “They might be less harmful to people around me than cigarettes,” “E-liquid comes in flavors I like,” “Using electronic nicotine products helps people quit smoking cigarettes,” “Electronic nicotine products don’t smell,” “Using an electronic nicotine product feels like smoking a regular cigarette,” “Electronic nicotine products are more acceptable to non-tobacco users,” and “I use electronic nicotine products as a way of cutting down on cigarette smoking.”
2.2.2. Dependent variables: E-cigarette use frequency and cigarette smoking abstinence
E-cigarette use frequency at wave 4 was dichotomized as daily versus non-daily use. Participants were classified as vaping daily if they reported using e-cigarettes “every day”, and as non-daily if they reported using e-cigarettes “some days” at the wave 4 survey. We additionally examined cigarette smoking abstinence at wave 5, defined as currently smoking cigarettes “not at all” at the time of the wave 5 survey.
2.2.3. Covariates
We examined wave 4 socio-demographic characteristics and tobacco history variables. Socio-demographic characteristics included age (18–24 y, 25–34 y, 35–54 y, ≥55 y), biological sex (female, male), race (Black, White, another race/multiple races), ethnicity (Hispanic, Not Hispanic) education (<High school or GED, high school graduate, some college, ≥college degree), and receipt of state or federal financial assistance (yes/no). Tobacco history variables included wave 4 measures of cigarettes smoked per day (1–5, 6–10, ≥10), smoking frequency (daily, non-daily), use of nicotine replacement therapy or a prescription cessation aid in the past 12-months (yes, no), and past 30-day use of other tobacco products (yes/no use of hookah, pipe tobacco, cigars/cigarillos, smokeless tobacco).
2.3. Analysis
Latent class analysis (LCA) was used to identify distinct sub-groups of e-cigarette users based on self-reported motivations for using e-cigarettes. LCA is a person-centered statistical method based on conditional probabilities that classifies participants into discrete categories (i.e., latent classes) based on similar individual response patterns to categorical variables. We used an increasing number of classes and model fit statistics to identify the optimal number of e-cigarette user sub-groups (Collins and Lanza, 2009). The model fit statistics used included Bayesian information criteria (BIC), Akaike information criteria (AIC), adjusted BIC (aBIC), consistent AIC (cAIC), and entropy. We used the SAS procedure PROC LCA (Lanza et al., 2007), which incorporates a full information maximum likelihood approach to account for missing data (missing <1 % for all variables included in LCA model). To determine the optimal number of classes, we considered the iteration with the lowest BIC/AIC/aBIC/cAIC and highest entropy (range 0–1.0, higher values representing more distinguishable groups) that also produced groups that could be assigned interpretable and meaningful class labels.
Upon identification of the optimal number of e-cigarette user sub-groups, posterior probabilities were used to classify participants into the most likely class membership (i.e., the sub-group for which they have the highest posterior probability, also known as the maximum-probability assignment rule). We examined the distribution and average of the posterior probabilities to assess the confidence in true class membership (Lanza et al., 2007). We then examined whether e-cigarette user sub-groups were associated with e-cigarette use frequency at wave 4 and cigarette smoking abstinence at wave 5. We fit separate logistic regression models examining the association of latent classes (independent variable) with daily e-cigarette use at wave 4 (dependent variable) and the association of latent classes (independent variable) with smoking abstinence at wave 5 (dependent variable).
In a separate series of logistic regression models, we examined the association of daily vs. nondaily e-cigarette use at wave 4 (independent variable) with smoking abstinence at wave 5 (dependent variable), before and after adjustment for the latent classes. We created a causal diagram (i.e., directed acyclic graph) to determine covariate adjustment for each of the aforementioned models and to avoid over-adjustment of potential mediators (Fig. 1)(Greenland, 2003; Greenland et al., 1999). Models adjusted for all covariates except for wave 4 smoking frequency and smoking intensity, which were assumed to be key mediators of vaping and smoking abstinence. In secondary analyses, we examined the association of individual reasons for using e-cigarettes with smoking abstinence at wave 5. We also compared wave 4 covariates, vaping motivations, and vaping frequency between those retained and lost to follow-up between waves 4 and 5.
Fig. 1.
Directed acyclic graph for the association of latent class motivations for vaping, vaping frequency, smoking cessation, and covariates. Bolded arrows represent associations of interest that are examined in models. RQ1 examines the association of motivations for vaping with vaping frequency, RQ2 examines the association of motivations for vaping and smoking cessation, and RQ3 examines the association of vaping frequency and smoking cessation, accounting for motivations for vaping, RQ=Research Question; SES=Socioeconomic Status; W4 =Wave 4; W5 = Wave 5.
All analyses were weighted by PATH survey sample weights with standard errors computed using the Balanced Repeated Replication method and Fay adjustment (p = 0.03). Analyses using the full sample of dual users at wave 4 regardless of retention at wave 5 were weighted using Wave 4 cross-sectional weights, while longitudinal analyses of wave 5 smoking outcomes were weighted using wave 5 longitudinal sample weights. PATH survey weights re-weight the sample to be representative of the U.S. civilian population based on inverse probabilities of survey selection and non-response (see PATH User Guide for detailed description of weighting procedures) (United States Department of Health and Human, 2021). This study involved secondary analyses of publicly available de-identified data, and thus does not constitute human subjects research.
3. Results
Among 1544 participants who were dual users of e-cigarettes and cigarettes at wave 4, 375 (24.3 %) used e-cigarettes daily, and 1169 (75.7 %) used e-cigarettes non-daily. Table 1 presents descriptive statistics of the total sample and stratified by e-cigarette use frequency. There were substantial differences in demographics and smoking behaviors between daily and non-daily vapers. Compared to participants who vaped daily, participants who vaped non-daily were more likely to be female, to receive government financial assistance, to be daily cigarette smokers, to smoke ≥ 10 cigarettes per day, and to have used other tobacco products in the past 30-days.
Table 1.
Descriptive Statistics of Covariates and Reasons for Using E-cigarettes Among 1,544 Dual Users of E-cigarettes and Cigarettes at Wave 4.
| Covariate measured at wave 4 | Full Sample (n=1,544) N(%) |
Daily E-cigarette User (n=375) N(%) |
Non-Daily E-cigarette User (n=1,169) N(%) |
p-valuea |
|---|---|---|---|---|
|
| ||||
| Age (years) | 0.8666 | |||
| 18–24 | 497 (21.0) | 120 (19.4) | 377 (21.6) | |
| 25–34 | 416 (29.3) | 101 (29.3) | 315 (29.4) | |
| 35–54 | 460 (36.7) | 114 (38.2) | 346 (36.2) | |
| ≥55 | 171 (12.9) | 40 (13.1) | 131 (11.2) | |
| Female Sex | 761 (46.1) | 152 (38.2) | 609 (48.9) | 0.0048 |
| Race | 0.1345 | |||
| White | 1,224 (81.0) | 308 (84.8) | 916 (79.6) | |
| Black | 140 (9.3) | 25 (6.9) | 115 (10.2) | |
| Another race/multiple races | 180 (9.7) | 42 (8.3) | 138 (10.2) | |
| Hispanic | 200 (10.7) | 39 (9.2) | 161 (11.2) | 0.3125 |
| Education | 0.8168 | |||
| <High school or GED | 388 (23.7) | 91 (24.7) | 297 (23.4) | |
| High school graduate | 423 (28.7) | 98 (26.8) | 325 (29.3) | |
| Some College | 584 (37.6) | 143 (37.9) | 441 (37.5) | |
| ≥College degree | 149 (10.1) | 43 (10.7) | 106 (9.8) | |
| Financial assistance | 431 (28.6) | 83 (21.9) | 348 (31.1) | 0.0104 |
| Past 12-month cessation aid use | 246 (17.0) | 67 (17.2) | 179 (17.0) | 0.9381 |
| Past 30-day use of other tobacco | 681 (39.9) | 146 (33.7) | 535 (42.1) | 0.0185 |
| Cigarettes smoked per day | <.0001 | |||
| 1–5 | 544 (32.5) | 196 (49.6) | 348 (26.4) | |
| 6–10 | 402 (45.6) | 87 (25.5) | 315 (25.7) | |
| ≥10 | 598 (41.9) | 92 (24.9) | 506 (47.9) | |
| Daily smoker | 1,026 (67.6) | 170 (46.9) | 856 (75.0) | <.0001 |
| Reasons for using e-cigarettes | ||||
| They are affordable | 912 (60.7) | 242 (63.4) | 670 (59.8) | 0.3391 |
| I can use electronic nicotine products at times when or in places where smoking cigarettes isn’t allowed. | 1,171 (77.4) | 289 (78.6) | 882 (77.0) | 0.6824 |
| They might be less harmful to me than smoking cigarettes. | 1,106 (72.5) | 314 (84.6) | 792 (68.2) | <.0001 |
| They might be less harmful to people around me than cigarettes. | 1,196 (78.6) | 315 (84.6) | 881 (76.5) | 0.0024 |
| E-liquid comes in flavors I like | 1,206 (78.3) | 309 (81.5) | 897 (77.2) | 0.1515 |
| Using electronic nicotine products helps people quit smoking cigarettes. | 1,047 (70.3) | 303 (83.2) | 744 (65.7) | <.0001 |
| Electronic nicotine products don’t smell. | 1,021 (68.2) | 269 (72.9) | 752 (66.6) | 0.0546 |
| Using an electronic nicotine product feels like smoking a regular cigarette. | 673 (46.5) | 180 (51.0) | 493 (44.9) | 0.0921 |
| Electronic nicotine products are more acceptable to nontobacco users. | 1,062 (69.8) | 268 (70.9) | 794 (69.3) | 0.6823 |
| Use electronic nicotine products as a way of cutting down on cigarette smoking | 1.039 (70.5) | 313 (84.6) | 726 (65.5) | <.0001 |
Estimates presented as unweighted counts and weighted percentages. Estimates weighted using PATH wave 4 cross-sectional weights.
p-value from chi-square tests for comparisons between vaping frequency groups.
The most commonly reported individual reasons for using e-cigarettes among the full sample were because they might be less harmful to other people (78.6 %), because e-cigarettes come in flavors liked by participants (78.3 %), and because they can be used at times or in places when smoking cigarettes is not allowed (77.4 %). The least commonly reported reason for using e-cigarettes among the full sample was because e-cigarettes feel like smoking a regular cigarette (46.5 %). Compared to participants who vaped non-daily, participants who vaped daily were more likely to report using e-cigarettes because they might be less harmful to themselves (84.6 % vs. 68.2 %), because they might be less harmful to others (84.6 % vs. 76.5 %), because they help people quit smoking (83.2 % vs. 65.7 %), and as a way to cut down on cigarette smoking (84.6 % vs. 65.5 %).
3.1. E-cigarette user motivation sub-groups
The LCA analysis indicated that a 4-class solution was the best fitting model based on fit statistics and group interpretability (see Supplemental Table 1 for fit statistics of each iteration). The 4-class solution had the lowest BIC and cAIC. In addition, the AIC and aBIC values reduced only negligibly after four classes. Although entropy was lower for a 4-class solution (0.73) compared to a 5-class solution (0.77), we found the 4-class solution could be assigned meaningful and interpretable sub-groups while still maintaining sufficiently large sample sizes in each class. The smallest class in the 5-class solution represented just 4 % (n = 60) of participants, compared to ~10 % (n = 149) in the smallest class of the 4-class solution (see Supplemental Table 2 for 5-class solution conditional probabilities). The following four e-cigarette user motivation sub-groups were identified: (1) Convenience and Social Acceptability (20.5 %), (2) Vaping Enthusiasts (54.0 %), (3) Experimenters (9.6 %), and (4) Quitting Smoking and Harm Reduction (15.9 %) (Table 2).
Table 2.
Conditional Probabilities of Reasons for Using E-cigarettes by Each Latent Class.
| Convenience and Social Acceptability | Vaping Enthusiasts | Experimenters | Quitting Smoking and Harm Reduction | |
|---|---|---|---|---|
|
| ||||
| Unweighted percentage among analytic sample based on posterior probability | 0.205 | 0.540 | 0.096 | 0.159 |
| Mean posterior probability | 0.836 | 0.880 | 0.911 | 0.757 |
| Membership probability | 0.202 | 0.524 | 0.085 | 0.188 |
| Reasons | ||||
| They are affordable | 0.525 | 0.780 | 0.052 | 0.478 |
| Can use them at times or in places where smoking cigarettes are not allowed | 0.832 | 0.899 | 0.280 | 0.595 |
| It might be less harmful to me than cigarettes | 0.416 | 0.968 | 0.042 | 0.704 |
| They might be less harmful to people around me than cigarettes | 0.677 | 0.995 | 0.055 | 0.660 |
| Using them helps people quit smoking cigarettes | 0.162 | 0.938 | 0.079 | 0.926 |
| They come in flavors I like | 0.754 | 0.886 | 0.381 | 0.717 |
| They don’t smell | 0.735 | 0.810 | 0.179 | 0.513 |
| They feel like smoking a regular cigarette | 0.428 | 0.585 | 0.054 | 0.361 |
| They are more acceptable to nontobacco users | 0.711 | 0.903 | 0.086 | 0.396 |
| Use as a way to cut down cigarette smoking | 0.331 | 0.875 | 0.173 | 0.885 |
Conditional probabilities >0.60 for variables included in LCA model are bolded for interpretability
The Convenience and Social Acceptability class had high probabilities of reporting using e-cigarettes because they can use them at times when cigarettes are not allowed (83 %), because they don’t smell (74 %), because they may be less harmful to others (68 %) and more acceptable to others than cigarettes (71 %), and because of the flavor offerings (75 %). The Vaping Enthusiast class had high probabilities (78–99 %) of endorsing nearly all potential reasons for using e-cigarettes, including harm reduction and quitting smoking, affordability, flavors, and convenience and social acceptability reasons. The lowest conditional probability in the Vaping Enthusiast class was for endorsing using e-cigarettes because they feel like smoking a regular cigarette (59 %). The Experimenter class had low probabilities (<39 %) for all 10 reasons for using e-cigarettes. The highest conditional probability was for using e-cigarettes because they come in appealing flavors (38 %). The Quitting Smoking and Harm Reduction class had very high conditional probabilities for using e-cigarettes because they help people quit smoking (93 %) and to cut down on their own cigarette smoking (89 %). They also had relatively high probabilities for reporting using e-cigarettes because they are less harmful to themselves (70 %) and others (66 %), and because of appealing flavors (72 %).
3.2. E-cigarette user motivation sub-groups and E-cigarette use frequency
Table 3 presents results examining the association between the motivation sub-groups and e-cigarette use frequency at wave 4. Compared to participants in the Convenience and Social Acceptability sub-groups, participants in the Vaping Enthusiast and Quitting Smoking and Harm Reduction sub-groups were 2–3 times more likely to use e-cigarettes daily (vs. non-daily). In adjusted models, participants in the Convenience and Social Acceptability sub-group were 59 % less likely to vape daily compared to those in the Quitting Smoking and Harm Reduction sub-group (OR=0.41, 95 % CI: 0.24–0.71), and 71 % less likely to vape daily compared to those in the Vaping Enthusiast sub-group (OR=0.29, 95 % CI: 0.20–0.42). Participants in the Vaping Enthusiast group were also more likely than those in the Quitting Smoking and Harm Reduction (OR=1.45, 95 % CI: 0.99–2.11) and Experimenter (OR=2.01, 95 % CI: 1.15–3.51) classes to vape daily.
Table 3.
Association between E-cigarette Motivation Sub-Groups and Wave 4 Daily Vaping Among 1,544 Dual Users of E-cigarettes and Cigarettes at Wave 4.
| Convenience and Social Acceptability | Vaping Enthusiasts | Experimenters | Quitting Smoking and Harm Reduction | |
|---|---|---|---|---|
|
| ||||
| n(%) of total analytic sample | 316 (20.5) | 834 (54.0) | 149 (9.6) | 245 (15.9) |
| n(%) vape daily | 36 (11.8) | 257 (32.6) | 26 (19.3) | 56 (25.8) |
| Unadjusted Models | ||||
| OR (95% CI) | Reference | 3.64 (2.54–5.22) | 1.80 (0.94–3.45) | 2.61 (1.57–4.34) |
| OR (95% CI) | 0.28 (0.19–0.39) | Reference | 0.50 (0.29–0.86) | 0.72 (0.49–1.05) |
| OR (95% CI) | 0.45 (0.29–1.07) | 2.02 (1.17–3.49) | Reference | 1.45 (0.74–2.84) |
| OR (95% CI) | 0.38 (0.23–0.64) | 1.39 (0.95–2.05) | 0.69 (0.35–1.35) | Reference |
| Adjusted Models a | ||||
| OR (95% CI) | Reference | 3.50 (2.40–5.11) | 1.74 (0.88–3.47) | 2.42 (1.40–4.19) |
| OR (95% CI) | 0.29 (0.20–0.42) | Reference | 0.50 (0.29–0.87) | 0.59 (0.49–1.01) |
| OR (95% CI) | 0.57 (0.29–1.14) | 2.01 (1.15–3.51) | Reference | 1.39 (0.70–2.76) |
| OR (95% CI) | 0.41 (0.24–0.71) | 1.45 (0.99–2.11) | 0.72 (0.36–1.43) | Reference |
Outcome is Daily Vaping (1) vs. Non-Daily Vaping (0). Percentages and model estimates are weighted using W4 cross-sectional survey sample weights. Table presents row comparisons in which models were re-tested four times switching the sub-group reference group to capture all pairwise comparisons.
Adjusted for wave 4 values of age, sex, race, ethnicity, financial assistance, past 30-day use of other tobacco products, and past 12-month use of prescription cessation aids.
3.3. E-cigarette User Motivation Sub-Groups and Cigarette Smoking Abstinence at Wave 5
Among 1544 participants at wave 4, 1244 (81 %) had wave 5 follow-up data. Participants lost to follow-up between wave 4 and wave 5 were younger than retained participants; there were few other differences in covariates, motivations, or vaping frequency between participants retained vs. lost to follow-up (Supplemental Table 3). Among participants with follow-up data, 243 (19.5 %) reported smoking abstinence at Wave 5 (i.e., reported not smoking cigarettes at all at the time of the survey) and 668 (53.7 %) were still vaping. Results from the crude and multivariable logistic regression models either indicated no association between user sub-groups and smoking abstinence (i.e., ORs close to 1.0), or were imprecise estimates (i.e., wide confidence intervals) (Table 4).
Table 4.
Association between E-cigarette Motivation Sub-Groups and Wave 5 Smoking Abstinence Among 1,244 Dual Users of E-cigarettes and Cigarettes at Wave 4 with Wave 5 Follow-up Data.
| Convenience and Social Acceptability | Vaping Enthusiasts | Experimenters | Quitting Smoking and Harm Reduction | |
|---|---|---|---|---|
|
| ||||
| n(%) of total analytic sample | 264 (20.3) | 658 (54.6) | 119 (8.5) | 203 (16.6) |
| n(%) smoking abstinent | 47 (16.6) | 137 (19.5) | 28 (21.0) | 31 (15.1) |
| Unadjusted Models | ||||
| OR (95% CI) | Reference | 1.22 (0.80–1.86) | 1.34 (0.72–2.47) | 0.89 (0.50–1.57) |
| OR (95% CI) | 0.82 (0.54–1.26) | Reference | 1.10 (0.61–1.97) | 0.73 (0.46–1.16) |
| OR (95% CI) | 0.75 (0.40–1.38) | 0.91 (0.51–1.63) | Reference | 0.67 (0.34–1.29) |
| OR (95% CI) | 1.13 (0.64–1.99) | 1.37 (0.86–2.18) | 1.50 (0.77–2.92) | Reference |
| Adjusted Models a | ||||
| OR (95% CI) | Reference | 1.12 (0.72–1.73) | 1.24 (0.64–2.43) | 0.88 (0.49–1.60) |
| OR (95% CI) | 0.90 (0.49–1.39) | Reference | 1.12 (0.58–2.15) | 0.79 (0.49–1.27) |
| OR (95% CI) | 0.80 (0.41–1.57) | 0.90 (0.47–1.73) | Reference | 0.71 (0.35–1.45) |
| OR (95% CI) | 1.13 (0.63–2.05) | 1.26 (0.79–2.03) | 1.41 (0.69–2.87) | Reference |
Outcome is no current smoking at wave 5 (1) vs. current smoking (0). Percentages and model estimates weighted by Wave 5 longitudinal survey sample weights. Table presents row comparisons in which models were re-tested four times switching the sub-group reference group to capture all pairwise comparisons.
Adjusted for wave 4 values of age, sex, race, ethnicity, financial assistance, past 30-day use of other tobacco products, and past 12-month use of prescription or nicotine replacement therapy cessation aids.
3.4. E-cigarette use frequency and cigarette smoking abstinence at wave 5
Compared to non-daily vaping, daily vaping was positively associated with smoking abstinence at wave 5 after adjusting for covariates but prior to adjustment for motivation sub-groups (adjusted OR: 1.42, 95 % CI: 0.97–2.08) (Table 5). Confidence intervals were consistent with a range of associations, including none. There was no meaningful change to the odds ratio after adjustment for motivation sub-groups (adjusted OR: 1.41, 95 % CI: 0.96–2.06).
Table 5.
Association between Vaping Frequency and Wave 5 Smoking Abstinence Among 1,244 Dual Users of E-cigarettes and Cigarettes at Wave 4 with Wave 5 Follow-up Data: Before and After Adjustment for E-cigarette Motivation Sub-Groups.
| N(%) Smoking Abstinence W5 | Unadjusted OR | Adjusted ORa | Adjusted ORb | |
|---|---|---|---|---|
|
| ||||
| Daily Vaping | 72 (23.7) | 1.39 (0.96–2.02) | 1.42 (0.97–2.08) | 1.41 (0.96–2.06) |
| Non-Daily Vaping | 171 (17.0) | 1.0 (REF) | 1.0 (REF) | 1.0 (REF) |
Outcome is no current smoking at wave 5 (1) vs. current smoking (0). Percentages and model estimates weighted by Wave 5 longitudinal survey sample weights
Adjusted for wave 4 values of age, sex, race, ethnicity, financial assistance, other tobacco product use, cessation aid use.
Adjusted for wave 4 values of age, sex, race, ethnicity, financial assistance, other tobacco product use, cessation aid use, vaping motivation sub-groups.
3.5. Individual reasons for E-cigarette use and cigarette smoking abstinence at wave 5
There was either no association between individual reasons for e-cigarette use and smoking abstinence (i.e., ORs close to 1.0), or estimates were too imprecise to make meaningful conclusions (i.e., wide confidence intervals) (Supplemental Table 4).
4. Discussion
In this study of adults who were dual users of e-cigarettes and cigarettes, four distinct sub-groups of reasons for using e-cigarettes emerged. The most prevalent sub-group had high conditional probabilities for nearly all surveyed reasons for using e-cigarettes. The second most common sub-group represented using e-cigarettes for convenience or social acceptability reasons, followed by those who reported using e-cigarettes specifically for harm reduction and smoking reduction, and finally those who represented experimental use. Participants who used e-cigarettes because of convenience or social acceptability were less frequent e-cigarette users than those who used e-cigarettes for smoking reduction. However, e-cigarette user sub-groups were not associated with cigarette smoking abstinence after one-year of follow-up.
It is well documented that many adults use e-cigarettes to help quit or reduce their cigarette smoking (Benmarhnia et al., 2018; Pepper and Brewer, 2014; Spears et al., 2018; Tackett et al., 2015). This study enhances the literature by demonstrating that there is a sub-group of adults who primarily vape for convenience (i.e., when cigarettes are not allowed) and because e-cigarettes are more acceptable to nontobacco users than cigarettes, rather than to quit smoking. Participants in this Convenience and Social Acceptability class had low probabilities of using e-cigarettes because they help people quit smoking and to cut down on cigarette smoking, but relatively high probability of using e-cigarettes because they may be less harmful to those around them than cigarettes. Interventions or education campaigns may seek to target individuals who vape for convenience or social acceptability, given that the discreteness of e-cigarettes and the ability to vape indoors have been identified as primary drivers of the popularity of pod-based e-cigarettes among young adults (Barrington-Trimis and Leventhal, 2018). Although secondhand e-cigarette aerosol is likely safer than secondhand cigarette smoke, emerging research demonstrates that vaping indoors leads to passive exposure to fine particulate matter and airborne toxic chemicals (e.g., aldehydes and nicotine)(Son et al., 2020). Additionally, biomarker studies indicate that some benefits of switching from combustible cigarettes to e-cigarettes are likely contingent on complete substitution (Mainous et al., 2020; Xie et al., 2020). Therefore, interventions targeting individuals who vape for convenience or social acceptability could focus on changing vaping motivations towards using e-cigarettes as a complete substitute for combustible cigarettes, rather than a product to only use when combustible cigarettes are not allowed or tolerated.
Results of this study indicate that vaping frequency may be associated with underlying motivations for e-cigarette use. Participants who reported using e-cigarettes because they are less harmful than cigarettes and to cut down on their own smoking were more likely to be daily e-cigarette users than participants who reported using e-cigarettes for convenience or social acceptability reasons. In prior observational studies, vaping daily is positively associated with smoking abstinence, while non-daily vaping is inversely associated with smoking abstinence relative to no e-cigarette use (Berry et al., 2019; Biener and Hargraves, 2015; Brose et al., 2015; Verplaetse et al., 2019). The current study provides preliminary evidence that e-cigarette use frequency may at least partially represent underlying sub-groups and motivations of e-cigarette users.
Although we found that e-cigarette use motivations were associated with e-cigarette use frequency, motivations were not prospectively associated with smoking abstinence one-year later. A large majority (80–85 %) of participants in the two latent classes with high probabilities of using e-cigarettes to quit or reduce smoking continued cigarette smoking at follow-up. Additionally, consistent with prior literature, we found a positive (albeit imprecise) association between daily vaping (vs. nondaily vaping) and smoking cessation after one year. However, further adjustment for e-cigarette motivations did not change the odds ratio for the association of daily vaping with cigarette smoking abstinence. Our results are inconsistent with at least two prior studies that found participants who report using e-cigarettes for quitting smoking were more likely to quit or reduce cigarette smoking than those who use e-cigarettes for reasons other than quitting (Vickerman et al., 2017; Rutten et al., 2015) Inconsistencies may be due to the fact that prior studies were conducted during earlier time-frames (e.g., 2013–2015), before substantial changes to vaping products and policies which may impact the effectiveness of e-cigarettes for smoking cessation (e.g., flavor restrictions, highly potent nicotine-salt e-liquid). Inconsistencies may also be due to the use of different study populations (e.g., U.S. population vs. quit-line data.) Assuming internal validity of the current study, our findings indicate that underlying motivations may not be an important source of confounding in analyses examining the association of vaping frequency with cigarette smoking abstinence.
There are limitations to this work. We did not adjust for smoking frequency or intensity at wave 4 because these variables are likely key mediators of the relationship between vaping at wave 4 and smoking cessation at wave 5. However, there may be residual confounding by smoking or vaping history prior to wave 4, including by duration of e-cigarette or cigarette use, nicotine dependence, or past quit attempts. The temporal relationship between vaping frequency and vaping motivations is unclear (i.e., whether motivations impact vaping frequency or vice versa). Furthermore, there may be reasons for using e-cigarettes that were not included on the PATH survey, such as recreational or social motivations. This is a concern given that one of the motivation sub-groups in our data (“Experimenters”) was poorly defined given low probabilities of all the potential reasons for using e-cigarettes. Finally, there was some loss-to-follow up (19 %) between wave 4 and wave 5 surveys, and there is the potential for selection bias due to differential attrition in analyses examining smoking abstinence. However, PATH survey sample weights account for survey nonresponse and there were little differences in covariates between those retained and lost to follow-up.
5. Conclusions
In this nationally representative study of adults who smoke cigarettes and use e-cigarettes, we identified four sub-groups of motivations for using e-cigarettes that were associated with vaping frequency. Participants who vaped for convenience and social acceptability were less frequent e-cigarette users than participants with other vaping motivations, such as harm reduction. However, vaping motivations were not associated with cigarette smoking abstinence at the one-year follow-up and did not explain an association between vaping frequency and smoking abstinence. Findings indicate that among individuals who smoke cigarettes, those who vape daily have different motivations for using e-cigarettes than those who vape non-daily, which should be considered when interpreting past and future studies on associations of e-cigarette use frequency and smoking abstinence.
Supplementary Material
Role of Funding Source
This work was supported by the National Cancer Institute at the National Institutes of Health and the Food and Drug Administration (country: United States; grant number: 1U54CA180905). The funders had no role in the design and conduct of the study; collection, management, analysis, or interpretation of the data; or preparation, review, or approval of the manuscript.
Footnotes
Conflict of Interest
No conflict declared.
CRediT authorship contribution statement
Dr. Harlow conducted statistical analyses and drafted and revised the manuscript, Dr. Cho provided consultation to statistical analyses and reviewed/revised the manuscript, Drs. Tackett, McConnel, Stokes, Leventhal, and Barrington-Trimis aided in interpretation of results and critically reviewed/revised the manuscript, All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
Declaration of interests
None.
Appendix A. Supporting information
Supplementary data associated with this article can be found in the online version at doi:10.1016/j.drugalcdep.2022.109583.
References
- Barrington-Trimis JL, Leventhal AM, 2018. Adolescents’ Use of “Pod Mod” E-cigarettes — urgent concerns. New Engl. J. Med. 379, 1099–1102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Benmarhnia T, Pierce JP, Leas E, White MM, Strong DR, Noble ML, Trinidad DR, 2018. Can E-cigarettes and pharmaceutical AIDS increase smoking cessation and reduce cigarette consumption? findings from a nationally representative cohort of American smokers. Am. J. Epidemiol. 187, 2397–2404. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Berry KM, Reynolds LM, Collins JM, et al. , 2019. E-cigarette initiation and associated changes in smoking cessation and reduction: the population assessment of tobacco and health study, 2013–2015. Tob. Control 28, 42–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Biener L, Hargraves JL, 2015. A longitudinal study of electronic cigarette use among a population-based sample of adult smokers: association with smoking cessation and motivation to quit. Nicotine Tob. Res. 17, 127–133. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Borland R, Murray K, Gravely S, Fong GT, Thompson ME, McNeill A, O’Connor RJ, Goniewicz ML, Yong HH, Levy DT, Heckman BW, Cummings KM, 2019. 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. Addiction 114, 24–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- United States Department of Health and Human Services. National Institutes of Health. National Institute on Drug Abuse, and United States Department of Health and Human Services. Food and Drug Administration. Center for Tobacco Products., 2021. Population Assessment of Tobacco and Health (PATH) Study [United States] Public-Use Files User Guide. Inter-university Consortium for Political and Social Research [distributor], 2021-December-16. 10.3886/ICPSR36498.v16. [DOI] [Google Scholar]
- Brose LS, Hitchman SC, Brown J, et al. , 2015. Is the use of electronic cigarettes while smoking associated with smoking cessation attempts, cessation and reduced cigarette consumption? A survey with a 1-year follow-up. Addiction 110, 1160–1168. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Collins LM, Lanza ST, 2009. Latent class and latent transition analysis: with applications in the social behavioral, and health sciences. John Wiley & Sons,. [Google Scholar]
- Evans-Polce RJ, Patrick ME, Lanza ST, Johnston LD, et al. , 2018. Reasons for vaping among U.S. 12th graders. J. Adolesc. Health 62, 457–462. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Greenland S, 2003. Quantifying biases in causal models. Epidemiology 14, 300–306. [PubMed] [Google Scholar]
- Greenland S, Pearl J, Robins JM, 1999. Causal diagrams for epidemiologic research. Epidemiology 10, 37–48. [PubMed] [Google Scholar]
- Hyland A, Ambrose BK, Conway KP, et al. , 2017. Design and methods of the Population Assessment of Tobacco and Health (PATH) Study. Tob. Control 26, 371–378. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kinouani S, Leflot C, Vanderkam P, et al. , 2020. Motivations for using electronic cigarettes in young adults: a systematic review. Subst. Abus. 41, 315–322. [DOI] [PubMed] [Google Scholar]
- Lake S, Nosova E, Buxton J, et al. , 2020. Characterizing motivations for cannabis use in a cohort of people who use illicit drugs: a latent class analysis. PLOS ONE 15, e0233463. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lanza ST, Collins LM Lemmon, et al. , 2007. PROC LCA: A SAS procedure for latent class analysis. Struct. Equ. Model.: a Multidiscip. J. 14, 671. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mainous AG, Yadav S, Hong YR, et al. , 2020. e-cigarette and conventional tobacco cigarette use, dual use, and C-reactive protein. J. Am. Coll. Cardiol. 75, 2271–2273. [DOI] [PubMed] [Google Scholar]
- Martínez Ú, Martínez-Loredo V, Simmons VN, et al. , 2020. How does smoking and nicotine dependence change after onset of vaping? a retrospective analysis of dual users. Nicotine Tob. Res. 22, 764–770. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Morgan Snell L, Barnes AJ, Nicksic NE, 2020. A longitudinal analysis of nicotine dependence and transitions from dual use of cigarettes and electronic cigarettes: Evidence from waves 1–3 of the path study. J. Stud. Alcohol Drugs 81, 595–603. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pepper JK, Brewer NT, 2014. Electronic nicotine delivery system (electronic cigarette) awareness, use, reactions and beliefs: a systematic review. Tob. Control 23, 375–384. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rutten LJF, Blake KD, Agunwamba AA, et al. , 2015. Use of E-cigarettes among current smokers: associations among reasons for use, quit intentions, and current tobacco use. Nicotine Tob. Res. 17, 1228–1234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Son Y, Giovenco DP, Delnevo C, et al. , 2020. Indoor air quality and passive e-cigarette aerosol exposures in vape-shops. Nicotine Tob. Res. 22, 1772–1779. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Spears CA, Jones DM, Weaver SR, et al. , 2018. Motives and perceptions regarding electronic nicotine delivery systems (ENDS) use among adults with mental health conditions. Addict. Behav. 80, 102–109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Sutton SK, Brandon KO, Harrell PT, et al. , 2022. Identifying prospective subpopulations of combustible and electronic cigarette dual users in the United States via finite mixture modeling. Addict. Online Print. 10.1111/add.15906. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tackett AP, Lechner WV, Meier E, et al. , 2015. Biochemically verified smoking cessation and vaping beliefs among vape store customers. Addiction 110, 868–874. [DOI] [PubMed] [Google Scholar]
- Verplaetse T, Moore K, Pittman B, et al. , 2019. Intersection of E-cigarette use and gender on transitions in cigarette smoking status: findings across waves 1 and 2 of the population assessment of tobacco and health study. nicotine & tobacco research: official journal of the society for research on nicotine and tobacco, 21, 1423–1428. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Vickerman KA, Schauer GL, Malarcher AM, et al. , 2017. Reasons for electronic nicotine delivery system use and smoking abstinence at 6 months: a descriptive study of callers to employer and health plan-sponsored quitlines. Tob. Control 26, 126–134. [DOI] [PubMed] [Google Scholar]
- Votaw VR, Mchugh RK, Witkiewitz K, 2019. Substance use & misuse alcohol use disorder and motives for prescription opioid misuse: a latent class analysis Alcohol use disorder and motives for prescription opioid misuse: a latent class analysis. Subst. Use Misuse 54, 1558–1568. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Xie W, Wilson AE, Harlow AF, et al. , 2020. Association of cigarette and electronic cigarette use patterns with levels of inflammatory and oxidative stress biomarkers among US adults: population assessment of tobacco and health study. Circulation 142, 1–3. [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.