Abstract
Aims:
To understand dual users’ cigarette and e-cigarette use patterns including the contexts in which they vape vs smoke and to understand how environmental and internal contexts and smoking patterns differ between dual users and exclusive smokers.
Design:
Longitudinal observational trial.
Participants:
Adult dual users (n=162) and adults who exclusively smoked (n=143), with no plans to quit smoking or vaping in the next 30 days
Setting:
Research center.
Measurements:
Participants carried smartphones for 2 weeks at baseline to record each use event for the two products and report on the context of their product use.
Findings:
Hierarchical linear regression models with random slopes and intercepts examined the within- and between-subject effects of context on the likelihood of vaping (vs. smoking); significant fixed effects were tested for moderation by e-cigarette dependence (based on first product used in the morning). Dual users reported significantly more puffs/cigarette (M=13.1, SD=10.2) than puffs/vape event (M=11.7, SD=11.5; p=0.01). E-cigarette dependence moderated the influence of social cues (t-ratio=2.4, p=0.02) and smoking restrictions (t-ratio=3.1, p=0.003) on the likelihood of vaping versus smoking (OR=2.30, p=0.02). Context was more related to which product was used in those of low versus higher e-cigarette dependence. Reports of strong cravings to smoke and positive expectancies for cigarettes were associated with reduced likelihood of vaping whereas strong cravings to vape and positive vaping expectancies were related to increased likelihood of vaping.
Conclusions:
Vaping was related to internal cues and more highly linked with social contexts and smoking restrictions (i.e., under stronger external stimulus control) in dual users of low to moderate vs. high e-cigarette dependence. These findings illustrate the importance of contextual factors in tobacco product use amongst dual users with the influence of context being reduced at high levels of e-cigarette dependence.
Keywords: e-cigarettes, ENDS, dependence, cigarettes, vaping, contexts, cues
INTRODUCTION
Since their introduction into the U.S. in 2008, e-cigarettes have become highly disruptive products with potential for significant public health effects. E-cigarette policy decisions would be informed by additional knowledge about factors that affect e-cigarette use and dependence. One knowledge gap highlighted by former CDC Director Tom Frieden, is that we do not know how the use of e-cigarettes influences the use of commercial, combustible cigarettes [1]. This is of public health importance since dual use might cause people who smoke to develop a dual use pattern (using both products) that prolongs their use of combustible cigarettes. Or, e-cigarette use, or “vaping”, might enhance public health by serving as a smoking cessation aid or via harm reduction to the extent that e-cigarette use displaces smoking and reduces smoking-related toxicant exposure [2].
Researchers have studied numerous facets of e-cigarettes and their effects even as e-cigarettes themselves have changed markedly (e.g., the development of nicotine salt [3] products): the content of aerosols [4, 5]; prevalence rates in the general population and by demographic factors [6, 7]; the dynamic profiles of nicotine delivery [8, 9]; the subjective effects of e-cigarette use [10]; whether e-cigarettes are a gateway product for youth [11]; e-cigarette dependence [12, 13]; and the ability of e-cigarettes to help people who smoke to quit using combustible tobacco [14, 15]. Very little research, however, has addressed how contextual features might affect combustible versus electronic cigarette use.
A large body of research shows that contexts such as smoking bans, exposure to others smoking, and internal contexts, can powerfully affect addictive drug use, including cigarette use [16–20]. Moreover, the extent to which stimuli or contexts influence drug use (i.e., stimulus control), may index or reflect the type of dependence. For instance, there is evidence that cigarette dependence has two major dimensions: primary and secondary dependence. Primary dependence (characterized by drug use that is automatic, relatively heavy, accompanied by strong urges, and perceptions of loss of control [20–21]), is associated with fairly continuous smoking versus intermittent and spurred by discrete cues [23: cf stereotypy [17]). Secondary dependence is smoking that is instrumental (e.g., to influence affect or cognition), and is relatively cue dependent: i.e., influenced by contexts such as social situations [21, 22, 24]. Some evidence suggests that secondary dependence is a developmental phase that may or may not transition to primary dependence [25, 26]. The relative strength of these types of dependence confers information regarding the persistence of smoking [21–23, 26, 27].
The current research uses real-time assessments of the use of cigarettes and e-cigarettes amongst exclusive smokers and dual users to determine the extent the use of these products is under stimulus control (i.e., is contextually contingent). We compare dual users’ and exclusive smokers’ smoking patterns (e.g., duration, number of puffs) and the influence of smoking prohibitions on smoking and vaping. We then use data from dual users to compare the association of smoking vs. vaping with external and internal cues, as an index of stimulus control. We further explore whether e-cigarette dependence accounts for variance in degree of contextual influence in product use, using a measure of relative dependence (which product is used first in the morning) that is especially strongly associated with likelihood of future vaping and with primary dependence [13]. Finally, we compare daily product use contexts between exclusive smokers and dual users and then examine whether dual users’ contexts differ by e-cigarette dependence. Such data may constitute an innovative and theoretically meaningful index of the nature and intensity of dependence on cigarettes and e-cigarettes amongst dual users.
METHOD
People who smoke commercial combustible cigarettes exclusively and dual users of combustible and e-cigarettes were recruited for this longitudinal observational trial via television and Facebook advertisements in the greater Madison and Milwaukee, WI areas from October 2015 to July, 2017. Eligible participants were ≥ 18 years old, able to read and write English, had no plans to quit smoking and/or e-cigarette use in the next 30 days, were not currently using smoking cessation medication, and were not currently in treatment for psychosis or bipolar disorder. Dual users had to have used nicotine-containing e-cigarettes at least once a week for the past month and have smoked daily for the last 3 months. Initially, dual users were required to smoke a minimum of 5 cigarettes/day but to increase recruitment, approximately 6 months into our 2-year recruitment we revised this criterion so that dual users just needed to have smoked daily for the last 3 months (Note: there was no statistically significant difference in combustible or e-cigarette dependence among participants who were recruited using the initial or revised criterion). Participants in the exclusive smoking group had to report smoking at least 5 cigarettes per day for the past 6 months and no use of e-cigarettes within that time.
Participants provided informed written consent and completed baseline assessments of demographics, smoking and e-cigarette history, cigarette and e-cigarette use patterns, and beliefs about cigarettes and e-cigarettes. All participants completed measures of combustible cigarette dependence (the FTCD [28]; the Brief WISDM [26, 29], and the PS-CDI [30]). Dual users also completed three parallel e-cigarette dependence measures (the e-FTCD, the e-WISDM [13], and the PS-ECDI [30]). Participants carried a smartphone to record each cigarette and e-cigarette use event for 2 weeks as part of the ecological momentary assessment (EMA) protocol. Using these data, e-cigarette dependence was calculated based on which product, cigarettes or e-cigarettes, was used first in the morning; i.e., if e-cigarettes were always used first, the product use ratio = 100%. This was categorized as: high e-cigarette dependence=vaping first on ≥50% of mornings; moderate e-cigarette dependence=vaping first on <50% of mornings; and low e-cigarette dependence=always smoking first in the morning [13].
Smoking and vaping contexts were assessed for 2 weeks at baseline via EMA on a semi-random basis with at least 2 product use reports per product per day triggering assessments of contexts. The EMA prompts assessed the presence of a smoking ban, social contexts (i.e., being around others smoking or vaping), and internal cues (positive and negative affect, craving, difficulty concentrating). We have used the internal cue items in previous EMA research [31–33]; many items were adapted from established measures including the PANAS and the WSWS [34, 35]. Expectations of reward from product use were also assessed. Participants were also asked “How much time did you just spend vaping/smoking?” and “How many puffs did you just take from the cigarette/e-cig?”. Participants also completed an evening assessment prior to going to bed that assessed smoking restrictions, social contexts, and internal cues over the course of the day as well as daily ratings of anhedonia, stressors, overall smoking and vaping, and alcohol use. The evening assessments also queried attitudes towards vaping and exposure to vaping contexts; all participants, including exclusive smokers, completed these vaping items. See Appendix 1 for EMA items.
Analytic plan
Of the 256 dual users recruited for the study, only 162 (63.3%) reported at least one vape event and completed at least one random prompt during the baseline EMA and were therefore included in the analyses (i.e., only EMA-confirmed dual users were included in the analyses). Dual users who did not provide EMA data for analysis (n=94) were more likely to be African-American or multi-racial, have higher exhaled CO, vape much less frequently, and have lower e-WISDM PDM and Total scores compared to the participants who were included in the analyses (see Supplemental Tables 1 and 2). Of the 166 exclusive smokers recruited, 143 (86.1%) completed at least one random EMA prompt and were included in the analyses. Compared to the exclusive smokers who were included in the analyses, those who did not provide EMA data (n=23) reported lower levels of education, were older, and had smoked for longer (see Supplemental Tables 1 and 2).
Within the analytic sample (N=305), we examined group differences (exclusive smokers vs. dual users) for continuous variables reported during each EMA context assessment, including smoking duration, puffs/cigarette, internal contexts (e.g., affect) and external context (e.g., around smokers) as well as product differences in use duration and puffs amongst dual users. The main outcome for the context analyses was a dichotomous indicator of whether participants opted to smoke or vape, given the reported context. To account for the interdependence of EMA data within-subjects, hierarchical regression models were used to assess differences between dual users and smokers. Specifically, the main effects of context variables (e.g., affect, around smokers) on the probability of a use event being a vape event (vs. a smoking event) were analyzed using hierarchical logistic regression with random slopes and intercepts using Hierarchical Linear Modeling (HLM) Version 8.0. For significant fixed effects, we evaluated moderation effects through cross-level interactions between subject variables and within-subject (i.e., context) variables, and statistically tested through t-statistics derived from the ratio of the fixed effect cross-level interaction over its standard error. E-cigarette dependence was modeled as both as a continuous and a dummy coded categorical variable. The general model for the analyses examining the interactions is displayed in Figure 1. If models failed to converge, effects were modeled using random intercepts only. We used the Benjamini-Hochberg [36] approach to control the false discovery rate for families of predictors (i.e., for the analyses involving multiple internal and external contexts) but not for single context variables (e.g., being around someone vaping).
Figure 1.
Hierarchical logistic regression models with random intercepts and slopes to compute interactions
Descriptive summaries of outcomes where no inferential tests were conducted ignored the nested structure of the data. For analyses examining inferential differences between dual users and smokers or differences by dependence within dual users, person-level descriptive summaries were provided to account for the EMA data structure. Evening report data were examined to identify descriptive patterns and were analyzed by aggregating data into a single mean per person, followed by group comparisons using standard One-Way ANOVA or independent t-tests. While we provide descriptive summaries of variables to aid in interpretation in places, all inferential testing accounted for the nested nature of the data. These analyses were not pre-registered and should be considered exploratory.
RESULTS
Participants who smoked exclusively reported 118.3 (SD=87.3) smoking events and completed 98.3% of the context assessment prompts. Dual users reported 95.8 (SD=86.3) smoking events and 98.2 (SD=225.7) vaping events and completed 98.4% of the smoking context assessment prompts and 97.9% of the vaping context assessment prompts. Tables 1 and 2 characterize the demographics and tobacco use of the sample. Exclusive smokers differed significantly from dual user groups in that dual users were significantly: more likely to be White, report a psychiatric history, live with someone who vapes, report smoking significantly fewer cigarettes per day, report more motivation to quit smoking (albeit only a mean of 3.8 out of a possible 10), and have lower FTCD scores. The most common type of device used was a refillable tank (66.3%), followed by replaceable cartridges (21.3%), and disposables (8.8%%). The most commonly used e-liquid flavors were fruit (45.3%) and menthol (16.4%). The preferred nicotine content in the e-liquid (listed in order of prevalence) was 1-6 mg=27.8%, 7-12 mg=26.4%, 18-24 mg=25.0%, 13-17 mg=13.9%, >24 mg=4.9%, and 0 mg=2.1%. Analyses did not account for product differences among dual users.
Table 1.
Group differences in demographic variables - N (%)
| Total (N= 305) | Smokers (n=143) | Dual Users (n=162) | Inferential Test | df | p-value | ||
|---|---|---|---|---|---|---|---|
| Site | Madison | 136 (44.6) | 56 (39.2) | 80 (49.4) | X2=3.21 | 1 | 0.07 |
| Milwaukee | 169 (55.4) | 87 (60.8) | 82 (50.6) | ||||
| Gender | Women | 152 (49.8) | 72 (50.4) | 80 (49.4) | X2=0.03 | 1 | 0.87 |
| Men | 153 (50.2) | 71 (49.7) | 82 (50.6) | ||||
| Race | White | 201 (69.8) | 78 (57.4) | 123 (80.9) | X2=23.41 | 2 | < 0.001 |
| African American | 69 (24) | 50 (36.8) | 19 (12.5) | ||||
| Multi-racial | 18 (6.3) | 8 (5.9) | 10 (6.6) | ||||
| Hispanic | 14 (4.8) | 5 (3.7) | 9 (5.7) | 1 | 0.73 | ||
| Education | More than high school | 239 (91.6) | 115 (90.6) | 124 (92.5) | X2=0.35 | 2 | 0.84 |
| High school/GED | 20 (7.7) | 11 (8.7) | 9 (6.7) | ||||
| Less than high school | 2 (0.8) | 1 (0.8) | 1 (0.8) | ||||
| Psychiatric history | Any history | 163 (53.4) | 64 (44.8) | 99 (61.1) | X2=8.17 | 1 | 0.004 |
| Depression | 136 (44.6) | 52 (36.4) | 84 (51.9) | X2=7.37 | 1 | 0.006 | |
| Anxiety Disorder | 89 (29.2) | 34 (23.8) | 55 (34.0) | X2=3.80 | 1 | 0.05 | |
| ADD/ADHD | 41 (13.4) | 12 (8.4) | 29 (17.9) | X2=5.90 | 1 | 0.02 | |
| Lives with partner who smokes | 149 (50.3) | 73 (52.9) | 76 (48.1) | X2=0.68 | 1 | 0.68 | |
| Lives with partner who vapes | 59 (20.1) | 9 (6.7) | 50 (31.7) | X2=28.24 | 1 | 28.24 | |
| Age (Mean [SD]) | 39.8 (13.7) | 41.3 (13.6) | 38.4 (13.6) | t=1.84 | 303 | 1.84 | |
Table 2.
Group differences in smoking, vaping, and cigarette dependence (mean [SD])
| Smokers (n=143) | Dual Users (n=162) | t-test | df | p-value | |
|---|---|---|---|---|---|
| Smoking Behavior | |||||
| Years of daily smoking | 24.4 (14.1) | 21.8 (13.6) | 1.62 | 299 | 0.11 |
| Cigarettes/day | 16.0 (11.2) | 12.0 (7.6) | 3.69 | 299 | <0.001 |
| Motivation to quit smoking combustible cigarettes (1-10 scale with 10 = extremely) | 3.3 (1.8) | 3.8 (1.6) | −2.34 | 302 | 0.02 |
| Smoking Dependence | |||||
| Expired CO | 16.7 (9.8) | 15.3 (10.3) | 1.26 | 302 | 0.20 |
| FTCD | 4.8 (2.2) | 4.2 (2.5) | 2.3 | 300 | 0.02 |
| Smoke in first 30 min (N [%]) | 54 (42.5) | 56 (42.4) | X2=0.0002 | 1 | 0.99 |
| WISDM PDM | 4.5 (1.5) | 4.3 (1.5) | 1.29 | 301 | 0.20 |
| WISDM SDM | 4.1 (1.2) | 4 (1.2) | 0.25 | 300 | 0.79 |
| Vaping Behavior | |||||
| Age of first vape | 35.6 (13.8) | ||||
| Years of vaping | 2.8 (2.4) | ||||
| Vape events/day* | 6.1 (7.8) | ||||
| Motivation to quit vaping (1-10 scale with 10 = extremely) | 2.5 (1.8) | ||||
| E-Cigarette Dependence | |||||
| e-FTCD | 2.9 (1.5) | ||||
| Vape in first 30 min (N [%]) | 9 (16.1) | ||||
| % days vape before smoking in the morning | 25.3 (31.8) | ||||
| e-WISDM PDM | 2.9 (1.5) | ||||
| e-WISDM SDM | 3.1 (1.2) | ||||
| e-WISMD Total | 33.1 (13.5) |
FTCD = Fagerstrom Test of Cigarette Dependence [28]; WISDM = Wisconsin Inventory of Smoking Dependence Motives [26]; PDM = Primary Dependence Motives; SDM = Secondary Dependence Motives; e-FTCD = Fagerstrom Test of E-Cigarette Dependence; e-WISDM = Wisconsin Inventory of E-Cigarette Dependence Motives;
Vape events/day based on mean vapes/day from the 2-week EMA sample.
Comparing duration and puffs during smoking and vaping events among dual users and exclusive smokers
Exclusive smokers smoked for approximately 8.4 (SD=8.9) minutes/cigarette and took 12.4 (SD=9.3) puffs/cigarette. Dual users reported smoking for a similar duration as exclusive smokers, approximately 8.2 (SD=9.4) minutes/cigarette and reported similar duration for vape events (i.e., 8.8 (SD=11.3) minutes/vape event). Among dual users, duration of use for either product was not moderated by e-cigarette dependence: the interaction between moderate vs low dependence and product: t-ratio=0.69, df=154, p=0.49; the interaction between high vs low dependence and product: t-ratio=−1.18, df=154, p=0.29. Dual users reported significantly more puffs/cigarette (M=13.1, SD=10.2) than per vape event (M=11.7, SD=11.5; t=3.04, p=0.01). This was not moderated by e-cigarette dependence: the interaction between moderate vs low dependence and product: t-ratio=0.69, df=155, p=0.49; the interaction between high vs low dependence and product: t-ratio=−1.18, df=155, p=0.29.
Association of smoking restrictions on the likelihood of smoking and vaping
On average, exclusive smokers reported that 10.2% of smoking events occurred when smoking was prohibited. Dual users reported that 20% of their smoking events occurred when smoking was not permitted. The presence of a smoking ban also significantly increased the probability that a dual user would vape rather than smoke (OR=2.77, p <0.001). Based on model estimates, a dual user had a 25.1% chance of a vape event when smoking was allowed, versus a 69.5% probability of a vape event when smoking was prohibited (see Figure 1). The impact of a smoking ban on use behavior was moderated by e-cigarette dependence (t-ratio=3.1, p=0.003). Dual users with low or moderate e-cigarette dependence were significantly more likely to vape when smoking was prohibited than when it was permitted (Figure 2). Participants with high e-cigarette dependence showed relatively little difference in the predicted probability of a vape event (vs. a smoking event) based on whether smoking was allowed or prohibited (76.6% vs. 84.6%).
Figure 2. Predicted probability of a vape event when smoking is allowed vs. prohibited among dual users.
Note. * p<.05. Low E-cig Dep = Low E-cigarette Dependence (always smokes first in the morning); Mod E-cig Dep = Moderate E-cigarette Dependence (vapes first on fewer than 50% of mornings); High E-cig Dep = High E-cigarette Dependence (vapes first on half or more mornings)
Associations between social context on use behavior among dual users
Amongst dual users, being around a person vaping was significantly related to choosing to vape rather than smoke (OR=2.3, p<.001). A dual user had a 38.6% predicted probability of a vape event when others were vaping, based on model estimates, compared to a 21.5% predicted probability in the absence of others vaping. A moderation effect showed that the presence of others vaping was more highly related to vaping likelihood amongst participants with low and moderate e-cigarette dependence than amongst those of higher dependence (Figure 3). Participants with low e-cigarette dependence were significantly more likely to vape than smoke in the context of others’ vaping versus when others were not vaping (t-ratio=2.4, p=0.02). There was no difference in moderation of product use by social context when low and moderate e-cigarette dependent participants were compared (t-ratio=0.4, p=0.70). However, high e-cigarette dependent participants were less likely to vape versus smoke in the context of others’ vaping (vs. not vaping) compared to low e-cigarette dependent participants (t-ratio=−2.1, p=0.04).
Figure 3. Predicted probability of a vape event when someone is vaping nearby or not among dual users.
Note. * p<.05. Low E-cig Dep = Low E-cigarette Dependence (always smokes first in the morning); Mod E-cig Dep = Moderate E-cigarette Dependence (vapes first on fewer than 50% of mornings); High E-cig Dep = High E-cigarette Dependence (vapes first on half or more mornings)
Conversely, being around someone who was smoking was inversely associated with the likelihood of vaping vs. smoking (OR=0.7, p=0.02). Due to a failure to converge, this model was run with a random intercept only. Model estimates showed similar probabilities of smoking in dual users (33.2%) and exclusive smokers (35.5%), when in the presence of others smoking. According to model estimates, a dual user had a 20.0% probability of a vape event when around others smoking versus a 26.7% probability without others smoking. This effect was moderated by e-cigarette dependence. Among those with low e-cigarette dependence, exposure to someone smoking significantly reduced the probability of a vape event (predicted probabilities: 5.4% events were vape events when someone was smoking vs. 10.3% were vape events when someone was not smoking; t-ratio = −2.11, p=0.04; see Figure 4). However, among dual users with moderate to high e-cigarette dependence, the probability of vaping was not significantly influenced by exposure to someone smoking (predicted probabilities: in moderate e-cigarette dependence vaping occurred on 14.9% of occasions where smoking occurred vs. in 19.3% of occasions in the absence of smoking; in high e-cigarette dependence vaping occurred in 79.4% of occasions where smoking occurred vs. in 80.3% of occasions in the absence of smoking).
Figure 4. Predicted probability of a vape event when someone is smoking nearby or not among dual users.
Note. * p<.05. Low E-cig Dep = Low E-cigarette Dependence (always smokes first in the morning); Mod E-cig Dep = Moderate E-cigarette Dependence (vapes first on fewer than 50% of mornings); High E-cig Dep = High E-cigarette Dependence (vapes first on half or more mornings)
Associations of internal cues on use behavior among dual users
According to the hierarchical logistic regression analyses, there were no differences in positive affect, negative affect, hunger, or difficulty concentrating between smoking and vaping events among dual users. Because of a failure to converge, these models were run with random intercepts only. When dual users reported a smoking event, they also reported higher cigarette craving than during vape events (OR=0.71, p=.004, see Table 3) and conversely, when they were vaping they reported higher vape craving than during smoking events (OR=2.87, p<.001). A similar pattern was seen for product-specific expectancies, such that product-specific positive expectancies were significantly higher when that product was being used. However, we note these effects did not control for individual differences (through use of random slopes).
Table 3.
Mean and standard deviation of affect, craving, and expectancies during smoking and vaping events among dual users
| Smoking | Vaping | OR | 95% CI | t-ratio | df | p-value | |
|---|---|---|---|---|---|---|---|
| Positive Affect c | 3.1 (1.3) | 3.1 (1.5) | 0.92 | (0.67, 1.25) | −0.55 | 2401 | 0.58 |
| Negative Affect c | 2.6 (1.3) | 2.6 (1.6) | 0.90 | (0.75, 1.09) | −1.09 | 2402 | 0.28 |
| Cigarette Craving | 3.8 (1.3) | 3.4 (2.0) | 0.24 | (0.15, 0.40) | −4.65 | 160 | 0.004 |
| E-Cigarette Craving | 2.6 (1.5) | 3.7 (1.7) | 1.19 | (2.22,4.89) | 5.95 | 160 | <0.001 |
| Positive Cigarette Expectanciesa | 3.9 (1.4) | 3.3 (2.0) | 0.31 | (0.19, 0.53) | −4.41 | 159 | <0.001 |
| Positive E-Cigarette Expectancies b | 2.7 (1.5) | 3.7 (1.7) | 2.30 | (1.60, 3.31) | 4.55 | 159 | <0.001 |
| Hunger c | 2.9 (1.3) | 2.9 (1.6) | 1.07 | (0.82, 1.40) | 0.51 | 2402 | 0.61 |
| Concentration c | 2.2 (1.2) | 2 (1.6) | 0.93 | (0.72, 1.19) | −0.59 | 2383 | 0.55 |
Note.
Mean of 2 Likert items scored from 1=strongly disagree to 7=strongly agree: 1) I would enjoy having a cigarette right now; and 2) Smoking right now would help me feel better.
Mean of 2 Likert items scored from 1=strongly disagree to 7=strongly agree: 1) I would enjoy vaping right now; and 2) Vaping right now would help me feel better.
Models were run with random intercepts only due to failure to converge with random slopes.
All effects remain statistically significant after using the Benjamini-Hochberg approach to control for the false discovery rate.
We examined whether the association of craving and positive expectancies with product use was moderated by e-cigarette dependence. The HLM model predicting likelihood of a vape event based on cigarette craving suggests that while cigarette craving reduces the likelihood that the dual user will vape (OR=0.31, 95% CI: 0.18, 0.55), the more e-cigarette dependent the dual user is, the stronger this effect (OR=.99, 95% CI: .989, .999; see Table 4). Thus, cigarette craving is associated with decreased vaping likelihood especially amongst those high in e-cigarette dependence. These findings were consistent when we used a trichotomized split on the e-cigarette dependence variable (i.e., participants high in e-cigarette dependence had significantly stronger associations between cigarette cravings and use of a combustible cigarettes rather than vaping compared to participants low in e-cigarette dependence). The likelihood of a vape event was: 1) positively related to e-cigarette craving (OR=3.15, 95% CI: 2.11, 4.72) and e-cigarette expectancies (OR=2.20, 95% CI: 1.53, 3.17), but inversely related to positive cigarette expectancies (OR=0.34, 95% CI: 0.19, 0.62) (Table 4). None of these effects was moderated by e-cigarette dependence.
Table 4.
Fixed effects and e-cigarette dependence moderated effects in predicting likelihood that a use event was a vape event
| Main effect |
How main effect varies by first product use ratio |
|||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| β10 | OR | t-ratio | df | p-value | β11 | OR | t-ratio | df | p-value | |
| Cigarette Craving | −1.16 | 0.31 | −4.14 | 159 | <0.001 | −0.0006 | 0.99 | −2.53 | 159 | 0.01 |
| E-cigarette Craving | 1.15 | 3.15 | 5.63 | 159 | <0.001 | −0.0002 | 1.00 | −0.06 | 159 | 0.95 |
| Positive Cigarette Expectancies | −1.08 | 0.34 | −3.34 | 158 | 0.001 | −0.0019 | 1.00 | −0.70 | 158 | 0.49 |
| Positive E-cigarette Expectancies | 0.79 | 2.20 | 4.26 | 158 | <0.001 | −0.0014 | 1.00 | −0.56 | 158 | 0.57 |
Note. Product use ratio = the percent of mornings e-cigarettes are used first.
All effects remain statistically significant after using the Benjamini-Hochberg approach to control for the false discovery rate.
General context based on evening report
We used ANOVA to examine evening reports of daily context by group (exclusive smokers vs. dual users) and e-cigarette dependence (see Table 5). Compared to exclusive smokers, dual users reported higher levels of negative affect, greater e-cigarette craving, more time around vapers and in places where vaping was occurring, more time in situations that made them want to vape, higher levels of stress and greater cigarette and e-cigarette cravings during these stressors. Dual users with high e-cigarette dependence reported greater e-cigarette craving, lower positive smoking expectations, higher positive vaping expectations, a lower percent of the day spent in smoking areas, reduced cigarette craving in situations that made participants want to smoke, higher frequency of situations that made participants want to vape and greater e-cigarette craving in such situations, lower cigarette craving during stressors, and higher e-cigarette craving during stressors compared to dual users with low or moderate e-cigarette dependence.
Table 5.
Evening report EMA smoking restrictions, social contexts, and internal cues by user group and e-cigarette dependence among dual users
| Exclusive Smokers | Dual Users | Low Dep | Mod Dep | High Dep | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Mean | SD | Mean | SD | T-test (df) | p-value† | Mean | SD | Mean | SD | Mean | SD | F (df) | p-value† | |
| Positive Affect | 3.1 | 1.3 | 3.3 | 1.3 | −0.98 (294) | 0.33 | 3.2 | 1.4 | 3.2 | 1.3 | 3.5 | 1.1 | 0.81 (2) | 0.45 |
| Negative Affect | 2.4 | 1.2 | 2.8 | 1.3 | −2.4 (294) | 0.02 | 2.6 | 1.3 | 2.8 | 1.3 | 2.9 | 1.3 | 0.74 (2) | 0.48 |
| Cigarette Craving | 3.6 | 1.3 | 3.6 | 1.3 | 0.01 (294) | 0.996 | 3.6 | 1.3 | 3.8 | 1.3 | 3.1 a | 1.3 | 3.29 (2) | 0.04 †† |
| E-cigarette Craving | 1.1 | 0.5 | 3 | 1.4 | −14.47 (294) | <0.001 | 2.3 | 1.3 | 2.7 | 1 | 4.3 ab | 1.4 | 28.76 (2) | <0.001 |
| Positive Smoking Expectancies | 3.5 | 1.4 | 3.5 | 1.5 | 0 (294) | 0.996 | 3.9 | 1.6 | 3.6 | 1.3 | 2.8 ab | 1.4 | 5.77 (2) | 0.004 |
| Positive Vaping Expectancies | 1.2 | 0.6 | 2.9 | 1.5 | −12.89 (294) | <0.001 | 2.3 | 1.4 | 2.6 | 1.2 | 4.3 ab | 1.4 | 26.62 (2) | <0.001 |
| Percent of Day in Smoking Area | 2.9 | 1 | 2.7 | 1 | 1.5 (294) | 0.13 | 2.9 | 0.9 | 2.9 | 0.9 | 2.2 ab | 1 | 7.14 (2) | 0.001 |
| Percent of Day in Vaping Area | 1.6 | 1.5 | 2.9 | 1 | −9.10 (293) | <0.001 | 2.8 | 1.2 | 3.1 | 1 | 2.8 | 0.9 | 1.32 (2) | 0.27 |
| Percent of Day Around Smokers | 1.7 | 1.2 | 1.5 | 1.1 | 1.16 (294) | 0.25 | 1.8 | 1.2 | 1.6 | 1.1 | 1.1 a | 0.8 | 4.22 (2) | 0.02 |
| Percent of Day Around Vapers | 0.2 | 0.5 | 1 | 1 | −8.42 (294) | <0.001 | 1.2 | 1.1 | 0.8 | 1 | 1.1 | 0.9 | 2.33 (2) | 0.10 |
| Freq. of situations that made you want to smoke (1-7) | 3.4 | 1.4 | 3.7 | 1.4 | −1.75 (294) | 0.08 | 3.8 | 1.6 | 3.9 | 1.3 | 3.2 a | 1.5 | 3.37 (2) | 0.04 †† |
| Strength of cigarette craving in these situations (1-7) | 3.7 | 1.5 | 4 | 1.5 | −1.62 (294) | 0.11 | 4.2 | 1.6 | 4.3 | 1.3 | 3.3 ab | 1.5 | 5.75 (2) | 0.004 |
| Freq. of situations that made you want to vape (1-7) | 1.2 | 0.5 | 3 | 1.4 | −14.72 (293) | <0.001 | 2.5 | 1.4 | 2.9 | 1.2 | 4 ab | 1.4 | 14.49 (2) | <0.001 |
| Strength of e-cigarette craving in these situations (1-7) | 1.1 | 0.4 | 3.1 | 1.5 | −14.65 (294) | <0.001 | 2.5 | 1.4 | 2.9 | 1.4 | 4.3 ab | 1.4 | 17.11 (2) | <0.001 |
| Highest level of stress (1-7) | 3.2 | 1.4 | 3.7 | 1.4 | −3.12 (294) | 0.002 | 3.7 | 1.5 | 3.8 | 1.4 | 3.6 | 1.3 | 0.32 (2) | 0.73 |
| Strength of cigarette craving during the stressor (1-7) | 3.4 | 1.6 | 3.9 | 1.6 | −2.70* (294) | 0.01 | 4.2 | 1.7 | 4.2 | 1.5 | 3.1 ab | 1.5 | 6.12 (2) | 0.003 |
| Strength of e-cigarette craving during the stressor (1-7) | 1.1 | 0.4 | 3 | 1.5 | −14.18 (294) | <0.001 | 2.5 | 1.5 | 2.7 | 1.3 | 4.1 ab | 1.4 | 14.70 (2) | <0.001 |
| Hedonic Tone (mean of pleasure from interpersonal contacts, work, and fun; 1-7) | 4.2 | 1.2 | 4.3 | 1.2 | −0.82 (293) | 0.04 †† | 4.2 | 1.4 | 4.3 | 1.1 | 4.5 | 1.0 | 0.62 (2) | 0.56 |
| Number of alcoholic drinks | 0.8 | 1.4 | 1.1 | 1.8 | −1.19 (287) | 0.02 †† | 0.9 | 2.3 | 1.1 | 1.5 | 1 | 1.6 | 0.20 (2) | 0.82 |
Note.
All effects remain statistically significant after using the Benjamini-Hochberg approach to control for the false discovery rate unless noted.
No longer statistically significant after the Benjamini-Hochberg correction.
Mean for high e-cigarette dependence dual users is statistically significantly different from moderate e-cigarette dependence dual users.
Mean for high e-cigarette dependence dual users is statistically significantly different from low e-cigarette dependence dual users.
Low Dep = low e-cigarette dependence; Mod Dep = moderate e-cigarette dependence; High Dep = high e-cigarette dependence.
DISCUSSION
This research provides new information on the product use patterns of dual users who were in a sustained dual use pattern, reporting no intention to quit smoking or vaping. The results show that dual users used e-cigarettes and cigarettes for about the same amount of time per use event, although they did take 1-2 more puffs from their cigarettes than from their e-cigarettes (albeit individuals may underestimate their puffs on e-cigarettes [37]). The duration of smoking events and puffs/cigarette were similar for people who smoked exclusively and dual users, possibly reflecting the fact that most dual users were smokers first.
The exclusive smoking and dual using groups differed meaningfully in symptomatology, in the contexts in which they lived, and how they reacted to such contexts. The evening report data showed that compared to exclusive smokers, dual users reported spending more time around vapers and in places where vaping was occurring and they reported higher levels of stress and negative affect. In addition, level of e-cigarette dependence was significantly related to a host of factors such a level of exposure to social and other contexts and reactions to such contexts. Dual users were also more likely to smoke in contexts with smoking prohibitions than were those smoking exclusively. The reason(s) for the last finding is unknown but could involve dual users, in general, being less compliant with social proscriptions. While this study does not comprise population-based data, it suggests that dual users and those who smoke differ in multiple ways and these would be difficult to control statistically in observational studies comparing dual users and those smoking exclusively.
There was evidence that e-cigarette dependence moderated the relations between external cues or contexts and vaping likelihood amongst dual users. Dual users of low or moderate e-cigarette dependence were especially likely to concentrate their vaping where smoking was prohibited (e.g., in work settings) vs. allowed. This suggests that established dual users with low e-cigarette dependence are more likely to be using e-cigarettes as a cigarette replacement.
Dual users of low and moderate e-cigarette dependence were also more likely to vape in the presence of others vaping than when no one around them is vaping. This accords with smoking data that show that cigarette dependence is related to level of stimulus control [17, 25, 26]; the greater the level of dependence, the less use behavior is isolated to particular external cues [24]. Shiffman and Bickel have suggested that as dependence increases, the impact of external stimulus control over smoking decreases, possibly because stimulus control shifts from external cues to internal temporally associated cues such as falling levels of drug in the body [16, 17]; also [39, 40]. Smoking appears to be under much greater stimulus control in those who smoke intermittently than in those smoking more heavily [24, 41]. The extent of stimulus control may have implications for treatment (e.g., suggesting the need for pharmacotherapy for those higher in primary dependence and for interventions that focus on external smoking cues for those relatively high in secondary dependence). These results suggest that assessment of e-cigarette dependence should include measures of the stimulus control over product use. For instance, the development of dependence on e-cigarettes may be indexed by the extent to which e-cigarette use becomes less contextually isolated: i.e., occurs across a greater range of contexts.
The relative dependence on e-cigarettes and cigarettes was measured based on the pattern of initial product use in the morning. This is an especially sensitive measure of relative dependence on the two products [38] and of higher primary e-cigarette dependence versus secondary dependence [38]. A progression from secondary to primary dependence on e-cigarettes might predict the likelihood that vaping will replace smoking and that e-cigarette use will be sustained.
E-cigarette dependence also moderated the impact of craving on product use. Dual users who were high in e-cigarette dependence were especially likely to smoke rather than vape when strongly craving a cigarette. This is anomalous in that those with high e-cigarette dependence tended to report lower cigarette craving than those lower in e-cigarette dependence (see Table 5). Moreover, they had lower positive expectancies from smoking. One potential explanation of these findings may be that strong cigarette cravings were relatively rare amongst dual users who were highly dependent upon e-cigarettes making them more salient or motivationally potent. In any event, these results suggest that even though dual users may have had sufficient exposure to e-cigarettes to develop relatively high dependence, the motivation to smoke remains strong. This may be why these dual users did not to quit smoking despite their e-cigarette use; there is sufficient specificity in the motivation to use the two products so as to undercut the fungibility of product use. However, given the unusual nature of these findings they certainly need to be replicated.
There are clear limitations to the conclusions that can be drawn from this research. First, participants in this study used their own e-cigarettes which were all free-base nicotine products (i.e., prior to the introduction of nicotine salt products). We were unable to account for product differences within this sample. It is not clear whether these findings would replicate among current dual users using nicotine salt products [3], which have a pharmacokinetic profile more similar to cigarettes than do free-base e-liquids [42] and perhaps result in greater dependence [43]. Second, we were studying dual users for whom e-cigarette use had not prevented regular, daily smoking who were required to use e-cigarettes once/week use for the last month. Requiring heavier e-cigarette use might have produced greater differential use of cigarettes. Third, these dual users were not currently highly motivated to quit vaping or quit smoking and they were almost all people who smoked prior to dual use, which is not representative of all dual users. Fourth, the EMA data on use behavior (e.g., number of puffs, duration of event) was based on self-report unlike other studies which have measured actual topography [37, 44]. Fifth, there were demographic and dependence differences between those who did and did not provide sufficient data to be included in these analyses. Although, the dual users who dropped out appeared to have used e-cigarettes very little and might not have been representative of dual use. Finally, in several instances, HLM models introducing random slopes failed to achieve convergence. There may be varying explanations for the underlying causes of lack of convergence; our use of random intercept-only models may underestimate standard errors for the effects of predictors in the affected analyses.
In summary, this observational study provides some insight into the vaping and smoking behavior and e-cigarette dependence of people who smoke and vape and are not motivated to quit using either product. These findings suggest that dual users vary markedly in the strength and nature of their e-cigarette dependence and that this appears to have implications for when and where they vape and their motives for doing so. It will be important to take all these factors, including dependence, into account when designing relevant public health interventions.
Supplementary Material
Funding:
Research reported in this publication was supported by the NCI and FDA Center for Tobacco Products (CTP) grant R01CA190025-01. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH or the Food and Drug Administration.
Footnotes
Declaration of Competing Interests: The authors have no conflicts of interest to declare.
REFERENCES
- 1.Centers for Disease Control and Prevention. About one in five U.S. adult cigarette smokers have tried an electronic cigarette2013 29 Mar 2021. Available from: https://www.cdc.gov/media/releases/2013/p0228_electronic_cigarettes.html.
- 2.Meltzer LR, Simmons VN, Pineiro B, Drobes DJ, Quinn GP, Meade CD, et al. Development of a self-help smoking cessation intervention for dual users of tobacco cigarettes and e-cigarettes. Int J Environ Res Public Health. 2021;18(5). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Leventhal AM, Madden DR, Peraza N, Schiff SJ, Lebovitz L, Whitted L, et al. Effect of exposure to e-cigarettes with salt vs free-base nicotine on the appeal and sensory experience of vaping: a randomized clinical trial. JAMA Netw Open. 2021;4(1):e2032757. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Margham J, McAdam K, Forster M, Liu C, Wright C, Mariner D, et al. Chemical composition of aerosol from an e-cigarette: a quantitative comparison with cigarette smoke. Chem Res Toxicol. 2016;29(10):1662–78. [DOI] [PubMed] [Google Scholar]
- 5.Dusautoir R, Zarcone G, Verriele M, Garcon G, Fronval I, Beauval N, et al. Comparison of the chemical composition of aerosols from heated tobacco products, electronic cigarettes and tobacco cigarettes and their toxic impacts on the human bronchial epithelial BEAS-2B cells. J Hazard Mater. 2021;401:123417. [DOI] [PubMed] [Google Scholar]
- 6.Villarroal MA, Cha AE, Vahratian A. Electronic cigarette use among U.S. adults, 20182020 29 Mar 2021. Available from: https://www.cdc.gov/nchs/products/databriefs/db365.htm.
- 7.Dai H, Leventhal AM. Prevalence of e-cigarette use among adults in the United States, 2014-2018. JAMA. 2019;322(18):1824–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.St Helen G, Havel C, Dempsey DA, Jacob P 3rd, Benowitz NL. Nicotine delivery, retention and pharmacokinetics from various electronic cigarettes. Addiction. 2016;111(3):535–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Ward AM, Yaman R, Ebbert JO. Electronic nicotine delivery system design and aerosol toxicants: A systematic review. PLoS One. 2020;15(6):e0234189. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Dawkins L, Corcoran O. Acute electronic cigarette use: nicotine delivery and subjective effects in regular users. Psychopharmacology (Berl). 2014;231(2):401–7. [DOI] [PubMed] [Google Scholar]
- 11.Barrington-Trimis JL, Urman R, Berhane K, Unger JB, Cruz TB, Pentz MA, et al. E-cigarettes and future cgarette use. Pediatrics. 2016;138(1):e20160379. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Vogel EA, Cho J, McConnell RS, Barrington-Trimis JL, Leventhal AM. Prevalence of electronic cigarette dependence among youth and its association with future use. JAMA Netw Open. 2020;3(2):e1921513. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Piper ME, Baker TB, Benowitz NL, Smith SS, Jorenby DE. E-cigarette dependence measures in dual users: reliability and relations with dependence criteria and e-cigarette cessation. Nicotine Tob Res. 2020;22(5):756–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Eisenberg MJ, Hebert-Losier A, Windle SB, Greenspoon T, Brandys T, Fulop T, et al. Effect of e-cigarettes plus counseling vs counseling alone on smoking cessation: a randomized clinical trial. JAMA. 2020;324(18):1844–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Hartmann-Boyce J, McRobbie H, Bullen C, Begh R, Stead LF, Hajek P. Electronic cigarettes for smoking cessation. Cochrane Database Syst Rev. 2016;9:CD010216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Bickel WK, Kelly TH. The relationship of stimulus control to the treatment of substance abuse. Natl Inst Drug Abuse Res Monogr Ser. 841988. p. 122–40. [PubMed] [Google Scholar]
- 17.Shiffman S, Dunbar MS, Ferguson SG. Stimulus control in intermittent and daily smokers. Psychol Addict Behav. 2015;29(4):847–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.U.S. Department of Health and Human Services. The health consequences of smoking - 50 years of progress: A report of the Surgeon General. Atlanta, GA. 2014. [Google Scholar]
- 19.Leventhal AM, Greenberg JB, Trujillo MA, Ameringer KJ, Lisha NE, Pang RD, et al. Positive and negative affect as predictors of urge to smoke: temporal factors and mediational pathways. Psychol Addict Behav. 2013;27(1):262–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Shiffman S, Dunbar M, Kirchner T, Li X, Tindle H, Anderson S, et al. Smoker reactivity to cues: effects on craving and on smoking behavior. J Abnorm Psychol. 2013;122(1):264–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Piper ME, Bolt DM, Kim SY, Japuntich SJ, Smith SS, Niederdeppe J, et al. Refining the tobacco dependence phenotype using the Wisconsin Inventory of Smoking Dependence Motives. J Abnorm Psychol. 2008;117(4):747–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Piasecki TM, Piper ME, Baker TB. Tobacco dependence: insights from investigations of self-reported smoking motives. Curr Dir Psychol Sci. 2010;19(6):395–401. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Baker TB, Piper ME, Schlam TR, Cook JW, Smith SS, Loh WY, et al. Are tobacco dependence and withdrawal related amongst heavy smokers? Relevance to conceptualizations of dependence. J Abnorm Psychol. 2012;121(4):909–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Shiffman S, Dunbar MS, Scholl SM, Tindle HA. Smoking motives of daily and non-daily smokers: a profile analysis. Drug Alcohol Depend. 2012;126(3):362–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Piasecki TM, Piper ME, Baker TB, Hunt-Carter EE. WISDM primary and secondary dependence motives: associations with self-monitored motives for smoking in two college samples. Drug Alcohol Depend. 2011;114(2-3):207–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Piper ME, Piasecki TM, Federman EB, Bolt DM, Smith SS, Fiore MC, et al. A multiple motives approach to tobacco dependence: the Wisconsin Inventory of Smoking Dependence Motives (WISDM-68). J Consult Clin Psychol. 2004;72(2):139–54. [DOI] [PubMed] [Google Scholar]
- 27.Piasecki TM, Piper ME, Baker TB. Refining the tobacco dependence phenotype using the Wisconsin Inventory of Smoking Dependence Motives: II. Evidence from a laboratory self-administration assay. J Abnorm Psychol. 2010;119(3):513–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Fagerstrom K Determinants of tobacco use and renaming the FTND to the Fagerstrom Test for Cigarette Dependence. Nicotine Tob Res. 2012;14(1):75–8. [DOI] [PubMed] [Google Scholar]
- 29.Smith SS, Piper ME, Bolt DM, Fiore MC, Wetter DW, Cinciripini PM, et al. Development of the Brief Wisconsin Inventory of Smoking Dependence Motives. Nicotine Tob Res. 2010;12(5):489–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Foulds J, Veldheer S, Yingst J, Hrabovsky S, Wilson SJ, Nichols TT, et al. Development of a questionnaire for assessing dependence on electronic cigarettes among a large sample of ex-smoking E-cigarette users. Nicotine Tob Res. 2015;17(2):186–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Bolt DM, Piper ME, Theobald WE, Baker TB. Why two smoking cessation agents work better than one: role of craving suppression. J Consult Clin Psychol. 2012;80(1):54–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Piper ME, Cook JW, Schlam TR, Smith SS, Bolt DM, Collins LM, et al. Toward precision smoking cessation treatment II: Proximal effects of smoking cessation intervention components on putative mechanisms of action. Drug Alcohol Depend. 2017;171:50–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Baker TB, Piper ME, Stein JH, Smith SS, Bolt DM, Fraser DL, et al. Effects of nicotine patch vs varenicline vs combination nicotine replacement therapy on smoking cessation at 26 weeks: a randomized clinical trial. JAMA. 2016;315(4):371–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Watson D, Clark LA, Tellegen A. Development and validation of brief measures of positive and negative affect: The PANAS scales. J Pers Soc Psychol. 1988;54(6):1063–70. [DOI] [PubMed] [Google Scholar]
- 35.Smith SS, Piper ME, Bolt DM, Kaye JT, Fiore MC, Baker TB. Revision of the Wisconsin Smoking Withdrawal Scale: Development of brief and long forms. Psychol Assess. 2021;33(3):255–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Benjamini Y, Hochberg Yl. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc, Series B Methodol. 1995;57(1):289–300. [Google Scholar]
- 37.Pearson JL, Elmasry H, Das B, Smiley SL, Rubin LF, DeAtley T, et al. Comparison of ecological momentary assessment versus direct measurement of e-cigarette use with a Bluetooth-enabled e-cigarette: a pilot study. JMIR Res Protoc. 2017;6(5):e84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Piper ME, Baker TB, Mermelstein R, Benowitz N, Jorenby DE. Relations among cigarette dependence, e-cigarette dependence and key dependence criteria amongst dual users of combustible and e-cigarettes Psychol Addict Behav. 2020;Epub ahead of print. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Shiffman S, Waters A, Hickcox M. The Nicotine Dependence Syndrome Scale: a multidimensional measure of nicotine dependence. Nicotine Tob Res. 2004;6(2):327–48. [DOI] [PubMed] [Google Scholar]
- 40.Baker TB, Piper ME, McCarthy DE, Majeskie MR, Fiore MC. Addiction motivation reformulated: an affective processing model of negative reinforcement. Psychol Rev. 2004;111(1):33–51. [DOI] [PubMed] [Google Scholar]
- 41.Shiffman S, Paty J. Smoking patterns and dependence: contrasting chippers and heavy smokers. J Abnorm Psychol. 2006;115(3):509–23. [DOI] [PubMed] [Google Scholar]
- 42.O’Connell G, Pritchard JD, Prue C, Thompson J, Verron T, Graff D, et al. A randomised, open-label, cross-over clinical study to evaluate the pharmacokinetic profiles of cigarettes and e-cigarettes with nicotine salt formulations in US adult smokers. Intern Emerg Med. 2019;14(6):853–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Prochaska JJ, Vogel EA, Benowitz N. Nicotine delivery and cigarette equivalents from vaping a JUULpod. Tob Control. 2021;Online ahead of print. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Dautzenberg B, Bricard D. Real-time characterization of e-cigarettes use; The 1 Million Puffs Study. J Addict Res Ther. 2015;6(2):2–5. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.