Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Sep 14.
Published in final edited form as: Drug Alcohol Depend. 2019 Aug 1;203:99–106. doi: 10.1016/j.drugalcdep.2019.05.020

Effects of non-tobacco flavors and nicotine on e-cigarette product appeal among young adult never, former, and current smokers

Adam M Leventhal a,b,*, Nicholas I Goldenson a, Jessica L Barrington-Trimis a, Raina D Pang a, Matthew G Kirkpatrick a
PMCID: PMC7489757  NIHMSID: NIHMS1621587  PMID: 31434028

Abstract

Background:

E-cigarette regulations targeting products that disproportionately appeal to never-smokers may optimize population health. This laboratory study of young adults tested whether differences in appeal between e-cigarettes with non-tobacco-flavored (vs. tobacco-flavored) and nicotine-containing (vs. nicotine-free) solutions varied by smoking history.

Methods:

Current (N = 53), former (N = 25), and never (N = 22) cigarette smokers who vape (Mean[SD] age = 25.4[4.4] years) administered standardized e-cigarette doses varied by a Flavor (fruit, menthol, tobacco) × Nicotine (nicotine-containing [6 mg/mL], nicotine-free) within-subject double-blind design. Participants rated each dose’s appeal (0–100 scale). Covariate-adjusted interactions tested whether smoking history moderated flavor and nicotine effects.

Results:

Appeal was higher for fruit and menthol than tobacco flavors in each group. The fruit vs. tobacco appeal difference was greater in never smokers (fruit–tobacco estimate = 19.6) than current smokers (estimate = 12.1) but not former smokers (estimate = 12.6). The menthol vs. tobacco difference was greater in never smokers (menthol–tobacco estimate = 17.3) than former (estimate = 6.0) and current (estimate = 7.2) smokers. Appeal was lower for nicotine-containing than nicotine-free solutions in each group; this difference was greater in never smokers (nicotine–nicotine-free estimate = −17.3) than former (estimate = −7.0) and current (estimate = −10.6) smokers. Compared to tobacco flavors, nicotine’s appeal-reducing effects were suppressed by fruit and menthol flavors in never smokers.

Conclusions:

Higher appeal of non-tobacco-flavored (vs. tobacco-flavored) and lower appeal of nicotine-containing (vs. nicotine-free) e-cigarettes may be widespread in young adults but disproportionately amplified in never smokers. Non-tobacco flavors may suppress nicotine’s appeal-lowering qualities in never smokers. The impact of regulating non-tobacco flavors in e-cigarettes may vary by smoking history.

Keywords: e-Cigarettes, Flavors, Tobacco regulatory science, Young adults, Product appeal

1. Introduction

Product standards prohibiting the manufacture, marketing, and sales of electronic cigarettes (e-cigarettes) with specific product characteristics can substantially affect prevalence of vaping in the population and in turn markedly impact public health (Food and Drug Administration, 2018). E-cigarettes currently on the market are available in numerous non-tobacco flavors, such as menthol and a variety of fruit flavors (e.g., candy, fruit), and are sold in various nicotine concentration levels, including nicotine-free solutions (Camenga et al., 2018; Zare et al., 2018). Non-tobacco flavors and nicotine concentration have been targeted in existing or proposed product standards by the U.S. Food and Drug Administration (2018), European Union Tobacco Product Directive (Berteletti et al., 2017; European Commision Public Health, 2018), and other regulatory agencies (Canada House of Commons’ Standing Committee on Health, 2015; Tulsieram et al., 2017).

Differences in user-reported product appeal in response to systematic administration of tobacco products that vary in flavor or nicotine content in the laboratory under double-blind conditions can guide product standards (Carter et al., 2009; Carter and Griffiths, 2009). Such experiments can isolate causal effects of flavorings, nicotine, and other product constituents on appeal while experimentally controlling exogenous influences common to observational research (e.g., marketing, pre-existing beliefs about products, social influences). Evidence from previous laboratory experiments demonstrate that product appeal is higher for e-cigarettes with fruit-flavored or dessert-flavored (Audrain-McGovern et al., 2016; Bono et al., 2018; Goldenson et al., 2016; Kim et al., 2016; DeVito et al., 2019) or menthol-flavored (Krishnan-Sarin et al., 2017; Rosbrook and Green, 2016) solutions than tobacco-flavored or flavorless solutions. Nicotine concentration in e-cigarettes produce counterpoising effects. In addition to increasing neuropharmacological-mediated reinforcement due to nicotine absorption, products with higher nicotine concentration produce aversive acute sensory effects (i.e., perceptions of harshness or bitterness) that may reduce immediate subjective appeal (Etter, 2016; Goldenson et al., 2016; Hajek et al., 2018; Rosbrook and Green, 2016; DeVito et al., 2019). Nicotine concentration may interact with menthol flavorings such that menthol-related cooling sensations suppress nicotine’s unappealing sensory qualities (Krishnan-Sarin et al., 2017; Rosbrook and Green, 2016).

While observational research suggests that use and appeal of non-tobacco flavored e-cigarettes is greater in never smokers than smokers (Berg, 2015; Bonhomme et al., 2016; Farsalinos et al., 2013; Harrell et al., 2016; Patel et al., 2016; Rodu and Plurphanswat, 2018; Russell et al., 2018), prior controlled laboratory product appeal experiments have aggregated results across never, former, and current combustible cigarette smokers, leaving unclear whether product preferences in experimental research vary by smoking history. Translating this literature into actionable evidence to base regulations that benefit the entire population is difficult because smokers and non-smokers garner divergent health effects from exposure to e-cigarette products they find appealing.

Restricting the sale of products that appeal to never smokers could prevent nicotine addiction and reduce exposure to toxins in e-cigarette aerosols in a population who garners no health benefits from vaping (National Academies of Sciences et al., 2018; U.S. Department of Health and Human Services, 2016). Prohibiting the sale of products that appeal to smokers could discourage transition to e-cigarettes and perhaps encourage former smokers who vape to resume smoking. Optimal regulatory restrictions would target particular products with greater appeal in never smokers than former or current smokers. By doing so, regulations could reduce the appeal of e-cigarettes among never smokers without discouraging smokers from transitioning to e-cigarette use. Consequently, it is critical to identify e-cigarette product characteristics that disproportionately appeal to never smokers relative to current or former smokers to inform regulatory policies that optimize health impact to the population as a whole.

Although most middle-aged and older adults who vape are current or former smokers, never-smoking is common among young adult vapers (Centers for Disease Control and Prevention and National Center on Health Statistics, 2016). This laboratory study of young adult e-cigarette users tested whether smoking history moderated the individual and interactive effects of fruit and menthol (vs. tobacco) flavors and nicotine-containing (vs. nicotine-free) e-cigarette solutions on product appeal ratings. Given past observational research results (Berg, 2015; Bonhomme et al., 2016; Farsalinos et al., 2013; Patel et al., 2016), we hypothesized that the relative preference of fruit or menthol (vs. tobacco) flavors would be heightened in never smokers relative to former and current smokers. No a priori hypotheses were made regarding whether nicotine concentration’s main effects or interactive effects with flavor differed by smoking history.

2. Methods

2.1. Participants

Participants (N = 101) in Los Angeles, CA, USA responded to advertisements about a study on reasons for e-cigarette use in 2015. Inclusion criteria were: 1) 18–35 years old; 2) current e-cigarette use > 1 day/week for > 1 month; 3) use nicotine-containing (> 0 mg/ mL) e-cigarettes; and 4) English fluency. Exclusion criteria were: 1) plan to immediately reduce or quit vaping; 2) current smoking cessation medication use; or 3) pregnancy or breastfeeding. One participant who did not complete the entire procedure after enrollment was excluded from analytic sample (N = 100). Participants provided written informed consent. The University of Southern California Institutional Review Board approved the protocol.

2.2. Design

A flavor (fruit, menthol, tobacco) × nicotine concentration (nicotine-containing vs. nicotine-free) within-participant fully crossed factorial design was applied using a double-blind product appeal testing procedure. In previous work, this methodology showed sensitivity to flavor manipulations and successful upholding of study blind (Goldenson et al., 2016). This single-visit procedure involved multiple controlled puffing trials administered via experimenter-provided e-cigarette devices loaded with solutions that systematically varied by flavor and nicotine according to the design. Devices were pre-prepared prior to study visits to keep conditions blind to staff administering the protocol. Nine flavors (5 fruit, 2 menthol, 2 tobacco) each in nicotine and nicotine-free formulations were used (18 total solutions). Each solution was administered at 3.3 and 4.3 V power settings to provide greater generalizability of the findings across devices of varying power available on the market. There were 36 total experimental trials (plus 4 filler trials in a flavorless solution) appearing in a randomly assigned order for each participant. Appeal ratings were collected following each trial.

2.3. Procedure

After eligibility confirmation, participants attended a 4 h laboratory session before which they were told to avoid use of all nicotine/tobacco products for 2 hs prior to arrival. After informed consent and breath carbon monoxide (CO) and salivary cotinine assessment, participants completed the product appeal testing procedure following previous work (Goldenson et al., 2016) in four 9-trial blocks. During each 30-minute inter-block interval, participants completed questionnaires assessing tobacco product use and other characteristics (see below).

At study outset, participants were informed that they would be vaping different e-cigarettes of various flavors, nicotine concentrations, and voltage settings. Participants then viewed an animated video tutorial depicting the controlled puffing parameters. The video cued participants when to inhale and exhale to standardize puffing parameters across trials and was played each trial. A cycle of a 10-s preparation interval, 4 s inhalation interval, 1 s hold interval, and 2 s exhale interval was repeated twice per trial (i.e., 2-puffs per trial). These puff parameters approximate typical e-cigarette use patterns (Helen et al., 2016; Yip and Talbot, 2013) and were used previously (Goldenson et al., 2016). Each trial was separated by a one-minute rest period during which participants were provided with drinking water to reduce cross-trial carryover of olfactory and gustatory sensations. The visible window portion of the atomizer was concealed to prevent unblinding of participants and staff from seeing the color of the e-cigarette solution in each trial. The laboratory was equipped with ventilation to clear aerosol, which reduced staff unblinding due to exposure to the scent of aerosols between trials.

2.4. Materials

Solutions were loaded into Joyetech “Delta 23 Atomizer” tanks with “eVic Supreme” batteries (i.e., tank style device), which allowed voltage to be varied with constant resistance (i.e., 1.5Ω [Ohms]). Solutions (Dekang Biotechnology Co., Ltd.) included those used in our previous study with this paradigm (Goldenson et al., 2016) and were labeled as containing 50/50 propylene glycol/vegetable glycerin (PG/VG) and either 6 mg/mL free-base nicotine or no nicotine. Five fruit-flavored (Blueberry, Strawberry, Peach, Watermelon, Blackberry), 2 menthol-flavored (Triple Menthol, Portal Blend), and 2 tobacco-flavored (Red USA, Desert Ship) solutions were used in addition to the flavorless solution.

2.5. Measures

2.5.1. Appeal

After each 2-puff trial, ratings using visual analogue scales [VAS; 0–100 range] included three appeal questions used previously (Goldenson et al., 2016): 1) “How much did you like it?” (“Not at all” - “Extremely” as anchors); 2) “How much did you dislike it?” (“Not at all” - “Extremely”); and 3) “Would you use it again?” (“Not at all” - “Definitely”). For each trial, composite mean scores of “Liking,” “Disliking” (reverse-scored), and “Willingness to use again” ratings were calculated.

2.5.2. Combustible cigarette smoking history and other participant characteristics

Participants completed a demographic and tobacco product use history questionnaire to characterize the sample and identify covariates (see Table 1). As in prior work (Berg, 2015; Rodu and Plurphanswat, 2018; Russell et al., 2018; Villanti et al., 2017), lifetime and past 30-day smoking questions were used to classify participants who smoked < 100 cigarettes lifetime as never smokers, > 100 cigarettes lifetime without past 30-day smoking as former smokers, and > 100 cigarettes lifetime and past 30-day smoking as current smokers.

Table 1.

Descriptive statistics of participant characteristics, by smoking history.

Stratified by Smoking History
Pooled Sample (N = 100) Never Smokers (N = 22) Former Smokers (N = 25) Current Smokers (N = 53) P-Value for Test of Group Differences
Demographics
Female Gender, N (%) 35 (35.0%) 6 (27.3%) 9 (36.0%) 20 (37.7%) .68
Age, mean (SD), years 25.4 (4.4) 22.7 (4.0)a 25.6 (4.0)b 26.5 (4.4)b .003
Race/Ethnicity, N (%) .14
 Hispanic Ethnicity 22 (22.0%) 6 (27.3%) 6 (24.0%) 10 (18.9%)
 White 29 (29.0%) 6 (27.3%) 11 (44.0%) 12 (22.6%)
 Black 25 (25.0%) 3 (13.6%) 3 (12.0%) 19 (35.8%)
 Asian 15 (15.0%) 6 (27.3%) 3 (12.0%) 6 (11.3%)
 Other 9 (9.0%) 1 (4.5%) 2 (8.0%) 6 (11.3%)
Tobacco Product Use Characteristics
Biomarkers, mean (SD)
 Carbon monoxide, ppm 5.00 (5.51) 3.05 (4.73)a 2.40 (3.04)a 7.04 (5.95)g .001
 Salivary cotinine semi-quantitative levela 2.85 (1.22) 2.14 (0.89)a 3.00 (1.41)g 3.08 (1.15)g .007
Combustible cigarettes
 Age started smoking everyday, mean (SD), years 18.46 (4.11) N/A 18.83 (1.17) 18.42 (4.33) .82
 Current cigarettes/day, mean (SD) 5.89 (5.67) N/A N/A 5.89 (5.67) N/A
 Cigarettes/day when smoking heaviest, mean (SD) 11.60 (7.35) N/A 12.20 (5.40) 11.55 (7.54) .85
 FTCD, mean (SD) 2.45 (1.94) N/A N/A 2.45 (1.94) N/A
 Smoke(d) Mentholated Cigarettes, N (%) 25 (43.1%) N/A N/A 23 (43.4%) N/A
e-Cigarettes
 PSECD, mean (SD) 7.01 (4.51) 4.73 (3.82)a 8.32 (4.89)b 7.34 (4.32)b .02
 Puffs per day, mean (SD) 74.3 (124.3) 49.7 (53.3) 123.0 (205.6) 61.5 (84.8) .07
 Nicotine concentration typically used, mean (SD) 8.77 (13.92) 5.76 (6.23) 6.57 (5.35) 11.1 (18.1) .22
 Duration of e-cigarette use, mean (SD), months 19.8 (14.4) 20.6 (13.3) 21.9 (13.2) 18.4 (15.4) .59
 e-Cigarette device type typically used, N (%) .03
  Cig-a-like 12 (12%) 2 (9.1%)ab 0 (0%)a 10 (18.9%)b
  Tank/pen 30 (30%) 3 (13.6%)a 8 (32.0%)a 19 (35.8%)a
  Advanced personal vaporizer/mod 58 (58%) 17 (77.3%)a 17 (68.0%)ab 24 (45.3%)b
 Preferred e-cigarette flavor, N (%) .58
  Fruit 80 (80.0%) 20 (90.9%) 19 (76.0%) 41 (77.4%)
  Menthol 13 (13.0%) 2 (9.1%) 4 (16.0%) 7 (13.2%)
  Tobacco 7 (7.0%) 0 (0%) 2 (8.0%) 5 (9.4%)
Open in a new tab
a

NicAlert Strip (Range 1–6). P-value for omnibus group differences from χ2 (categorical variable) or ANOVA (continuous variable) test. Groups that do not share superscripts are significantly different without correction for multiple testing (p < .05). FTCD = Fagerström Test for Cigarette Dependence; PSECD = Penn State Electronic Cigarette Dependence Index; ppm = parts per million.

All participants were administered the Penn State Electronic Cigarette Dependence Index (Foulds et al., 2015), a 10-item measure of current e-cigarette use and dependence, that assesses frequency of use, prior quit attempts, difficulty quitting, experience of craving, withdrawal symptoms, and urge use e-cigarettes (range: 0–20). Current smokers completed the Fagerström Test of Cigarette Dependence (FTCD) (Heatherton et al., 1991), a 6-item measure of combustible cigarette dependence severity (range: 0–10). CO was assessed using Smokelyzer breath monitor (Bedfont Scientific Ltd.). Salivary cotinine was measured with NicAlert™ test strips (LiveWellTesting.com, San Diego, CA), which provided a semi-quantitative index (0 = 0–10, 1= 10–30, 2 = 30–100, 3 = 100–200, 4 = 200–500, 5 = 500–1000, 6 = > 1000 ng/mL) (Cooke et al., 2008).

2.5.3. Chemical Assay of e-Cigarette Solutions.

The solutions were analyzed for nicotine, PG, and VG concentrations, and presence of constituents that are known flavorants via gas chromatography with nitrogen-phosphorous detector using previously-published methods (Leigh et al., 2016).

2.6. Data analyses

Following descriptive comparisons of participant characteristics by smoking history groups, the primary analysis used multilevel linear models (MLMs) of trial-level appeal outcome data nested within participants (36 observations per participant). MLMs included the independent categorical variables of flavor (fruit, menthol, and tobacco [reference category]) and nicotine concentration (nicotine and nicotine-free [reference]) fixed effects. Flavor × nicotine concentration interaction effects were also tested to determine whether nicotine effects differed with fruit and menthol flavors compared to tobacco flavor. The MLMs were first tested separately in three subsamples separated into never (n = 22), former (n = 25), and current (n = 53) smokers. Then, to determine whether effects were moderated by smoking history, the subsamples were pooled into a combined dataset (N = 100) to test interactions with the categorical smoking history variable (never [reference], former, current smokers). Flavor × smoking history interactions indicated whether flavor effects differed by smoking history, nicotine × smoking history interactions tested differences in nicotine effects by smoking history, and flavor × nicotine × smoking history interactions indicated whether flavor × nicotine interaction effects varied by smoking history. To address confounds associated with smoking history that could moderate flavor and nicotine effects on appeal, all relevant participant characteristics that significantly differed by smoking history were selected as covariates and adjusted for in the pooled sample moderation by smoking history tests. Four covariate constructs differed by smoking history (i.e., age, e-cigarette device type, e-cigarette dependence, salivary cotinine; see Table 1). Adjustment involved simultaneous inclusion of each covariate construct’s main effect and corresponding interactions with flavor and nicotine (e.g., tests of flavor × smoking history interactions included age, Electronic Cigarette Dependence Index score, cotinine, device type, age × flavor, Electronic Cigarette Dependence Index × flavor, cotinine × flavor, and device type × flavor terms as covariates). The three-way interaction tests included all two-way and three-way interaction term covariates for the 4 covariate constructs (e.g., adjustment for age involved inclusion of age, age × flavor, age × flavor, and age × flavor × nicotine effects).

Primary results are reported as unstandardized effect estimates (differences in outcomes by condition in 0–100 VAS units; Table 2). To correct for multiple testing, the Benjamini-Hochberg method was used to identify two-tailed raw P-values that were significant with a study-wide false discovery rate of 0.05 (see Table 2; Benjamini and Hochberg, 1995). Data were analyzed using IBM SPSS Statistics Version 24 (IBM Corp., Armonk, NY). Additional supplementary analyses are described below.

Table 2.

Effect estimates of fruit, menthol and tobacco-flavored e-cigarettes by smoking

 
  
  
 Comparisons of Effects by Smoking History
Never Smokers (N = 22)
 Former Smokers (N = 25)
 Current Smokers (N = 53)
 Former vs. Never Current vs. Never
M (SE) P M (SE) P M (SE) P P P
Main Effects
Fruit vs. Tobacco 19.6 (2.1) < .001* 12.6 (2.3) < .001* 12.1 (1.5) < .001* .05 .006*
Menthol vs. Tobacco 17.3 (2.5) < .001* 6.0 (2.7) .03* 7.2 (1.8) < .001* < .001* .001*
Nicotine vs. Nicotine-free −17.3 (1.7) < .001* −7.0 (1.8) < .001* −10.6 (1.2) < .001* .002* .01*
Interactions
Fruit vs. Tobacco × Nicotine vs. Nicotine-free 9.0 (4.2) .03* 4.6 (4.5) .31 6.0 (3.0) .05 .78 .62
Menthol vs. Tobacco × Nicotine vs. Nicotine-free 18.0 (5.0) < .001* 4.0 (5.4) .46 13.0 (3.6) < .001* .16  .99
Open in a new tab

Note. Smokers = smoked cigarette in past 30-days and smoked > 100 cigarettes in lifetime and; Former Smokers = Did not smoke in past 30-days but smoked > 100 cigarettes in lifetime; Never Smokers = did not smoke in past 30-days and did not smoke > 100 cigarettes in lifetime. Nicotine = 6 mg/mL; Nicotine-Free = 0 mg/ mL. Appeal = Average of “liking,” “willingness-to-use-again” and “disliking” (reverse-scored) (range 0–100).

*

Significant effect after Benjamini-Hochberg correction to maintain study-wise false discovery rate of 0.05.

3. Results

3.1. Sample characteristics

Descriptive statistics of demographics and tobacco product use characteristics, by smoking history, are reported in Table 1. The pooled sample had a mean age of 25.4 (SD = 4.4) years, was 35% female, was racially/ethnically diverse, and reported, on average, moderate e-cigarette dependence levels on the Electronic Cigarette Dependence Index. Never smokers were younger on average than former and current smokers. E-cigarette dependence was lower in never smokers than former and current smokers. Salivary cotinine levels were higher in current and former (vs. never) smokers. The sample majority (80%) reported preferring fruit-flavored e-cigarette solutions, which did not differ by smoking status. A greater proportion of current smokers used cig-a-like devices than former smokers. Personal vaporizers/mods were used by a greater proportion of never than current smokers (Table 1).

3.2. Chemical assay of e-cigarette solutions

All 9 nicotine-free solutions had nicotine levels below limits of quantitation. The 9 nicotine-containing solutions had a mean nicotine concentration of 6.08 mg/mL (SD = 0.53, range: 5.08–6.72). The mean PG/VG ratio of the 18 solutions was 49.2/50.8 (SD = 4.3/4.3; range: 39.2/60.8 – 56.4/43.6). Numerous flavoring compounds consistent with the labeled flavor were detected in the study e-cigarette solutions detailed in the Online Supplement.

3.3. Effects of flavor and nicotine on product appeal

Table 2 depicts flavor, nicotine, and flavor × nicotine interaction effect estimates in the form of difference scores, stratified by smoking history (middle columns), and interaction tests of moderation by smoking history (right-hand columns).

3.3.1. Flavor effects, by smoking history

Flavor main effects averaged across nicotine conditions showed that appeal ratings were significantly higher for fruit vs. tobacco flavored solutions in each smoking history group (Table 2, Fig. 1). Covariate-adjusted fruit vs. tobacco × smoking history interaction tests indicated that the magnitude of difference in appeal ratings favoring fruit (vs. tobacco) solutions was significantly greater in never smokers (estimate [i.e., fruit – tobacco difference score]: 19.6) compared to current smokers (estimate: 12.1) but was not different from former smokers (estimate: 12.6).

Fig. 1.

Fig. 1.

Open in a new tab

Product appeal of e-cigarettes with fruit, menthol, and tobacco flavored solutions, by smoking history (M ± SE).

*Appeal rating significantly different between respective flavor and tobacco flavor within smoking history group (Pcorr < 0.05).

†Extent of difference in appeal between respective flavor and tobacco flavor significantly differs between respective group and never smokers (Pcorr < 0.05).

Appeal = Average of “liking,” “willingness-to-use-again” and “disliking” (reverse-scored) (range 0–100).

Pcorr = P-value corrected for multiple tests to control study-wise false discovery rate of .05.

Appeal was greater for menthol-flavored than tobacco-flavored solutions in each group. Menthol × smoking history interaction tests showed that the magnitude of preference for menthol vs. tobacco was greater in never smokers (estimate: 17.3) as compared to former smokers (estimate: 6.0) and current smokers (estimate: 7.2).

3.3.2. Nicotine concentration effects, by smoking history

Averaged across flavor conditions, main effect analyses showed that appeal was significantly lower for nicotine-containing (vs. nicotine-free) e-cigarette solutions in each group, and the relative magnitude of this effect was amplified in never smokers (estimate [i.e., nicotine – nicotine free difference score]: −17.0) compared to former (estimate: −7.0) and current (estimate: −10.6) smokers (Fig. 2).

Fig. 2.

Fig. 2.

Open in a new tab

Product appeal of e-cigarettes with nicotine and nicotine-free solutions, by smoking history (M ± SE).

*Appeal rating significantly different from nicotine-free within smoking history group (Pcorr < 0.05).

†Extent of difference in appeal between nicotine and nicotine-free significantly differs between respective group and never smokers (Pcorr < 0.05).

Appeal = Average of “liking,” “willingness-to-use-again” and “disliking” (reverse-scored) (range 0–100).

Pcorr = P-value corrected for multiple tests to control study-wise false discovery rate of .05.

3.3.3. Flavor × Nicotine concentration interaction effects, by smoking history

Among never smokers, the extent to which nicotine-containing solutions produced lower appeal ratings than nicotine-free solutions was suppressed in menthol relative to tobacco flavors (Table 2; Fig. 3). This effect is quantified in the menthol × nicotine estimate (i.e., nicotine – nicotine-free difference for menthol [−8.3] subtracted by nicotine – nicotine-free difference score for tobacco [−26.3]) of 18.0 in never smokers. Evidence of menthol-induced suppression of nicotine’s unappealing effects was also observed in current smokers (menthol × nicotine estimate: 13.0) but not former smokers (estimate: 4.0). Head-to-head comparisons in the magnitude of menthol × nicotine interaction effects did not significantly vary by smoking history group (i.e., non-significant menthol × nicotine × smoking history effects).

Fig. 3.

Fig. 3.

Open in a new tab

Product appeal of e-cigarettes with nicotine and nicotine-free solutions, by flavor and smoking history (M ± SE).

*Extent of difference in appeal between nicotine and nicotine-free significantly differs between respective flavor and tobacco flavor within smoking history group (Pcorr < 0.05).

Appeal = Average of “liking,” “willingness-to-use-again” and “disliking” (reverse-scored) (range 0–100).

Pcorr = P-value corrected for multiple tests to control study-wise false discovery rate of .05.

Fruit vs. tobacco flavors suppressed the extent to which nicotine produced lower appeal than nicotine-free solutions among never smokers (fruit × nicotine estimate: 9.0) but not among former (estimate: 4.6) or current (estimate: 6.0) smokers. The magnitude of fruit × nicotine interaction effects did not significantly vary by smoking history group (i.e., non-significant fruit × nicotine × smoking history effects; Fig. 3).

3.4. Supplementary analysis

Given that participants were exposed to numerous products within a single session, it is possible that participants habituated or became fatigued or (de)sensitized during the procedure, which could alter the flavor and nicotine effects on product appeal ratings. To assess for habituation or fatigue effects, we tested the effects of trial order (i.e., continuous variable, range: 1–36) on appeal as well as the trial order × flavor, order × nicotine, order × nicotine × flavor interactions. None of these effects were significant for each group, providing no evidence that cross-product differences in appeal diminished or increased during the procedure.

4. Discussion

This study provides new experimental evidence indicating that while most young adult e-cigarette users preferred fruit and menthol over tobacco flavors and found nicotine-containing solutions less appealing than nicotine-free solutions, these effects were amplified in never smokers relative to current and former smokers. We also found that menthol and fruit flavors suppressed nicotine’s appeal-reducing qualities among never smokers. In doing so, this study provides some of the most rigorous experimental evidence to date indicating that the appeal of non-tobacco e-cigarette flavors in young adult vapers may be disproportionately augmented in a subpopulation—never smokers—who garner no health benefits and only risks from vaping.

There are several explanations for differences in product appeal by smoking history. Perhaps inter-individual factors that differ between never, former, and current smokers directly influence product preferences and could confound the findings. The likelihood of quitting smoking increases with age (Lee and Kahende, 2007), and a developmentally-normative propensity toward preferring fruit tastes and disliking bitter tastes declines with age (Hoffman et al., 2016). Thus, mere differences in age, rather than smoking history, could drive younger people toward fruit-flavored products and away from bitter-tasting products; nicotine is a bittering agent (Hummel et al., 1992). Never, former, and current smokers may also differ in their nicotine dependence and access to retailers that sell e-cigarette products that sensitize or habituate the user to certain flavors or nicotine concentrations.

To address such alternative explanations, we considered numerous potential confounds as covariates. Given that results were robust to control of the four covariates that differed by group, it is unlikely that these variables or other characteristics reported in Table 1 that did not differ across groups explain differences in flavor and nicotine concentration effects by smoking history. While it remains possible that other unmeasured factors may explain the pattern of findings, such confounds (whether measured or unmeasured) have ambiguous implications for regulatory policies. Even if, for example, age drives preference for non-tobacco flavors regardless of smoking history, because most never smokers who vape are adolescents and young adults (Centers for Disease Control and Prevention and National Center on Health Statistics, 2016), regulations of non-tobacco flavors might still target never smokers more than smokers.

The current findings may also reflect the possibility that smoking combustible cigarettes directly alters preferences for certain e-cigarette flavors and nicotine concentrations. Repeated exposure to the taste of tobacco and the bitter qualities of nicotine from combustible cigarettes could habituate smokers to the aversiveness of these types of orosensory effects and, through associative conditioning with nicotine’s rewarding pharmacological effects, may cause these sensations to become reinforcing in their own right (Pritchard et al., 1996). When similar orosensory perceptions are encountered in tobacco-flavored and nicotine-laden e-cigarette solutions, they may be perceived as more appealing (or less aversive) in current or former smokers than never smokers who may be naive to such orosensory effects. This explanation concords with the current pattern of findings. Group-by-flavor interactions appear to be driven by never smokers’ low appeal of tobacco flavors, especially tobacco-flavored nicotine-containing solutions. Appeal of fruit flavors was similar across groups.

Consistent with previous research in samples with heterogenous smoking histories (Krishnan-Sarin et al., 2017; Rosbrook and Green, 2016; DeVito et al., 2019), this study found that menthol suppressed the tendency of nicotine-containing (vs. nicotine-free) solutions to be perceived as less appealing but only in never and current smokers. Nicotine’s direct sensory-stimulating effects can be aversive because nicotine causes bitter taste perceptions and irritation to the airways. By suppressing these effects, menthol may make the consumption of e-cigarettes with higher nicotine concentrations more palatable. This study also produced novel evidence that, in never smokers only, fruit flavors also suppressed nicotine’s appeal-reducing effects. For never smokers who were particularly sensitive to nicotine’s appeal-attenuating properties in this study, the suppression of nicotine’s aversive sensory effects by non-tobacco flavors could neutralize a barrier that would otherwise prevent never smokers from vaping e-cigarettes in higher nicotine concentrations.

The premise behind product appeal testing is to provide controlled experimental evidence that may be indicative of the likelihood of continued use after initial exposure to a product. While laboratory product appeal testing may not unequivocally translate to use in the natural ecology, previous research using similar paradigms with other addictive consumer products (e.g., alcoholic beverages) have found laboratory-derived measures predict subsequent use persistence and dependence over long-term follow-up (King et al., 2014). Further, the current results concord with population-based observational research showing that ratios of use of non-tobacco flavored vs. tobacco-flavored e-cigarettes are substantially larger in non-smokers than smokers (Bonhomme et al., 2016). Collectively, there appears to be a triangulation of data from complimentary methodologies implicating non-tobacco flavors as a stronger driver of e-cigarette use in never smokers than smokers.

4.1. Limitations

Limitations of this study’s generalizability warrant consideration in interpreting its policy implications. This study included only 100 vapers, all of whom were young adults from one region. Whether this sample is representative of the entire population of young adult vapers or generalizes to other age groups is unknown. Only a specific set of e-cigarette solutions with a single device were tested. Pod-mod e-cigarette products, which have recently become popular (Barrington-Trimis and Leventhal, 2018), operate at lower power settings than the tank-style device used in this study. Pod-mods also utilize solutions with salt-based nicotine formulations, which may produce less aversive sensory effects than the free-base nicotine e-cigarette solutions used here (Barrington-Trimis and Leventhal, 2018). Whether similar results are observed with pod-mods and other e-cigarette products warrants future inquiry.

Conclusions drawn from this study’s methodology do not address certain drivers of product appeal. This method isolates whether constituents in e-cigarette solutions alter the subjective appeal of the immediate user experience. As ratings are provided immediately following product administration, appeal is likely driven by the momentary smells, tastes, airway sensations, and vapor cloud appearances generated while puffing from the product rather than neuropharmacologically-mediated psychoactive drivers of product appeal. Extraneous factors associated with certain products that may also influence product appeal (e.g., labels, advertising, cultural trends) are intentionally muted in this methodology.

Products were grouped by fruit, menthol, and tobacco classes of flavors rather than the specific constituents. An ideal product standard would identify quantitative limits on particular constituents in e-cigarette products. Like previous constituent analyses of e-cigarette solutions (Leigh et al., 2016), dozens of compounds known to produce certain sensory effects from the food science literature were found in this study’s e-cigarette solution. Some constituents may have synergistic effects with one another on taste perceptions, making it difficult to link particular constituents or constituent combinations to specific outcomes. Consequently, classifying flavors into common classes that each have unique defining features, such as what was applied here and previously (Krüsemann et al., 2018; Yingst et al., 2017), makes pragmatic sense.

5. Conclusions

All young adult vapers in this study, regardless of smoking history, found non-tobacco flavors more appealing than tobacco flavored e-cigarettes. Thus, an unfortunate reality is that regulatory policies that reduce the availability of non-tobacco flavors may not benefit the health of every segment of the population, namely smokers who aim to quit smoking with the aid of e-cigarettes and find non-tobacco flavors an appealing alternative to combustible cigarettes (Chen, 2018; Farsalinos et al., 2013; Soule et al., 2019). Yet, findings from the larger literature including those reported here suggest that if tobacco was the primary e-cigarette flavor available, the entire product category would be less appealing (and potentially less addictive) for young never smokers than it would be for smokers interested in transitioning to a lower-harm product.

Supplementary Material

Supplemental Materials

Acknowledgments

Role of funding source

The funding agencies had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Footnotes

Declaration of Competing Interest

Dr. Goldenson left USC on February 10, 2019 and started as an employee of JUUL Labs as of March 4, 2019. He met criteria for authorship prior to leaving USC, and he had no role in revising the paper after leaving USC and joining JUUL Labs. There are no other interests declared by authors.

Appendix A. Supplementary data

Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.drugalcdep.2019.05.020.

References

  1. Audrain-McGovern J, Strasser AA, Wileyto EP, 2016. The impact of flavoring on the rewarding and reinforcing value of e-cigarettes with nicotine among young adult smokers. Drug Alcohol Depend 166, 263–267. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Barrington-Trimis JL, Leventhal AM, 2018. Adolescents’ use of “pod mod” e-cigarettes—urgent concerns. N. Engl. J. Med 379, 1099–1102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Benjamini Y, Hochberg Y, 1995. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. Royal Stat. Soc 57, 289–300. [Google Scholar]
  4. Berg CJ, 2015. Preferred flavors and reasons for e-cigarette use and discontinued use among never, current, and former smokers. Int. J. Public Health 61, 225–236. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Berteletti F, King J, Burch J, Toma Friedlaender A, 2017. Campaign for a revised tobacco products directive in the European Union: lessons learnt. Tob. Control 26, 464–467. [DOI] [PubMed] [Google Scholar]
  6. Bonhomme MG, Holder-Hayes E, Ambrose BK, Tworek C, Feirman SP, King BA, Apelberg BJ, 2016. Flavoured non-cigarette tobacco product use among US adults: 2013–2014. Tob. Control 25, ii4–ii13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bono RS, Barnes AJ, Lester RC, Cobb CO, 2018. Effects of electronic cigarette liquid flavors and modified risk messages on perceptions and subjective effects of e-cigarettes. Health Educ. Behav 46, 197–203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Camenga DR, Morean M, Kong G, Krishnan-Sarin S, Simon P, Bold K, 2018. Appeal and use of customizable e-cigarette product features in adolescents. Tob. Reg. Sci 4, 51–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Canada House of Commons’ Standing Committee on Health, 2015. Vaping: Toward a Regulatory Framework for e-cigarettes. 41st Parliament, Second Session Canada House of Commons, Ottawa, Ontario. [Google Scholar]
  10. Carter LP, Griffiths RR, 2009. Principles of laboratory assessment of drug abuse liability and implications for clinical development. Drug Alcohol Depend 105, S14–S25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Carter LP, Stitzer ML, Henningfield JE, O’Connor RJ, Cummings KM, Hatsukami DK, 2009. Abuse liability assessment of tobacco products including potential reduced exposure products. Cancer Epidemiol. Biomarkers Prev 18, 3241–3262. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Centers for Disease Control and Prevention, National Center on Health Statistics, 2016. QuickStats: cigarette smoking status among current adult e-cigarette users, by age group — national health interview survey, United States, 2015. MMWR 65, 1177. [DOI] [PubMed] [Google Scholar]
  13. Chen JC, 2018. Flavored e-cigarette use and cigarette smoking reduction and cessation— a large national study among young adult smokers. Subst. Use Misuse 52, 2017–2031. [DOI] [PubMed] [Google Scholar]
  14. Cooke F, Bullen C, Whittaker R, McRobbie H, Chen MH, Walker N, 2008. Diagnostic accuracy of NicAlert cotinine test strips in saliva for verifying smoking status. Nicotine Tob. Res 10, 607–612. [DOI] [PubMed] [Google Scholar]
  15. Devito EE, Jensen KP, O’Malley SS, Gueorguieva R, Krishnan-Sarin S, Valentine G, Jatlow PI, Sofuoglu M, 2019. Modulation of “protective” nicotine perception and use profile by flavorants: preliminary findings in e-cigarettes. Nicotine Tob Res Epub ahead of print. [DOI] [PMC free article] [PubMed]
  16. Etter JF, 2016. Throat hit in users of the electronic cigarette: an exploratory study. Psychol. Addict. Behav 30, 93–100. [DOI] [PubMed] [Google Scholar]
  17. European Commision Public Health, 2018. European Tobacco Products Directive 2014/ 14/EU European Commision, Brussels, Belgium. [Google Scholar]
  18. Farsalinos KE, Romagna G, Tsiapras D, Kyrzopoulos S, Spyrou A, Voudris V, 2013. Impact of flavour variability on electronic cigarette use experience: an internet survey. Int. J. Environ. Res. Public Health 10, 7272–7282. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Food and Drug Administration, 2018. Statement From FDA Commissioner Scott Gottlieb, MD, on New Enforcement Actions and a Youth Tobacco Prevention Plan to Stop Youth Use of, and Access to, JUUL and Other e-cigarettes Food and Drug Administration, Silver Spring, MD. [Google Scholar]
  20. Foulds J, Veldheer S, Yingst J, Hrabovsky S, Wilson SJ, Nichols TT, Eissenberg T, 2015. Development of a questionnaire for assessing dependence on electronic cigarettes among a large sample of ex-smoking E-cigarette users. Nicotine Tob. Res 17, 186–192. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Goldenson NI, Kirkpatrick MG, Barrington-Trimis JL, Pang RD, McBeth JF, Pentz MA, Samet JM, Leventhal AM, 2016. Effects of fruit flavorings and nicotine on the appeal and sensory properties of e-cigarettes among young adult vapers: application of a novel methodology. Drug Alcohol Depend 168, 176–180. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hajek P, Przulj D, Phillips-Waller A, Anderson R, McRobbie H, 2018. Initial ratings of different types of e-cigarettes and relationships between product appeal and nicotine delivery. Psychopharmacology 235, 1083–1092. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Harrell M, Weaver S, Loukas A, Creamer M, Marti CN, Jackson CD, Heath JW, Nayak P, Perry CL, Pechacek TF, Eriksen MP, 2016. Flavored e-cigarette use: characterizing youth, young adult, and adult users. Prev. Med. Rep 5, 33–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Heatherton TF, Kozlowski LT, Frecker RC, Fagerström KO, 1991. The Fagerström test for nicotine dependence: a revision of the Fagerstrom Tolerance Questionnaire. Br. J. Addict 86, 1119–1127. [DOI] [PubMed] [Google Scholar]
  25. Helen GS, Ross KC, Dempsey DA, Havel CM, Jacob PIII, Benowitz NL, 2016. Nicotine delivery and vaping behavior during ad libitum e-cigarette access. Tob. Reg. Sci 2, 363–376. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Hoffman AC, Salgado RV, Dresler C, Faller RW, Bartlett C, 2016. Flavour preferences in youth versus adults: a review. Tob. Control 25, ii32–ii39. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Hummel T, Livermore A, Hummel C, Kobal G, 1992. Chemosensory event-related potentials in man: relation to olfactory and painful sensations elicited by nicotine. Electroencephalogr. Clin. Neurophysiol 84, 192–195. [DOI] [PubMed] [Google Scholar]
  28. Kim H, Lim J, Buehler SS, Brinkman MC, Johnson NM, Wilson L, Cross KS, Clark PI, 2016. Role of fruit and other flavours in liking and disliking of electronic cigarettes. Tob. Control 25, ii55–ii61. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. King AC, McNamara PJ, Hasin DS, Cao D, 2014. Alcohol challenge responses predict future alcohol use disorder symptoms: a 6-year prospective study. Biol. Psychiatry 75, 798–806. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Krishnan-Sarin S, Green BG, Kong G, Cavallo DA, Jatlow P, Gueorguiva R, Buta E, O’Malley SS, 2017. Studying the interactive effects of menthol and nicotine among youth: an examination using e-cigarettes. Drug Alcohol Depend 180, 193–199. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Krüsemann EJZ, Boesveldt S, de Graaf K, Talhout R, 2018. An e-liquid flavor wheel: a shared vocabulary based on systematically reviewing e-liquid flavor classifications in literature. Nicotine Tob Res Epub ahead of print. [DOI] [PMC free article] [PubMed]
  32. Lee CW, Kahende J, 2007. Factors associated with successful smoking cessation in the United States, 2000. Am. J. Public Health 97, 1503–1509. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Leigh NJ, Lawton RI, Hershberger PA, Goniewicz ML, 2016. Flavourings significantly affect inhalation toxicity of aerosol generated from electronic nicotine delivery systems (ENDS). Tob. Control 25, ii81–ii87. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. National Academies of Sciences, Engineering, and Medicine, Health and Medicine Division, Board on Population Health and Public Health Practice, Committee on the Review of the Health Effects of Electronic Nicotine Delivery Systems, 2018. Public Health Consequences of E-Cigarettes National Academies Press, Washington, D.C. [Google Scholar]
  35. Patel D, Davis KC, Cox S, Bradfield B, King BA, Shafer P, Caraballo R, Bunnell R, 2016. Reasons for current E-cigarette use among US adults. Prev. Med 93, 14–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Pritchard WS, Robinson JH, Guy TD, Davis RA, Stiles MF, 1996. Assessing the sensory role of nicotine in cigarette smoking. Psychopharmacology 127, 55–62. [DOI] [PubMed] [Google Scholar]
  37. Rodu B, Plurphanswat N, 2018. E-cigarette use among US adults: population Assessment of Tobacco and Health (PATH) study. Nicotine Tob. Res 20, 940–948. [DOI] [PubMed] [Google Scholar]
  38. Rosbrook K, Green BG, 2016. Sensory effects of menthol and nicotine in an e-cigarette. Nicotine Tob. Res 18, 1588–1595. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Russell C, McKeganey N, Dickson T, Nides M, 2018. Changing patterns of first e-cigarette flavor used and current flavors used by 20,836 adult frequent e-cigarette users in the USA. Harm Reduct. J 15, 33. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Soule EK, Plunk AD, Harrell PT, Hayes RB, Edwards KC, 2019. Longitudinal analysis of associations between reasons for electronic cigarette use and change in smoking status among adults in the Population Assessment of Tobacco and Health Study. Nicotine Tob Res Epub ahead of print. [DOI] [PMC free article] [PubMed]
  41. Tulsieram KL, Rinaldi S, Shelley JJ, 2017. Recommendations: will the tobacco and vaping products act go far enough? Can. J. Public Health 108, 328–330. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. U.S. Department of Health and Human Services, 2016. E-cigarette Use among Youth and Young Adults. A Report of the Surgeon General US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health, Atlanta, GA. [Google Scholar]
  43. Villanti AC, Johnson AL, Ambrose BK, Cummings KM, Stanton CA, Rose SW, Feirman SP, Tworek C, Glasser AM, Pearson JL, Cohn AM, Conway KP, Niaura RS, Bansal-Travers M, Hyland A, 2017. Flavored tobacco product use in youth and adults: findings from the first wave of the PATH study (2013–2014). Am. J. Prev. Med 53, 139–151. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Yip H, Talbot P, 2013. Mining data on usage of electronic nicotine delivery systems (ENDS) from YouTube videos. Tob. Control 22, 103–106. [DOI] [PubMed] [Google Scholar]
  45. Yingst JM, Veldheer S, Hammett E, Hrabovksy S, Foulds J, 2017. A method for classifying user-reported electronic cigarette liquid flavors. Nicotine Tob. Res 19, 1381–1385. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Zare S, Nemati M, Zheng Y, 2018. A systematic review of consumer preference for e-cigarette attributes: flavor, nicotine strength, and type. PLoS One 13, e0194145. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Materials
NIHMS1621587-supplement-Supplemental_Materials.doc (2MB, doc)

RESOURCES