Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Feb 18.
Published in final edited form as: Prev Sci. 2025 May 1;26(5):760–772. doi: 10.1007/s11121-025-01811-0

Longitudinal Associations of Impulsivity, Sensation-Seeking, and Peer E-Cigarette Use on the Frequency of E-Cigarette Use Across Young Adulthood

Caroline North 1, Keryn E Pasch 1, Miguel Pinedo 1, Anna V Wilkinson 2, Alexandra Loukas 1
PMCID: PMC12911321  NIHMSID: NIHMS2138767  PMID: 40310498

Abstract

We examined the associations of impulsivity, sensation-seeking, and peer electronic nicotine delivery systems (ENDS) use on longitudinal changes in ENDS use frequency across ages 19–29 years old. Data were drawn from a larger multi-wave study of college students in Texas. Participants were 1227 initially 19–25-year-old young adults who currently used ENDS at least once across six waves (baseline: fall 2015, final wave: spring 2019). At baseline, participants were 21.3 years old on average, 43.6% male, 35.5% non-Hispanic White, 32.5% Hispanic/Latino, 16.2% Asian, 6.7% Black, and 9.1% another racial/ethnic identity. Growth curve modeling with an accelerated longitudinal design was used to test direct and interactive associations of age, impulsivity, sensation-seeking, and peer ENDS use on ENDS use frequency across young adulthood, 19–29 years old. Findings indicated that the trajectory of ENDS use frequency increased with increasing age. Impulsivity, but not sensation-seeking, was associated with an increase in ENDS use frequency across increasing age. Impulsivity and sensation-seeking significantly interacted with peer ENDS use: those high in impulsivity or sensation-seeking used ENDS less frequently as they aged when they had fewer peers who use ENDS, and those high in sensation-seeking used ENDS more frequently when they had more peers who use ENDS. Peers play an important role for young adults with impulsivity and/or sensation-seeking—having few peers who use ENDS was protective of escalations in ENDS use, and having more peers who use ENDS increases the risk for escalations in ENDS use for those high in sensation-seeking only.

Keywords: E-cigarettes, Electronic nicotine delivery systems, Young adults, Personality, Impulsivity, Sensation-seeking, Peers

As of 2021, more than 16% of young adults aged 18–30 years old reported past 30-day electronic nicotine delivery systems (ENDS, also known as e-cigarette) use (Patrick et al., 2022). Use of ENDS in young adulthood (ages 18–25 years old) is concerning because ENDS often contain nicotine (United States Department of Health and Human Services, 2023) and nicotine use before the age of 25 negatively impacts brain development (McGrath-Morrow et al., 2020).

The use of ENDS is also concerning because ENDS often contain toxicants (e.g., formaldehyde and heavy metals) which are associated with an increased risk for respiratory problems, lung damage, cardiovascular disorders, and other neurological manifestations (Esteban-Lopez et al., 2022). Recent evidence indicates that among adolescents and young adults, the frequency of ENDS use escalates/increases as they age (Audrain-McGovern et al., 2021; Doran & Tully, 2018; Loukas et al., 2023). However, few studies examine the longitudinal changes in ENDS use frequency and the factors associated with these changes in young adulthood. Thus, the present study aimed to examine the intra- and inter-personal factors associated with intraindividual changes in ENDS use frequency across young adulthood.

An important intra-personal factor known to be associated with increased risk for substance use in young adulthood is personality (Krueger et al., 2000). Two personality characteristics related to risky behavior (i.e., high-risk personality characteristics) that are associated with young adult tobacco use are impulsivity and sensation-seeking (Quinn & Harden, 2013). Impulsivity is the inclination to engage in hasty, risky, inappropriate acts, and a decreased sensitivity or disregard for the long-term consequences or safety of a behavior (Brikmanis et al., 2017). Sensation-seeking is the seeking of novel, varied, complex, and intense sensations or experiences and the willingness to take risks for such experiences (Zuckerman, 2005). Both impulsivity and sensation-seeking may shape vulnerability to ENDS use.

Limited longitudinal research indicates that impulsivity is associated with changes in ENDS use frequency across young adulthood. One study indicated that higher levels of impulsivity were associated with a greater frequency of ENDS use over a 3-year period among 18- to 25-year-olds (Mittal et al., 2022). Another reported that higher levels of impulsivity were associated with a slower decline in ENDS use frequency across a 2-year period among 18- to 24-year-olds (Doran & Tully, 2018). Research examining the longitudinal associations between sensation-seeking and the trajectory of ENDS use frequency, however, is both limited and mixed. One study found no significant association between sensation-seeking and the trajectory of ENDS use frequency up to 4.5 years later among 18- to 26-year-olds (Loukas et al., 2023), whereas another found that higher levels of sensation-seeking were associated with more frequent ENDS use over a 2-year period among 18- to 24-year-olds (Doran & Tully, 2018). Inconsistent findings of this prior research may be due to differences in measurement of sensation-seeking, time period when ENDS use was assessed, and/or differences in samples.

It is also possible that impulsivity and sensation-seeking are differentially associated with ENDS use frequency. From a theoretical perspective, impulsivity may impact the continued use of ENDS to a greater degree than sensation-seeking. Young adults high in impulsivity have a decreased regard for future consequences (Moeller et al., 2001) and they may find the effects of nicotine as reinforcing their positive affect while decreasing their negative affect (Chase & Hogarth, 2011), both of which may contribute to the continuing and escalating use of ENDS. Comparatively, those high in sensation-seeking may become accustomed to the effects of nicotine, making the physical experiences associated with ENDS use less novel and exciting over time (Perkins et al., 2000). As a result, those high in sensation-seeking may desist their use of ENDS and search for new novel products to try. Given the limited longitudinal studies on the role of personality characteristics in trajectories of ENDS use frequency, research is needed to examine the associations of impulsivity and sensation-seeking on ENDS use frequency in young adulthood. Moreover, research indicates that impulsivity and sensation-seeking are positively correlated with one another (Quinn & Harden, 2013), and thus, research that examines both factors within a single model is needed to determine their unique contributions to the trajectory of ENDS use frequency.

Regardless of the potential differences in how impulsivity and sensation-seeking are associated with ENDS use, not all young adults with high-risk personality characteristics are equally vulnerable to ENDS use. There may be inter-personal, or social contextual factors, that interact with or moderate the associations of impulsivity and sensation-seeking on continued ENDS use, such as peer, or friend, ENDS use. Having a greater number of peers who use ENDS is associated with an increased likelihood of using ENDS up to 1 year later in young adulthood (North et al., 2021; Romm et al., 2022). Peers may provide a unique social context that interacts with impulsivity and sensation-seeking to moderate, or exacerbate, the positive associations between high-risk personality characteristics and the trajectory of ENDS use frequency. Not only are peers and fitting in important for young adults, peers also provide a social context where behaviors are encouraged (Arnett, 2005; Bandura, 1989). For young adults high in impulsivity who are low in selfcontrol, having peers who engage in a particular behavior provides greater opportunities to act upon impulses (Wright et al., 2001). Similarly, young adults high in sensation-seeking may feel more motivated to accede to social pressures and engage in the seemingly novel and stimulating behaviors being modeled by peers (Slater, 2003). Thus, young adults high in impulsivity and/or sensation-seeking may be more likely to escalate ENDS use across time, but primarily when they have a larger number of peers who use ENDS. However, research has yet to determine if peer ENDS use moderates the associations between impulsivity and sensation-seeking on the intraindividual longitudinal changes in ENDS use frequency across young adulthood.

Current Study

The present study examined the associations of impulsivity, sensation-seeking, and peer ENDS use on intraindividual changes/trajectories in ENDS use frequency across ages 19–29 years. The present study also determined if peer ENDS use moderated the direct associations of impulsivity and sensation-seeking on ENDS use frequency across young adulthood. Growth curve models were used to fit the ENDS use frequency trajectory and test study hypotheses, controlling for participant socio-demographic characteristics and current use of other tobacco products, cannabis, and binge drinking. Informed by prior research on escalating ENDS use trajectories among adolescents (Audrain-McGovern et al., 2021) and young adults (Loukas et al., 2023), and on research regarding the roles of impulsivity, sensation-seeking, and peers on ENDS and tobacco use (Doran & Tully, 2018; Mittal et al., 2022; North et al., 2021; Quinn & Harden, 2013; Romm et al., 2022), the following were hypothesized: (1) the trajectory of the frequency of past 30-day ENDS use will increase across increasing age during young adulthood, 19–29 years of age; (2) higher levels of impulsivity and sensation-seeking will be positively associated with ENDS use frequency across increasing age; (3) higher levels of impulsivity and sensation-seeking will be associated with a faster rate of escalation in ENDS use frequency as young adults increase in age; (4) having a greater number of peers who use ENDS will exacerbate the associations observed in Hypothesis 2, such that the associations of impulsivity and sensation-seeking with ENDS use frequency across increasing age in young adulthood will be stronger among those with a greater number of peers who use ENDS compared to those with fewer peers who use ENDS; and (5) there will be two significant three-way interactions: one between peer ENDS use × impulsivity × age and one between peer ENDS use × sensation-seeking × age on ENDS use frequency. The interactions between impulsivity and sensation-seeking with peer ENDS use will be associated with faster rate of escalation in ENDS use frequency across young adulthood. Specifically, the rate of escalation in the trajectory of ENDS use frequency will be faster among young adults who are high in impulsivity or sensation-seeking and have a greater number of peers who use ENDS, compared to those with low high-risk personality characteristics and/or fewer peers who use ENDS.

Method

Participants and Procedures

Participants were a convenience sample of 1227 young adults drawn from the final six full waves of the Marketing and Promotions Across Colleges in Texas project (Project M-PACT). In fall 2014, 5482 18- to 29-year-old college students were recruited via email to participate in an online study. Participants were recruited from 24 colleges and universities across five counties surrounding the four largest metropolitan areas in Texas. A total of 13,714 students were deemed eligible to participate, and ~ 40% (N = 5482) provided informed consent and completed the first web-based survey during fall 2014-early spring 2015. Students were then followed for 4.5 years until spring 2019 across 8 waves: There were five bi-annual surveys from spring 2015 to spring 2018 and two annual surveys in spring 2018 and spring 2019. During fall 2017, an abbreviated contact survey was administered, but this survey did not assess all key study variables; thus, it was not included in the present study. Retention rates from the original sample of 5482 college students across the 8 full study waves ranged from a low of 69% (final wave) to a high of 81% (wave 4). Participants received incentives in the form of a $10 e-gift card for each wave 1 and 2 and a $20 e-gift card for each subsequent wave. All study procedures were Institutional Review Board approved (2013–06–0034) and additional information on Project M-PACT procedures is reported elsewhere (Loukas et al., 2016).

The present study was limited to data from wave 3 (fall, 2015) to wave 8 (spring, 2019) because waves 1 and 2 did not assess past 30-day ENDS use frequency. Thus, six full waves of data were used from wave 3 (fall, 2015) through wave 8 (spring, 2019). Data were further limited to those aged 19–25 years at wave 3 (hereafter referred to as baseline) and those who reported past 30-day ENDS use in one or more of the six waves (n = 1277). The latter requirement was needed to capture intraindividual changes in ENDS use frequency across increasing ages in young adulthood (Nelson, 2021). The final sample included 1277 young adults who were 21.33 years old on average (SD = 1.60) at baseline, 43.6% male, 35.5% identified as non-Hispanic White, 32.5% identified as Hispanic/Latino, 16.2% identified as non-Hispanic Asian, 6.7% identified as non-Hispanic Black, 9.1% identified as another racial/ethnic identity, and 93.8% attended a 4-year university.

Measures

Outcome—ENDS Use Frequency

Frequency of ENDS use, assessed at all six study waves, was queried with: “How many of the past 30 days have you used any ENDS product (i.e., an e-cigarette, vape pen, e-hookah, or mod), even one or two puffs, as intended (i.e., with nicotine cartridges and/or e-liquid/e-juice)?” Starting in spring 2018, the description of ENDS was updated to include new pod vape products on the market (e.g., JUUL). Response options ranged from 0 to 30 days. Responses were log transformed by log (raw value + 1) to account for non-normality of the distribution and log-transformed values ranged from 0 to 3.43. Thus, the outcome is referred to as log ENDS use frequency.

Predictor Variables—High-Risk Personality Characteristics

Impulsivity, examined at all six study waves, was assessed with three items adapted from the Substance Use Risk Profile Scale (Woicik et al., 2009) asking participants to indicate how much they agree/disagree with the following: “I often don’t think things through before I speak,” “I usually act without stopping to think,” and “Generally, I am an impulsive person.” Response options ranged from “strongly disagree” (coded as 1) to “strongly agree” (coded as 5). The three items were averaged to create a total impulsivity score, and this average score was grand mean centered, with higher scores reflecting higher levels of impulsivity. Among the present sample, internal consistency reliability ranged from a low of α = 0.81 (fall, 2016) to a high of α = 0.84 (spring, 2017).

Sensation-seeking, examined at all six study waves, was assessed with four items adapted from the Brief Sensation Seeking Scale-4 (Stephenson et al., 2003). Participants were asked how much they agreed/disagreed with the following: “I would like to explore strange places,” “I like to do frightening things,” “I like new and exciting experiences, even if I have to break the rules,” and “I prefer friends who are exciting and unpredictable.” Response options ranged from “strongly disagree” (coded as 1) to “strongly agree” (coded as 5). The four items were averaged to create a total sensation-seeking score, and this average score was grand mean centered, with higher scores reflecting higher levels of sensation-seeking. Among the present sample, internal consistency reliability ranged from a low of α = 0.82 (fall, 2015) to a high of α = 0.84 (spring, 2018).

Moderating Variable—Perceived Peer ENDS Use

The moderating variable, perceived peer ENDS use (hereafter referred to as peer ENDS use), was assessed with a self-reported measure of number of friends who use ENDS. Peer ENDS use was examined at all six study waves with a question adapted from the Population Assessment of Tobacco and Health (PATH) study (National Institutes of Health, 2015): “How many of your close friends use ENDS products?” Responses ranged from “none” (coded as 0) to “all” (coded as 4). Peer ENDS use was grand mean centered with higher scores reflecting a greater amount of friends who use ENDS.

Covariates

Covariates included time-varying age (centered at 19 years), three time-invariant socio-demographic characteristics (sex, racial/ethnic identity, and college type) assessed at recruitment, and three time-varying substance use covariates (other tobacco use, cannabis use, and binge drinking). Study wave also was included in the model as a time-varying covariate to account for potential period effects. Sex (0 = female; 1 = male) and college type (0 = 2-year; 1 = 4-year) were dichotomus. Racial/ethnic identity was dummy coded (Hispanic/Latino, non-Hispanic Asian, non-Hispanic Black, another racial/ethnic identity) with non-Hispanic white identity as the reference group. Three time-varying substance use covariates were included: (1) current number of other tobacco products used assessed by querying and then summing past 30-day use of four tobacco products (cigarettes, cigars, hookah, smokeless; range 0 = 0 products in past 30, 1 = 1 to 4 products in past 30 days), (2) current cannabis use (0 = did not use cannabis in past 30 days, 1 = used cannabis at least one day in past 30), and (3) current binge drinking (0 = no binge drinking in past 14 days, 1 = at least one episode of binge drinking in past 14 days).

Data Analysis

Data analyses were conducted using R, version 4.2.1 (R Core Team, 2022). Linear mixed effect models were fit using the lme4 package (Bates et al., 2015) and interactions were graphed and probed with the interactions package (Wickman, 2016). Growth curve models, which are analysis of nested/longitudinal data that were implemented in the linear mixed effect model framework, allow the inclusion of fixed and random effects (Leiby, 2012). Growth curve models for an accelerated longitudinal design (Duncan et al., 1996) were conducted to examine intraindividual changes in log-transformed past 30-day ENDS use frequency across young adulthood and assess independent and moderating/interaction associations of impulsivity, sensation-seeking, and peer ENDS use on these changes. The accelerated longitudinal design uses participants’ age, instead of wave, to assess change in ENDS use frequency. Participants contributed up to 3.5 years of data and the age parameter represented the fitted trajectory in ENDS use frequency for the entire age range of the sample (19–29 years old), based on data contributed from participants’ overlapping age ranges. Participants were 19–25 years old at baseline and 22–29 at the final wave. Thus, the full trajectory represents ENDS use frequency from ages 19 to 29 years old, which is the entire age range of the sample. To account for non-independence of observations due to the nested structure of the data, all models included random intercepts for participant and for the college that they attended. Study wave was also included in analytic models to account for potential period effects.

Recommendations from Singer and Willett (2003) were followed during model building. First, an unconditional model was fit that included only time-varying age as an independent predictor of ENDS use frequency. The shape of the unconditional model was established by comparing models where age was treated as linear, quadratic, and logged, using the Bayesian information criterion (BIC). The linear model was selected because it had the lowest BIC (18,770) compared to the quadratic (BIC = 18,826) and logged (BIC = 18,826) models. The linear model exhibited a positive trajectory in ENDS use frequency across increasing age, t (5881) = 7.58, p < 0.001. Therefore, all subsequent analyses were conducted using a linear ENDS use frequency model. Results from the linear model indicated that as young adults got older, they used ENDS more frequently.

After establishing the shape of the model, a series of conditional linear growth curve models assessing intraindividual changes in log past 30-day ENDS use frequency were fit to test study hypotheses. The base model included time-invariant socio-demographic characteristics, time-varying study wave, and time-varying substance use as covariates. Timevarying impulsivity, sensation-seeking, and peer ENDS use were included as primary predictor variables. Following the base model, models were tested that included two-way and three-way interactions.

Interaction effects were examined by adding four two-way interactions sequentially to the base model: (a) age × impulsivity and age × sensation-seeking, to test whether higher levels of impulsivity and/or sensation-seeking were associated with a faster escalation in the rate of change in log past 30-day ENDS use frequency across increasing age (Hypothesis 3) and (b) impulsivity × peer ENDS use and sensation-seeking × peer ENDS use, to test if having a greater number of peers who use ENDS exacerbated the associations between impulsivity and sensation-seeking on ENDS use frequency (Hypothesis 4). In addition, two three-way interactions were also added to the base model: age × impulsivity × peer ENDS use and age × sensation-seeking × peer ENDS use, to determine if the moderating association between peer ENDS use and impulsivity/sensation-seeking was associated with a faster rate of escalation in ENDS use frequency across increasing age in young adulthood (Hypothesis 5). The two three-way interactions models also included the appropriate lower order two-way interactions within the same model (Aiken & West, 1991).

Missing Data

Data from the 1277 participants included in the present study represented a total of 7662 observations across the six study waves. Some participants did not participate in all study waves or answer all questions for primary study variables, resulting in 6526 complete observations (i.e., 14.8% missing). All models were fit using maximum likelihood estimation, allowing for the use of all available data; however, a series of generalized linear mixed models were run to determine if participants with missing data, compared to those with complete data, varied across study covariates. Findings indicated that males were more likely than females to have missing data (OR = 1.21; 95% CI, 1.06, 1.37). Participants identifying as non-Hispanic white were more likely than participants identifying as Hispanic/Latino (OR = 0.72; 95% CI, 0.62, 0.84), Asian (OR = 0.70; 95% CI, 0.57, 0.85), or another racial/ethnic identity (OR = 0.73; 95% CI, 0.57, 0.93) to have missing data. Those who used other tobacco products (i.e., other than ENDS) in the past 30 days (OR = 0.14; 95% CI, 0.11, 0.17) were less likely than their counterparts to have missing data. However, all odds ratios were small in size (Chen et al., 2010), and no significant differences were observed among any other socio-demographic characteristics or other substance use behaviors.

Results

Descriptive statistics were calculated for all time-varying study variables across the six study waves (see Table 1). Past 30-day ENDS use frequency appeared to slightly decrease from fall 2015 (mean = 3.29) to spring 2017 (mean = 2.85) and then began to increase to the highest average frequency during spring 2019 (mean = 5.44). Levels of impulsivity and sensation-seeking both remained relatively stable across the six waves. However, peer ENDS use appeared to slightly decrease from fall 2015 (mean = 2.16) to spring 2017 (mean = 1.95) and then began to increase to the highest average during spring 2019 (mean = 2.35).

Table 1.

Descriptive statistics for all time-varying covariates and primary study variables by wave; fall 2015-spring 2019 (N = 1277)

Fall 2015 Spring 2016 Fall 2016 Spring 2017 Spring 2018 Spring 2019
Age (mean [SD]) 21.33 (1.60) 21.79 (1.61) 22.31 (1.60) 22.79 (1.60) 23.82 (1.61) 24.84 (1.58)
Past 30-day cannabis (%) 38.06 39.93 38.45 39.78 35.87 34.61
Past 14-day binge drinking (%) 42.29 44.71 43.77 43.07 41.50 36.26
Past 30-day # of other tobacco (mean [SD]) 0.76 (0.90) 0.74 (0.92) 0.66 (0.88) 0.67 (0.88) 0.58 (0.83) 0.46 (0.78)
Impulsivity (mean [SD]) 2.50 (0.97) 2.49 (0.98) 2.47 (0.96) 2.47 (0.98) 2.49 (0.94) 2.42 (0.97)
Sensation-seeking (mean [SD]) 3.48 (0.93) 3.47 (0.92) 3.46 (0.91) 3.40 (0.90) 3.46 (0.90) 3.42 (0.92)
Peer ENDS use (mean [SD]) 2.16 (0.90) 2.09 (0.87) 1.99 (0.86) 1.95 (0.82) 2.17 (0.87) 2.35 (0.86)
ENDS use frequency (mean [SD]) 3.29 (7.34) 3.35 (7.67) 2.88 (7.26) 2.85 (7.29) 4.21 (8.72) 5.44 (9.98)
Open in a new tab

Base Model

Table 2 displays the coefficients and standard errors for all variables included in the base model. As noted above in the “Data Analysis” section, results from the unconditional accelerated growth model with only age as a predictor of the trajectory in ENDS use frequency indicated ENDS use frequency increased as young adults got older, γ = 0.06, p < 0.001 (Hypothesis 1). Results from the base model with all primary study variables and covariates indicated that impulsivity and sensation-seeking were not independently associated with ENDS use frequency (Hypothesis 2). However, peer ENDS use, age, wave, sex, racial/ethnic identity, past 30-day use of other tobacco products, and past 14-day binge drinking were significantly associated with ENDS use. Males reported more frequent ENDS use than females, and participants identifying as non-Hispanic white reported more frequent ENDS use compared to participants identifying as Hispanic/Latino, non-Hispanic Black, non-Hispanic Asian, or another racial/ethnic identity. Participants who endorsed using a greater number of other tobacco products in the past 30 days, those who reported binge drinking at least 1 day in the past 14, and those with a greater number of peers who use ENDS also reported more frequent ENDS use than their counterparts.

Table 2.

Longitudinal escalation in ENDS use frequency from ages 19 to 29 years: associations with impulsivity, sensation-seeking, peer ENDS use, socio-demographic characteristics, and other substance use (N = 1277)

Unconditional model

Estimate (SE)		 Base model

Estimate (SE)		 Impulsivity × age model

Estimate (SE)		 Impulsivity × peers
model
Estimate (SE)		 Sensation-seeking ×
peers model
Estimate (SE)
Intercept .46 (.05) *** .62 (.10) *** .62 (.10) *** .62 (.10) *** .61 (.10) ***
Age in years .06 (.01) *** .02 (.01) * .02 (.01) * .02 (.01) * .02 (.01) *
Wave .02 (.01) * .02 (.01) * .02 (.01) * .02 (.01) *
Male .16 (.04) *** .15 (.04) *** .16 (.04) *** .16 (.04) ***
Hispanic/Latino −.25 (.05) *** −.25 (.05) *** −.25 (.05) *** −.25 (.05) ***
Non-Hispanic Black −.20 (.08) * −.20 (.08) * −.21 (.08) * −.20 (.08) *
Non-Hispanic Asian −.27 (.06) *** −.27 (.06) *** −.27 (.06) *** −.27 (.06) ***
Another race/ethnicity −.23 (.07) ** −.23 (.07) *** −.23 (.07) *** −.22 (.07) ***
Four-year college −.11 (.09) −.11 (.09) −.11 (.09) −.11 (.09)
Past 30-day other tobacco .10 (.02) *** .10 (.02) *** .10 (.02) *** .10 (.02) ***
Past 30-day cannabis .02 (.03) .02 (.03) .02 (.03) .02 (.03)
Past 30-day binge drinking .05 (.03) * .05 (.03) * .05 (.03) * .06 (.03) *
Peer ENDS use .34 (.01) *** .34 (.01) *** .34 (.01) *** .34 (.01) ***
Impulsivity −.01 (.02) −.07 (.03) * −.01 (.02) −.01 (.02)
Sensation-seeking −.001(.02) −.001 (.02) −.0003 (.02) .0003 (.02)
Impulsivity × age –- .01 (.01) * –- –-
Impulsivity × peers –- –- .03 (.01) * –-
Sensation-seeking × peers –- –- –- .05 (.01) ***
Open in a new tab

Non-significant interactions are not shown. ***p <.001, **p <.01, *p <.05

Interaction Models

Hypothesis 3 to 5 were examined by including two-way and three-way interactions between age, impulsivity, sensation-seeking, and peer ENDS use into the base model.

High-Risk Personality Characteristics × Age

First to determine if higher levels of impulsivity and sensation-seeking were associated with a faster escalation in ENDS use frequency as young adults aged (Hypothesis 3), two two-way interactions between impulsivity × age and sensation-seeking × age were examined. Only the impulsivity × age interaction was significant (see Table 2). To determine the nature of the impulsivity × age interaction, post-hoc analyses were conducted by calculating the simple slope between age and ENDS use frequency at 1 SD above and below the mean of impulsivity (Aiken & West, 1991). Results of post hoc analyses (see Fig. 1) indicated that the association between age and frequency of ENDS use was significant at higher levels (i.e., 1 SD above the mean) of impulsivity (β = 0.04; 95% CI, [0.01, 0.06]) but not lower levels (i.e., 1 SD below the mean) of impulsivity (β = 0.01; 95% CI, [−0.01, 0.04]). Impulsivity significantly exacerbated the effect of age on ENDS use frequency; young adults high in impulsivity exhibited a faster escalation in ENDS use frequency across increasing age compared to their counterparts low in impulsivity.

Fig. 1.

Fig. 1

Open in a new tab

Impulsivity as a moderator of the association between age and escalation in ENDS use frequency from ages 19 to 29 years old (N = 1277). Note. Shaded areas represent 1 standard error above and below the fitted slopes. Age was centered at 18 years old (now year 0), and thus, the X axis represents the years following age 18

High-Risk Personality × Peers

To determine if having a greater number of peers who use ENDS exacerbated the associations between impulsivity and sensation-seeking on ENDS use frequency (Hypothesis 4), two two-way interactions between impulsivity × peer ENDS use and sensation-seeking × peer ENDS use were examined in separate models. Both interactions were significant (see Table 2) and were probed by calculating the simple slope between impulsivity on ENDS use frequency or sensation-seeking on ENDS use frequency at 1 SD above and below the mean of peer ENDS use (Aiken & West, 1991). Results of post-hoc analyses (see Fig. 2A) indicated that the association between impulsivity and frequency of ENDS use across increasing age in young adulthood was not significant at 1 SD below the mean of peer ENDS use (β = −0.04; 95% CI, [−0.07, 0.0001]) or 1 SD above the mean of peer ENDS use (β = 0.01; 95% CI, [−0.03, 0.05]). These results indicate that the two simple slopes between impulsivity on ENDS use frequency at 1 SD above and 1 SD below the mean of peer ENDS use did not significantly vary. Thus, we examined the simple slopes between on ENDS use frequency at 1.5 SD above and 1.5 SD below the mean of peer ENDS use. Results of post hoc analyses (see Fig. 2B) indicated that the association was significant at 1.5 SD below the mean of peer ENDS use (β = −0.05; 95% CI, [−0.09, −0.001]), but not significant at 1.5 SD above the mean of peer ENDS use (β = 0.02; 95% CI, [−0.02, 0.07]). Peer ENDS use significantly moderated the effect of impulsivity on ENDS use frequency, such that young adults high in impulsivity tended to use ENDS less frequently as they aged when they had very few peers who use ENDS.

Fig. 2.

Fig. 2

Open in a new tab

Peer ENDS use as a moderator of the association between impulsivity and escalations in ENDS frequency from ages 19 to 29 years old (N = 1277). Note. The shaded areas represent 1 standard error above and below the fitted slopes

Figure 3 illustrates the significant sensation-seeking × peer ENDS use interaction. Results from post hoc analyses indicated that the association between sensation-seeking and frequency of ENDS use across increasing age in young adulthood was significant at 1 SD below the mean of peer ENDS (β = −0.04; 95% CI, [−0.08, −0.001]) and significant at 1 SD above the mean of peer ENDS use (β = 0.05; 95% CI, [0.004, 0.09]). Peer ENDS use significantly moderated the association of sensation-seeking with ENDS use frequency, such that young adults low in sensation-seeking tended to use ENDS less frequently as they aged when they had few peers who use ENDS, while young adults high in sensation-seeking tended to use ENDS more frequently as they aged when they had a greater number of peers who use ENDS (see Fig. 3).

Fig. 3.

Fig. 3

Open in a new tab

Peer ENDS use as a moderator of the association between sensation-seeking and escalations in ENDS frequency from ages 19 to 29 years old (N = 1277). Note. The shaded areas represent 1 standard error above and below the fitted slopes

Age × High-Risk Personality × Peers

To determine if the moderating association between peer ENDS use and impulsivity/sensation-seeking was associated with a faster rate of escalation in ENDS use frequency across increasing ages in young adulthood (Hypothesis 5), three-way interactions between age, each high-risk personality characteristic, and peer ENDS use were examined. Neither of the three-way interactions were significant; peer ENDS use did not exacerbate the association between impulsivity and sensation-seeking on the rate of escalation in ENDS use frequency across young adulthood.

Discussion

The present study contributes to the existing body of research by examining the roles of impulsivity, sensation-seeking, and peer ENDS use on intraindividual changes/trajectories in ENDS use frequency across ages 19–29 years. Consistent with prior research on adolescents and young adults (Audrain-McGovern et al., 2021; Loukas et al., 2023), we observed an increasing trajectory in ENDS use frequency across young adulthood. Further, young adults high in impulsivity, but not sensation-seeking, exhibited a faster escalation in ENDS use frequency as they aged as compared to their counterparts lower in impulsivity. The present study extends the existing literature by indicating that young adults high in impulsivity and sensation-seeking used ENDS less frequently as they aged when they had few peers who use ENDS, while those high in sensation-seeking used ENDS more frequently as they aged when they had more peers who use ENDS. Findings point to the need for researchers designing interventions to consider tailoring them to young adults high in impulsivity and those with a greater number of peers who use ENDS, which may help prevent escalations in ENDS use in young adulthood.

In support of Hypothesis 1, young adults used ENDS more frequently as they aged. The present findings are consistent with research among adolescents and young adults (Audrain-McGovern et al., 2021; Loukas et al., 2023), but also inconsistent with at least two studies indicating that ENDS use frequency decreased as young adults aged (Doran & Tully, 2018; Stanton et al., 2020). The inconsistent findings with the two latter studies (Doran & Tully, 2018; Stanton et al., 2020) may be due to a period effect. Specifically, ENDS use frequency was assessed pre-2017 in the two studies showing decreasing ENDS use frequency across increasing age and, thus, did not capture the surge in ENDS use (Dai & Leventhal, 2019; Roberts et al., 2022) that the present study captured and that was hypothesized to have resulted from increases in the average nicotine strength in newer pod-vape ENDS products (Wang et al., 2023). Because newer pod-vapes contain high concentrations of nicotine (Wang et al., 2023), it is possible that young adults are transitioning into heavier ENDS use patterns, possibly due to the development of nicotine dependence (Chesaniuk et al., 2019). However, the present study did not assess nicotine dependence and therefore cannot determine if the increase in ENDS use frequency is associated with, or explained by, nicotine dependence. Nonetheless, given the limited research on young adult ENDS use trajectories, future research should continue to monitor longitudinal changes in young adult ENDS use.

Although impulsivity was not directly associated with a greater frequency of ENDS use across increasing age (Hypothesis 2), young adults high in impulsivity exhibited a faster escalation in their ENDS use frequency as they aged, compared to those low in impulsivity (Hypothesis 3). This finding is inconsistent with one study indicating that impulsivity did not predict a faster escalation in ENDS use frequency (Mittal et al., 2022), and somewhat consistent with a second study indicating that higher levels of impulsivity were associated with slower declines in ENDS use frequency in young adulthood (Doran & Tully, 2018). It is possible that inconsistencies in findings are due to factors such as differences in samples, recruitment contexts, or measurement of impulsivity. Nonetheless, young adults high in impulsivity may use ENDS at a greater frequency as they age because they are particularly susceptible to the rewarding effects of nicotine and therefore are more likely to transition to habitual/compulsive tobacco use (Chase & Hogarth, 2011). As such, young adults who exhibit impulsive traits represent a particularly vulnerable population that warrants the development of ENDS intervention programs specifically targeted to their needs.

Contrary to expectations, sensation-seeking was not directly associated with ENDS use frequency (Hypothesis 2) nor was sensation-seeking associated with a faster escalation in ENDS use frequency across young adulthood (Hypothesis 3). This finding is consistent with two studies (Case et al., 2017; Loukas et al., 2023) and inconsistent with two others (Doran & Tully, 2018; Mitttal et al., 2022). However, it is important to note that the two studies with findings inconsistent with those from this study, sensation-seeking was measured with a single item, which may partially explain the discrepancy in findings. One possible explanation why sensation-seeking was not associated with ENDS use frequency may be due to differences in how impulsivity and sensation-seeking affect substance use behaviors. Research on cigarette smoking indicates that sensation-seeking is associated with initiation of cigarette smoking, while impulsivity is associated with transitions to regular/more frequent cigarette smoking (Balevich et al., 2013). Thus, sensation-seeking may have a greater effect on the initial use of ENDS rather than the continued frequency of use. More research on the role of high-risk personality characteristics and intraindividual changes in ENDS use is warranted to determine if there are differences between impulsivity and sensation-seeking on transitions in ENDS use behaviors.

Findings partially supported Hypothesis 4 indicating that peer ENDS use exacerbated the associations of of impulsivity and sensation-seeking with ENDS use frequency. Young adults high in sensation-seeking used ENDS more frequently as they aged when they had a greater number of peers who use ENDS. This finding is consistent with prior research showing that having a greater number of peers who use substances provides greater opportunities to accede to social pressures among those high in sensation-seeking (Slater, 2003). Findings also extend existing evidence by indicating that peer ENDS use exacerbates the positive associations between high-risk personality characteristics and the frequency of ENDS use. Subsequent research should aim to determine how/why peer ENDS use exacerbates the associations between high-risk personality characteristics and ENDS use. Prior research has hypothesized that peers affect ENDS use through greater opportunities for use (Bandura, 1989), but peers may also influence other factors including attitudes towards ENDS and harm perceptions (Arnett, 2007; Jensen, 2008). Because peer ENDS use is merely one component of social norms, future research should aim to determine if other social normative factors, such as perceived social acceptability of vaping, parent ENDS use, and perceived prevalence of ENDS use (i.e., descriptive norm), play a role in the associations between high-risk personality characteristics and intraindividual changes in ENDS use frequency across young adulthood.

An interesting and unexpected finding was that young adults high in impulsivity and sensation-seeking used ENDS less frequently as they aged when they had few peers who use ENDS. It appears that having more friends who do not vape is protective for young adults with high-risk personality characteristics. Peer use is often conceptualized as a risk factor, but additional research that explores the protective role of low peer use is needed. These findings also raise questions for future lines of research. Prior research on adolescent substance use indicates that those with high-risk personality characteristics are not at an increased/additive risk of substance use when they have many peers who use substances (Dinges & Oetting, 1993; Yanovitzky, 2006). Rather, they are at an increased risk for selecting peers who are similar in personality and exhibit risk-taking behaviors, and it is these peers that directly influence substance use (Dinges & Oetting, 1993; Yanovitzky, 2006). In the context of the present study, peer ENDS use may be the factor that explains why those with high-risk personality characteristics use ENDS more frequently over time. Among the present sample, having a greater number of peers who use ENDS was the strongest predictor of an increased frequency in ENDS use in all analytic models (see Table 2). As such, peer ENDS use may mediate/explain the associations between high-risk personality characteristics and an increased ENDS use frequency, which also may explain the lack of support for Hypothesis 5—the interaction between peer ENDS use and high-risk personality characteristics was not associated with a faster rate of escalation in ENDS use frequency across young adulthood.

Strengths and Limitations

Strengths of the present study include the 3.5-year longitudinal design and the assessment of ENDS use frequency among a large diverse sample of young adults. There are also a few limitations. First, participants were recruited from colleges/universities in Texas and findings may not be generalizable to other young adult populations in the USA. However, college students make up ~ 41% of the young adult population (National Center for Education Statistics, 2021) and thus are representative of a significant portion of young adults. A second limitation is that ENDS use and peeer ENDS use were assessed with online self-report measures that are subject to recall bias. However, prior research supports the validity and reliability of measuring tobacco use behaviors in young adulthood via self-report online surveys (Ramo et al., 2011). Third, although the present study examined longitudinal changes in ENDS use frequency, the ENDS use frequency assessment was based on past 30-day use; therefore, it was not possible to determine the quantity of nicotine consumed when vaping, and if young adults in the sample transitioned to regular ENDS use. Future research, with contemporary data that captures changes in ENDS use post COVID-19 pandemic, should follow young adults as they transition into adult roles to capture possible changes in nicotine consumption and transitions in ENDS use behaviors.

Conclusion

In sum, the present study indicates that as young adults age, they use ENDS more frequently, and higher levels of impulsivity are associated with a faster rate of escalation in ENDS use frequency. Findings also indicate that for young adults high in impulsivity or sensation-seeking, having few peers who use ENDS is associated with a decrease in ENDS use frequency, while for those high in sensation-seeking having a greater number of peers who use ENDS is associated with an increase in ENDS use frequency. These results the first to suggest that having few peers who use ENDS is protective for young adults with high-risk personality characteristics, in that it decreases the likelihood that young adults will escalate in their ENDS use as they age. Future research should aim to examine the roles of impulsivity, sensation-seeking, and peer ENDS use on longitudinal transitions and patterns in ENDS use to determine when in a young adult’s ENDS use trajectory these factors are most influential.

Funding

This work was supported by the National Institutes of Health (1 P50 CA180906 and 1 R01 CA249883-01 A1), from the National Cancer Institute (NCI) and the Food and Drug Administration Center for Tobacco Products (CTP). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health (NIH) or the Food and Drug Administration (FDA). Neither NIH nor FDA had any role in the study design, data collection, analysis, or writing of this paper.

Footnotes

Ethics Approval The questionnaire and methodology for this study was approved by the Institutional Review Board of the University of Texas at Austin (Approval Number: 2013–06-0034). The authors certify that the study was performed in accordance with the ethical standards as laid down in the Declaration of Helsinki and its later amendments.

Informed Consent Informed consent was obtained from all individual participants included in the study.

Conflict of interest The authors declare no competing interests.

Data Availability

Data will be made available upon request.

References

  1. Aiken LS, & West SG (1991). Multiple regression: Testing and interpreting interactions. SAGE. [Google Scholar]
  2. Arnett JJ (2005). The developmental context of substance use in emerging adulthood. Journal of Drug Issues, 35(2), 235–254. 10.1177/002204260503500202 [DOI] [Google Scholar]
  3. Arnett JJ (2007). The myth of peer influence in adolescent smoking initiation. Health Education & Behavior, 34(4), 594–607. 10.1177/1090198105285330 [DOI] [PubMed] [Google Scholar]
  4. Audrain-McGovern J, Rodriguez D, Testa S, Alexander E, & Pianin S (2021). Adolescent E-cigarette onset and escalation: Associations with internalizing and externalizing symptoms. Journal of Adolescent Health, 68(4), 801–807. 10.1016/j.jadohealth.2020.09.033 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Balevich EC, Wein ND, & Flory JD (2013). Cigarette smoking and measures of impulsivity in a college sample. Substance Abuse : Official Publication of the Association for Medical Education and Research in Substance Abuse, 34(3), 256–262. 10.1080/08897077.2012.763082 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bandura A. (1989). Human agency in social cognitive theory. American Psychologist, 10. [DOI] [PubMed] [Google Scholar]
  7. Bates D, Mächler M, Bolker B, & Walker S (2015). Fitting linear mixed-effects models using lme4. Journal of Statistical Software, 67, 1–48. 10.18637/jss.v067.i01 [DOI] [Google Scholar]
  8. Brikmanis K, Petersen A, & Doran N (2017). Do personality traits related to affect regulation predict other tobacco product use among young adult non-daily smokers? Addictive Behaviors, 75, 79–84. 10.1016/j.addbeh.2017.07.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Case KR, Loukas A, Harrell MB, Wilkinson AV, Springer AE, Pérez A, Creamer MR, & Perry CL (2017). The association between sensation seeking and e-cigarette use in Texas young adults: A cross-sectional study. Journal of American College Health, 65(4), 277–285. 10.1080/07448481.2017.1282487 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Chase HW, & Hogarth L (2011). Impulsivity and symptoms of nicotine dependence in a young adult population. Nicotine & Tobacco Research, 13(12), 1321–1325. 10.1093/ntr/ntr114 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Chen H, Cohen P, & Chen S (2010). How big is a big odds ratio? Interpreting the magnitudes of odds ratios in epidemiological studies. Communications in Statistics - Simulation and Computation, 39(4), 860–864. 10.1080/03610911003650383 [DOI] [Google Scholar]
  12. Chesaniuk M, Sokolovsky AW, Ahluwalia JS, Jackson KM, & Mermelstein R (2019). Dependence motives of young adult users of electronic nicotine delivery systems. Addictive Behaviors, 95, 1–5. 10.1016/j.addbeh.2019.02.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Dai H, & Leventhal AM (2019). Prevalence of e-cigarette use among adults in the United States, 2014–2018. JAMA, 322(18), 1824–1827. 10.1001/jama.2019.15331 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Dinges M, & Oetting E (1993). Similarity in drug use patterns between adolescents and their friends. Adolescence, 28(110), 253–253. [PubMed] [Google Scholar]
  15. Doran N, & Tully L (2018). Impulsivity and tobacco product use over time. Addictive Behaviors, 85, 153–157. 10.1016/j.addbeh.2018.06.007 [DOI] [PubMed] [Google Scholar]
  16. Duncan SC, Duncan TE, & Hops H (1996). Analysis of longitudinal data within accelerated longitudinal designs. Psychological Methods, 1(3), 236–248. 10.1037/1082-989X.1.3.236 [DOI] [Google Scholar]
  17. Esteban-Lopez M, Perry MD, Garbinski LD, Manevski M, Andre M, Ceyhan Y, Caobi A, Paul P, Lau LS, Ramelow J, Owens F, Souchak J, Ales E, & El-Hage N (2022). Health effects and known pathology associated with the use of E-cigarettes. Toxicology Reports, 9, 1357–1368. 10.1016/j.toxrep.2022.06.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Jensen RLA, & Gary F (2008). The empirical status of social learning theory of crime and deviance: The past, present, and future. Routledge. [Google Scholar]
  19. Krueger RF, Caspi A, & Moffitt TE (2000). Epidemiological personology: The unifying role of personality in population-based research on problem behaviors. Journal of Personality, 68(6), 967–998. 10.1111/1467-6494.00123 [DOI] [PubMed] [Google Scholar]
  20. Leiby BE (2012). Growth curve mixture models. Shanghai Archives of Psychiatry, 24(6), 355–358. 10.3969/j.issn.1002-0829.2012.06.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Loukas A, Chow S, Pasch KE, Li X, Hinds JT III., Marti CN, Harrell MB, Creamer MR, & Perry CL (2016). College students’ polytobacco use, cigarette cessation, and dependence. American Journal of Health Behavior, 40(4), 514–522. 10.5993/AJHB.40.4.13 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Loukas A, Li X, Wilkinson AV, & Marti CN (2023). Longitudinal examination of ENDS use among young adult college students: Associations with depressive symptoms and sensation seeking. Prevention Science. 10.1007/s11121-023-01572-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. McGrath-Morrow SA, Gorzkowski J, Groner JA, Rule AM, Wilson K, Tanski SE, Collaco JM, & Klein JD (2020). The effects of nicotine on development. Pediatrics, 145(3), e20191346. 10.1542/peds.2019-1346 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Mittal A, Du A, Merz W, Myers MG, Crotty Alexander LE, & Doran N (2022). Impulsivity-related personality traits as predictors of E-cigarette use among young adults over time. Substance Use & Misuse, 57(7), 1007–1013. 10.1080/10826084.2022.2046101 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Moeller FG, Barratt ES, Dougherty DM, Schmitz JM, & Swann AC (2001). Psychiatric aspects of impulsivity. American Journal of Psychiatry, 158(11), 1783–1793. 10.1176/appi.ajp.158.11.1783 [DOI] [PubMed] [Google Scholar]
  26. National Center for Education Statistics. (2021). The NCES Fast Facts Tool provides quick answers to many education questions (National Center for Education Statistics). National Center for Education Statistics. https://nces.ed.gov/fastfacts/display.asp?id=98. Accessed May 2024. [Google Scholar]
  27. National Institutes of Health. (2015). Population assessment of tobacco and health (PATH) study [United States] restricted-use files: Version 37 (Version v37). ICPSR - Interuniversity Consortium for Political and Social Research. 10.3886/ICPSR36231.V37 [DOI] [Google Scholar]
  28. North C, Li X, Grossberg LA, & Loukas A (2021). A one year prospective examination of risk factors for pod-vape use among young adults. Drug and Alcohol Dependence, 229, 109141. 10.1016/j.drugalcdep.2021.109141 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Patrick ME, Schulenberg JE, Miech RA, Johnston LD, O’Malley PM, & Bachman JG (2022). Monitoring the future panel study annual report: National data on substance use among adults ages 19 to 60, 1976–2021. In Institute for Social Research. Institute for Social Research. https://eric.ed.gov/?id=ED623992. Accessed May 2024. [Google Scholar]
  30. Perkins KA, Gerlach D, Broge M, Grobe JE, & Wilson A (2000). Greater sensitivity to subjective effects of nicotine in nonsmokers high in sensation seeking. Experimental and Clinical Psychopharmacology, 8(4), 462–471. 10.1037/1064-1297.8.4.462 [DOI] [PubMed] [Google Scholar]
  31. Quinn PD, & Harden KP (2013). Differential changes in impulsivity and sensation seeking and the escalation of substance use from adolescence to early adulthood. Development and Psychopathology, 25(1), 223–239. 10.1017/S0954579412000284 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. R Core Team. (2022). R: A language and environment for statistical computing [Computer software]. R Foundation for Statistical Computing. https://www.R-project.org/ [Google Scholar]
  33. Ramo DE, Hall SM, & Prochaska JJ (2011). Reliability and validity of self-reported smoking in an anonymous online survey with young adults. Health Psychology: Official Journal of the Division of Health Psychology, American Psychological Association, 30(6), 693–701. 10.1037/a0023443 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Roberts ME, Keller-Hamilton B, Ferketich AK, & Berman ML (2022). Juul and the upsurge of e-cigarette use among college undergraduates. Journal of American College Health, 70(1), 9–12. 10.1080/07448481.2020.1726355 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Romm KF, Wang Y, Ma Y, Wysota CN, Blank MD, Huebner DM, Roche KM, & Berg CJ (2022). The reciprocal relationships of social norms and risk perceptions to cigarette, e-cigarette, and cannabis use: Cross-lagged panel analyses among US young adults in a longitudinal study. Drug and Alcohol Dependence, 238, 109570. 10.1016/j.drugalcdep.2022.109570 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Singer JD, & Willett JB (2003). Applied longitudinal data analysis: Modeling change and event occurrence. Oxford University Press. [Google Scholar]
  37. Slater MD (2003). Sensation-seeking as a moderator of the effects of peer influences, consistency with personal aspirations, and perceived harm on marijuana and cigarette use among younger adolescents. Substance Use & Misuse, 38(7), 865–880. 10.1081/JA-120017614 [DOI] [PubMed] [Google Scholar]
  38. Stanton CA, Sharma E, Edwards KC, Halenar MJ, Taylor KA, Kasza KA, Day H, Anic G, Gardner LD, Hammad HT, Bansal-Travers M, Limpert J, Borek N, Kimmel HL, Compton WM, & Hyland A (2020). Longitudinal transitions of exclusive and polytobacco electronic nicotine delivery systems (ENDS) use among youth, young adults and adults in the USA: Findings from the PATH Study Waves 1–3 (2013–2016). Tobacco Control, 29(Suppl 3), s147–s154. 10.1136/tobaccocontrol-2019-055574 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Stephenson MT, Hoyle RH, Palmgreen P, & Slater MD (2003). Brief measures of sensation seeking for screening and large-scale surveys. Drug and Alcohol Dependence, 72(3), 279–286. 10.1016/j.drugalcdep.2003.08.003 [DOI] [PubMed] [Google Scholar]
  40. United States Department of Health and Human Services. (2023). Know the risks: E-Cigarettes & young people ∣ U.S. Surgeon General’s Report. Know the risks: E-Cigarettes and young people ∣ U.S. Surgeon General’s Report. https://e-cigarettes.surgeongeneral.gov/ [Google Scholar]
  41. Wang X, Ghimire R, Shrestha SS, Borowiecki M, Emery S, & Trivers KF (2023). Trends in nicotine strength in electronic cigarettes sold in the United States by flavor, product type, and manufacturer, 2017–2022. Nicotine & Tobacco Research, 25(7), 1355–1360. 10.1093/ntr/ntad033 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Wickman H. (2016). ggplot2: Elegant graphics for data analysis [Computer software]. Springer International Publishing. [Google Scholar]
  43. Woicik PA, Stewart SH, Pihl RO, & Conrod PJ (2009). The substance use risk profile scale: A scale measuring traits linked to reinforcement-specific substance use profiles. Addictive Behaviors, 34(12), 1042–1055. 10.1016/j.addbeh.2009.07.001 [DOI] [PubMed] [Google Scholar]
  44. Wright BRE, Caspi A, Moffitt TE, & Silva PA (2001). The effects of social ties on crime vary by criminal propensity: A lifecourse model of interdependence*. Criminology, 39(2), 321–348. 10.1111/j.1745-9125.2001.tb00925.x [DOI] [Google Scholar]
  45. Yanovitzky I. (2006). Sensation seeking and alcohol use by college students: Examining multiple pathways of effects. Journal of Health Communication, 11(3), 269–280. 10.1080/10810730600613856 [DOI] [PubMed] [Google Scholar]
  46. Zuckerman M. (2005). The neurobiology of impulsive sensation seeking: Genetics, brain physiology, biochemistry, and neurology. In Stough C (Ed.), Neurobiology of Exceptionality (pp. 31–52). Springer US. 10.1007/0-306-48649-0_2 [DOI] [Google Scholar]

Associated Data

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

Data Availability Statement

Data will be made available upon request.

RESOURCES