Time to first report of signs of nicotine dependence among youth who use e-cigarettes and cigarettes in the United States: A nationally representative cohort study, findings from the Population Assessment of Tobacco and Health, 2013–2019

Abstract

Objective:

To determine the time to first report of signs of nicotine dependence among youth exclusive e-cigarette users and compare this time to that for exclusive cigarette users.

Methods:

Secondary analysis of data (Waves 1–5; 2013–2019) from the Population Assessment of Tobacco and Health was conducted. Youth never tobacco users in the United States who reported exclusive past-30-day (P30D) e-cigarette or cigarette use (n = 2940, N = 5,391,642) in at least one wave were included in the current analysis. Survival analysis was used to estimate the time to the first report of three nicotine dependence indicators (i.e., “use within 30 minutes of waking”; “cravings” and “really needing to use”) following the first report of P30D use. Multivariable Cox proportional hazard models were used to estimate adjusted hazard ratios (aHR).

Results:

There were no significant differences in the time to first report of “use within 30 minutes of waking” (aHR = 1.1, 95% CI = 0.87–1.40) and “cravings” (aHR = 1.09, 95% CI = 0.81–1.47) between exclusive P30D e-cigarette use and exclusive P30D cigarette use. However, compared to exclusive P30D e-cigarette use, the hazard of first reporting “really needing to use” tobacco was 39% (aHR 1.39; 95% CI: 1.05–1.84) times higher for those who reported exclusive P30D cigarette use after controlling for covariates.

Conclusion:

Compared to exclusive P30D cigarette use, no differences in the time to first report of signs of nicotine dependence (“use within 30 minutes” and “cravings”) were observed among exclusive P30D e-cigarette users. Policymakers and regulatory agencies should consider this evidence when assessing the abuse liability of e-cigarette products.

1. Introduction

In 2022, more than 2.5 million (9.4%) middle and high school students reported current (past 30-day) use of e-cigarettes in the United States (US) (Park-Lee, 2022). Nicotine, the primary psychoactive substance in tobacco products, is harmful to the growing adolescent brain (Arrazola et al., 2015; Douglass et al., 2020; Kwan et al., 2020). Therefore, exposure to nicotine during adolescence and young adulthood, irrespective of the product, can increase the risk of nicotine dependence and future use of nicotine products (Vogel et al., 2019; Gaiha and Halpern-Felsher, 2020; Kwan et al., 2020). Factors such as higher nicotine concentration, increased vaping/smoking frequency, device type, and exposure to flavored e-cigarettes or cigarettes increase the risk for nicotine dependence (Azagba et al., 2019; Vogel et al., 2019; Kwan et al., 2020; Peterson et al., 2021; Tackett et al., 2021; Sargent et al., 2022). However, there is limited evidence on the time it takes for youth e-cigarette users to report symptoms of nicotine dependence, and little contemporary evidence of the same for youth cigarette smokers.

Most of our understanding of the time it takes to report symptoms of nicotine dependence among youth is based on studies on cigarette smoking from the late 1990s and early 2000s (DiFranza et al., 2000, 2007; Gervais et al., 2006; Dierker et al., 2008, 2012; Zhan et al., 2012). However, since 2014, the prevalence of e-cigarette use has remained significantly higher among youth than the prevalence of cigarette smoking (Arrazola et al., 2015; Wang et al., 2020). About 1.6% (N = 440,000) of youth reported past 30-day (P30D) cigarette smoking in 2022, compared to 9.4% (N = 2.5 million) of youth who reported P30D e-cigarette use (Park-Lee, 2022). Consequently, research on nicotine dependence must account for changes in the preferred modality of tobacco use among youth.

Previous studies suggest that initiating cigarette smoking at a younger age is associated with an earlier transition to nicotine dependence, daily use, and significantly higher dependence severity than at an older age (DiFranza et al., 2000; Dierker et al., 2012; Huggett et al., 2018; Sharapova et al., 2020). For example, DiFranza et al. (2000) found that among younger adolescent ever-cigarette smokers aged 12–13 years, nearly a quarter reported their first symptoms of nicotine dependence within a month of transitioning to monthly use (i.e., smoking at least once per month) (DiFranza et al., 2000). In contrast, studies that included older adolescents indicated that the first signs of nicotine dependence occur within 1–5 years of cigarette smoking (Dierker et al., 2012; Zhan et al., 2012; Huggett et al., 2018). Among a cohort of teenagers who initiated cigarette smoking between the ages of 13–18, 25% of those who had smoked 100 or more cigarettes in their lifetime experienced at least one symptom of dependence within one year of ever smoking cigarettes (Zhan et al., 2012). Although heavier tobacco use correlates with a higher nicotine intake and increases the risk for dependence, the transition to dependence was much slower in those who started smoking at an older age (DiFranza et al., 2000; Zhan et al., 2012; Vogel et al., 2019; Hammond et al., 2021).

Additionally, nicotine dependence was previously thought to transition through phases, occurring after a regular tobacco use (e.g., daily use) pattern is established and a person has smoked at least 100 cigarettes in their lifetime (Evans et al., 1992; Lamkin and Houston, 1998). However, nicotine dependence can occur before adolescents’ daily or weekly cigarette use is established (DiFranza et al., 2000; Gervais et al., 2006; Dierker et al., 2008). About 62% of adolescent ever cigarette smokers who transitioned to monthly cigarette smoking reported experiencing their first dependence symptoms before daily cigarette use occurred (DiFranza et al., 2000). Similarly, 20% of teenagers who had smoked <100 cigarettes in their lifetime reported dependence symptoms within a year of initiating use (Zhan et al., 2012). Moreover, 50% of adolescents who had only inhaled cigarettes at baseline reported dependence symptoms by the time they smoked seven cigarettes per month (DiFranza et al., 2007). These estimates highlight the complex interplay of biological and behavioral factors that lead to the development of nicotine dependence following exposure to nicotine (DiFranza et al., 2000; Dierker et al., 2012; Huggett et al., 2018; Sharapova et al., 2020).

To our knowledge, no study has examined the time to nicotine dependence among youth e-cigarette users in the US, nor is there a recent study on the time to nicotine dependence in adolescent cigarette smokers. We, therefore, sought to fill this knowledge gap in our study using nationally representative data from Waves 1–5 (2013–2019) of the Population Assessment of Tobacco (PATH) study. The objectives of this study are to 1) determine the time to signs of nicotine dependence among youth with exclusive P30D e-cigarette use 2) compare the time to signs of nicotine dependence for exclusive e-cigarette use to exclusive cigarette use to assess the abuse liability potential of e-cigarette use among youth. Abuse liability is defined as the probability that a product (e.g., tobacco) will be abused or cause dependence (Carter et al., 2009). We hypothesize that the time to the signs of nicotine dependence will be shorter for youth with exclusive cigarette use than for those with exclusive e-cigarette use.

2. Methods

2.1. Data source, study design, and participants

Secondary analysis was conducted using longitudinal data from Waves 1 to 5 (2013–2019) of the Population Assessment of Tobacco and Health (PATH) study. The PATH Study is a nationally representative, cohort study of 45,971 adults (aged 18+ years) and youth (aged 12–17 years) enrolled from the US civilian noninstitutionalized population (CNP) using a four-stage stratified probability sampling design in Wave 1 (Hyland et al., 2017). PATH utilizes population-based interviews regarding the use of tobacco products, risk factors, and health outcomes like nicotine dependence. The sampling frame’s inclusion criteria and details are described elsewhere (Hyland et al., 2017). The Westat Institutional Review Board approved the PATH study design and protocol. This study received IRB approval from the University of Texas Health Science Center at Houston (HSC-SPH-22–1071). The current analysis consisted of 2940 (N = 5,391,642) youth never tobacco users (at Wave 1) who reported exclusive P30D e-cigarette use or cigarette smoking in one or more subsequent waves (Waves 2–5) (Supplemental Fig.). Retention rates for youth never tobacco users (at Wave 1) at follow-up interviews were: 88.2% (Wave 2), 84.4% (Wave 3), 79.3% (Wave 4), and 72.6% (Wave 5).

2.2. Measures

2.2.1. Dependent variable

Dependent variables were the time to the first report of three different validated nicotine dependence indicators following the first report of exclusive P30D use of e-cigarettes or cigarettes. 1.“How soon after you wake up do you want to use[ electronic nicotine products] or [tobacco products]?”[Responses categorized into ≤30 minutes vs. > 30 minutes], 2. “Do you ever have strong cravings to use [electronic nicotine products] or [tobacco products]?”[Yes/No] 3. Have you ever felt like you really needed to use [electronic nicotine products] or [tobacco products]? [Yes/No] (DiFranza et al., 2002; Prokhorov et al., 2017; Chaffee et al., 2022; Strong et al., 2023).

PATH participant follow-up occurred yearly. However, the follow-up times (i.e., the number of weeks since the last interview) varied within and between participants (Supplemental Table 1). In the analysis, the first report of each nicotine dependence indicator at Waves 2–5 was coded 1 (and all other outcomes 0). Missing information for the dependence status was assumed to be non-dependent prior to the first report of dependence or censoring. Thus, if a participant included in the analytic sample at Wave 1 was missing data at subsequent wave(s) (e.g., Wave 2) prior to first reporting P30D use at a later wave (e.g., Wave 3), follow-up began at the wave at which first P30D use was first reported, and they were followed until they first reported dependence or were censored at later waves. For example, if P30D use was first reported at Wave 2 and the first dependence outcome at Wave 3, the time at Wave 2 was coded as 0 and the number of weeks between Waves 2 and 3 was measured as the time to dependence. If no dependence was reported at Wave 3, follow-up continued further from Wave 3 until the first dependence was reported or censoring occurred at either Wave 4 or 5. For those who first reported P30D use in the same wave as their first symptom of nicotine dependence, the time to first report of nicotine dependence was coded as 0 (Pérez et al., 2020). Participants who never reported a symptom of nicotine dependence during all follow-up waves were right censored at Wave 5.

2.2.2. Independent variable

At each wave (Waves 2–5), never users were asked about ever e-cigarette use or cigarette smoking. Participants who reported ever e-cigarette or cigarette use (Waves 2–5) were then asked the question, “In the past 30 days, on how many days did you [use an e-cigarette?] or [smoke cigarettes]?” in the same wave. Use of a product on one or more days was then categorized as exclusive P30D e-cigarette use (e-cigarette use only but not cigarette use) or exclusive P30D cigarette use (cigarette use but not e-cigarette use). P30D use was modeled as a time-varying variable.

2.2.3. Covariates

Covariates included age, race/ethnicity (1 = Non-Hispanic White, 2 = Non-Hispanic Black, 3 = Hispanic, 4 = Other Race), gender (1 = Male, 2 = Female), highest parental education level attained (1 = Less than high school, 2 = High school graduate or equivalent 3 = Some college but no degree or associate degree 4 = Bachelor’s degree 5 = Master’s or higher), P30D other tobacco use (e.g., snus) (1 = Yes, 2 = No), frequency of P30D e-cigarette or cigarette use (1 = 1–19 days vs. 2 = 20–30 days) and PATH survey Wave. P30D other tobacco use and frequency of exclusive P30D e-cigarette or cigarette use were modeled as time-varying covariates and measured at prior waves before each dependence outcome.

2.3. Statistical analysis

Data coding and linkage of youth data to adult data set for persons who turned 18 years or older during follow-up were performed before statistical analysis. Sampling weights and 100 balance repeated replicate weights for each participant’s first wave of youth participation were incorporated in the analyses using the Fays- Balanced Repeated Replicated (Fays-BRR) method with Fay’s adjustment set to 0.3 (United States Department Of Health And Human Services. National Institutes Of Health. National Institute On Drug Abuse, 2016; Pérez et al., 2020).

Weighted frequencies and percentages were reported at baseline. Pearson’s Chi-square (X²) test was used to test for differences in proportions for categorical variables, and the Student’s t-test was used to test for differences in mean age (Table 1). Kaplan-Meier (KM) survival analysis was performed to determine the time to first report of each indicator of nicotine dependence following the first report of P30D use during follow-up (Fig. 1). Consistent with the presentation in prior literature, Fig. 2 shows the time (weeks) at which the cumulative incidence probability of first reporting each indicator of nicotine dependence was 25% (Gervais et al., 2006; Zhan et al., 2012).

Table 1.

Sociodemographic Characteristics for Exclusive Past 30-day users (E-cigarette use or Exclusive Cigarette smoking, n = 2940, N = 5,391,642) at the first report of P30D use, PATH, US (Waves 2–5; 2014–2019).

	Youth past 30-day e-cigarette and cigarette users n = 2940, N = 5,391,642), n (weighted %**)	P-value
Age (years)*	17.63 (0.05)	<0.001
Gender		<0.001
Male	1564 (53.5)
Female	1376 (46.5)
Missing	0
Race/ethnicity		<0.001
Non-Hispanic white	1566 (59.9)
Non-Hispanic black	302 (10.6)
Hispanic	786 (21.0)
Other race***	286 (8.5)
Missing	0
Parent education level		<0.001
Less than high school	536 (16.1)
High school and equivalent	739 (24.1)
Some college but no degree or associate degree	848 (29.2)
Bachelor’s degree	527 (19.9)
Master’s or higher	272 (10.6)
Missing	18

PATH restricted file received disclosure to publish: October 2023, United States Department of Health and Human Services. National Institutes of Health. National Institute on Drug Abuse, and United States Department of Health and Human Services. Food and Drug Administration. Center for Tobacco Products. Population Assessment of Tobacco and Health (PATH) Study [United States] Restricted-Use Files. Inter-university Consortium for Political and Social Research [distributor], 2023–10-04. https://doi.org/10.3886/ICPSR36231.v37

P-values from Chi-square test.

^*Weighted mean and standard errors are shown for age as a continuous variable.

^**Weighted column percentages.

^***Other race includes Asian, other race groups, and multiracial.

Fig. 1.

Cumulative probability for the time to first report of nicotine dependence outcomes by tobacco use behavior (exclusive e-cigarette use and exclusive cigarette smoking) among youth (mean age = 17.6 years) past 30-day users (n = 2940, N = 5,391,642), PATH, US (Waves 2–5; 2014–2019).

Fig. 2.

Number of weeks following P30D use (exclusive e-cigarette or cigarette smoking) at which the cumulative probability of first reporting an indicator of nicotine dependence is 25% among youth (mean age = 17.6 years) exclusive past 30-day users (n = 2940, N = 5,391,642) PATH, US (Waves 2–5; 2014–2019).

Unadjusted and adjusted weighted Cox proportional hazards regression models were used to estimate hazard ratios (HR, aHR) and 95% confidence interval (CI) to evaluate differences in the hazards of first reporting each nicotine dependence indicator (Table 2). The Cox proportional hazard assumption was checked for all covariates using Schoenfeld’s residual test. Statistical significance was tested using a two-sided P-value<0.05 for all analyses. All statistical analyses were done on the ICPSR server with SAS 9.4 (SAS Institute, Cary, North Carolina) and Stata version 17 (Stata Corp., College Station, Texas).

Table 2.

Hazard Ratios (95% CI) for the relationship between past 30-day use (exclusive cigarette smoking vs. exclusive cigarette use) and the time to first report of nicotine dependence (Waves 2–5, 2014–2019) among youth, PATH, US (Waves 2–5; 2014–2019).

	Unadjusted HR (95% CI)	Adjusted HR (95% CI)
Use within 30 minutes of waking Exclusive e-cigarette use	Ref (1.00)	Ref (1.00)
Exclusive cigarette smoking	0.92 (0.75–1.13)	1.10 (0.87–1.40)
Strong cravings Exclusive e-cigarette use	Ref (1.00)	Ref (1.00)
Exclusive cigarette smoking	0.97 (0.74–1.26)	1.09 (0.81–1.47)
Really needing to use Exclusive e-cigarette use	Ref (1.00)	Ref (1.00)
Exclusive cigarette smoking	1.39 (1.07–1.79)	1.39 (1.05–1.84)

PATH restricted file received disclosure to publish: October 2023, United States Department of Health and Human Services. National Institutes of Health. National Institute on Drug Abuse, and United States Department of Health and Human Services. Food and Drug Administration. Center for Tobacco Products. Population Assessment of Tobacco and Health (PATH) Study [United States] Restricted-Use Files. Inter-university Consortium for Political and Social Research [distributor], 2023–10-04. https://doi.org/10.3886/ICPSR36231.v37 Adjusted for: age, gender, race/ethnicity, highest parent grade level, other P30D tobacco use, frequency of P30D use and Wave at participation.

HR- Hazard Ratio.

CI- Confidence Interval.

3. Results

3.1. Sociodemographic factors

Table 1. Among P30D exclusive e-cigarette and cigarette users (n = 2940, N = 5,391,642, mean age = 17.63 years at first P30D use), 53.5% were males, 59.9% were non-Hispanic whites and 29.2% reported some college but no degree or associate degree for highest parental grade level attained.

3.2. Time to first report of indicators of nicotine dependence (Fig. 1)

3.2.1. Use within 30 minutes of waking

There was no difference in the time to first report of use within 30 minutes of waking by P30D use (aHR = 1.10, 95% CI = 0.87–1.40; Table 2). The cumulative incidence of first reporting use within 30 minutes of waking after first P30D use for youth with exclusive cigarette use relative to those with exclusive e-cigarette use were as follows, by year one (11% vs. 8%), by year two (37% vs. 40%), by year three (60% vs. 66%) and by year four (77% vs. 81%). The highest increase in the risk of first reporting use within 30 minutes of waking occurred primarily between 1 and 2 years of first reporting P30D use. The cumulative risk of first reporting use within 30 minutes increased from 11% at 1 year to 37% at 2 years, a 26% difference for youth with exclusive cigarette use and from 8% at 1 year to 40% at 2 years, a 32% difference for youth with exclusive e-cigarette use.

3.2.2. Cravings

There was no significant difference in the time to the first report of cravings by P30D use (aHR = 1.09, 95% CI = 0.81–1.47; Table 2). The cumulative incidence of first reporting cravings after first P30D use for youth with exclusive cigarette use relative to those with exclusive e-cigarette use was as follows, by year one (11% vs. 8.8%), by year two (29% vs. 31%), by year three (57% vs. 60%) and by year four (73% vs. 77%). The highest increase in the risk of first reporting cravings occurred between 2 and 3 years of first reporting P30D use. The cumulative risk of first reporting cravings increased from 29% at 2 years to 57% at 3 years, a 28% difference for youth with exclusive cigarette use and from 31% at 2 years to 60% at 3 years, a 29% difference for youth with exclusive e-cigarette use.

3.2.3. Really needed to use

Compared to exclusive P30D e-cigarette use, the risk of first reporting really needing to use was 39% (aHR 1.39; 95% CI: 1.05–1.84) times higher for those who reported exclusive P30D cigarette use after controlling for covariates (Table 2). The cumulative incidence of first reporting really needing to use after first P30D use for youth with exclusive cigarette use relative to those with exclusive e-cigarette use were as follows, by year one (14.7% vs. 9.9%), by year two (37% vs. 28%), by year three (62% vs. 54%) and by year four (85% vs. 71%). The highest increase in the risk of first reporting really needing to use occurred between 2 and 3 years of first reporting P30D use. The cumulative risk of first reporting really needing to use increased from 37% at 2 years to 62% at 3 years, a 25% difference for youth with exclusive cigarette use and from 28% at 2 years to 54% at 3 years, a 26% difference for youth with exclusive e-cigarette use.

3.3. Time to 25% cumulative probability (Fig. 2)

The 25% cumulative probability of first reporting use within 30 minutes of waking (i.e.,100 weeks vs. 99 weeks) and cravings (i.e., 101 weeks for both groups) were similar for youth with exclusive cigarette use and exclusive e-cigarette use. However, the time to the first report of really needing to use occurred earlier at 96 weeks for 25% of youth with exclusive cigarette use relative to 103 weeks for those with exclusive e-cigarette use.

4. Discussion

The current study estimated the time to the first report of three validated nicotine dependence indicators (i.e., use within 30 minutes of waking, really needing to use, and cravings) (DiFranza et al., 2002; Prokhorov et al., 2017; Chaffee et al., 2022; Strong et al., 2023) following the first report of P30D tobacco use (i.e., exclusive e-cigarette use or exclusive cigarette smoking) among a nationally representative cohort of youth never tobacco users at baseline (Wave 1:2013) who were followed from Wave 2 (2014) to Wave 5 (2019).

4.1. Time to dependence (Studies on cigarette smoking)

Symptoms of nicotine dependence were first reported by 25% of exclusive P30D cigarette smokers within 2 years after first reporting P30D use. Similarly, in the Social and Emotional Contexts of Adolescent Smoking Patterns Study (SECASPS), dependence symptoms (e.g., strong cravings) emerged within a year or two for 25% of 9th and 10th graders followed from 2008 to 2012 after initiating cigarette smoking (Zhan et al., 2012). However, among a younger cohort of 7th graders followed from 1999 to 2004, 25% reported dependence symptoms within months (<1 year) after their first puff of cigarettes (Gervais et al., 2006). Our study and prior studies (Gervais et al., 2006; Zhan et al., 2012) differ in the indicators of nicotine dependence, reference point (first puff vs. inhalation vs. first report of P30D) for follow-up, follow-up time (our study did not have data at 1-month follow-up), and age. Notably, exclusive P30D cigarette smokers in our study were, on average older than the 7th-grade cohort (12.7 years) but similar in age to the SECASPS cohort (15.5 years) in the studies discussed (Gervais et al., 2006; Zhan et al., 2012). It is known that nicotine’s effect on drug-rewarding centers in the brain is much more enhanced during early adolescence compared to later adolescence or adulthood (Kwan et al., 2020; Leslie, 2020). As such, DiFranza et al., 2007) also documented an instance where 6th graders reported dependence symptoms within days of first inhaling a cigarette (DiFranza et al., 2007). Collectively, this research suggests that age at first use is inversely related to time to nicotine dependence.

4.2. Exclusive e-cigarette use vs. Exclusive cigarette smoking

Preliminary evidence for the time to first report of dependence symptoms for exclusive e-cigarette use was compared to that for exclusive cigarette use to examine the abuse liability potential of e-cigarettes among youth with P30D use. As hypothesized, exclusive cigarette use was associated with a more rapid time to first report of the early dependence symptom really needing to use than exclusive e-cigarette use. About 25% of youth with exclusive cigarette use who first reported really needing to use did so 7 weeks earlier than 25% of youth with exclusive e-cigarette use. However, relative to exclusive cigarette use, the time to first reporting of severe (i.e., use within 30 minutes of waking) and early (i.e., cravings) dependence symptoms were similar for youth with exclusive e-cigarette use. This evidence is worrying because it suggests that for youth who report exclusive P30D use, the time to dependence (particularly severe dependence) for those with exclusive e-cigarette use may be similar to those with exclusive cigarette use.

Nicotine in combustible cigarettes is addictive (Benowitz, 2008) and adult studies suggest that persons who smoke cigarettes on average show significantly faster and greater levels of nicotine absorption relative to those who use e-cigarettes (Yingst et al., 2019; Goldenson et al., 2020, 2021). Therefore, we would expect persons with exclusive cigarette use to report symptoms of nicotine dependence earlier and more rapidly than those with exclusive e-cigarette use. However, the amount of nicotine that is delivered and ultimately available in the blood after e-cigarette use could vary widely (from negligible to equivalent or higher than cigarettes) based on user experience, product type, nicotine formulation, etc., even after accounting for individual-level variation and use frequency over time (Benowitz, 2008; Trtchounian et al., 2010; Spindle and Eissenberg, 2018; Yingst et al., 2019; Talih et al., 2020, 2022; Goldenson et al., 2021; Prochaska et al., 2022). For example, the urinary cotinine levels (a surrogate measure of blood nicotine levels) detected among youth who use JUUL and other pod mod devices were higher than that in a similar youth population after combustible cigarette smoking (Goniewicz et al., 2019). Similarly, some adults who use high nicotine content e-cigarette devices record higher serum nicotine levels than the average reported for cigarette use (Ramôa et al., 2016; Yingst et al., 2019). Besides, the majority of adults in the previous studies who reported e-cigarette use, formerly used or currently use cigarettes (Yingst et al., 2019; Goldenson et al., 2021) compared to the tobacco naïve youth who reported exclusive e-cigarette use in this study. Therefore, given that nicotine’s effect on the brain is more pronounced during adolescence and young adulthood, it is plausible that among youth, the time to signs of nicotine dependence for those who report exclusive e-cigarette use is similar to that for youth with exclusive cigarette use. Particularly with the continued shift towards the sale of higher nicotine strength e-cigarette products, devices that are more efficient at nicotine delivery (Romberg et al., 2019; Ali et al., 2023) and increased user experience, the time to reporting dependence symptoms among youth who report P30D exclusive e-cigarette use could potentially parallel or be shorter than that for exclusive cigarette use over time.

Further, compared to the estimated 440,000 (1.6%) youth who reported P30D cigarette use in 2022 (Park-Lee, 2022), the overall impact of nicotine dependence and the abuse liability potential of e-cigarettes at the population level among youth who use e-cigarettes may be greater, given the about 2.5 million (9.4%) youth who reported P30D e-cigarette use. Notably, nicotine dependence and e-cigarette use among youth could predict future cigarette smoking too, as young people potentially take up other tobacco product use to satisfy a physiologic need for nicotine (Doubeni et al., 2010; Dierker et al., 2012, 2015; Primack et al., 2015; Soneji et al., 2017; Barrington-Trimis et al., 2020). Interestingly, 46% to 61% of participants who reported dependence outcomes were 18 years or older (Supplemental Table 2).

4.3. Strengths and limitations

To our knowledge, the current study provides preliminary estimates of the time to the first report of nicotine dependence symptoms among a nationally representative sample of youth with exclusive P30D e-cigarette use and compares this to youth with exclusive P30D cigarette use. The use of longitudinal analysis allows risk estimation to infer causality. Study limitations include 1) The lack of data on e-cigarette product characteristics (e.g., device type) across all waves in the PATH data set. Therefore, we could not assess to what extent these factors affected the study findings. 2) Inability to generalize findings to youth who report dual use over time or initiate tobacco use earlier than 12 years. 3) Data analyzed in this study spanned 2013–2019; therefore, findings may not fully reflect the nicotine dependence profile among contemporary youth, given recent trends in P30D use and e-cigarette product characteristics. 4) PATH does not provide data on the exact time for P30D use or dependence outcomes. Additionally, there is a risk that our estimates could be biased since our analytic approach to handling missing data assumes that the missing status for P30D use or dependence is non-P30D use or non-dependent. 5) All measures for the current study were self-reported and there is a likelihood of information bias. However, self-reported data in PATH have been shown to be reliable (kappa = 0.79) and valid (Tourangeau et al., 2019).

5. Conclusion

Study findings indicate that compared to exclusive P30D cigarette use (a known addictive product), the time to first report of nicotine dependence (use within 30 minutes and cravings) for youth with exclusive P30D e-cigarette use is comparable. Therefore, the abuse liability potential (i.e., the likelihood that a product will be abused, cause dependence, or result in adverse health outcomes) (Carter et al., 2009), of e-cigarettes among youth should be considered by policymakers and regulatory bodies in their review of these products as youth e-cigarette use could potentially undermine the current gains in reducing cigarette smoking at the population level (General, U.S.P.H.S.O. Of the S, 2012). Further research is warranted to understand how evolving e-cigarette products (e.g., brands) and product characteristics (e.g., nicotine content, device type) impact the time to first report of nicotine dependence symptoms among youth with e-cigarette use (Adjei et al., 2024).

Supplementary Material

Supplementary data to this article can be found online at https://doi.org/10.1016/j.ypmed.2024.107924.

Data availability

PATH Restricted-Use Files (RUF) were analyzed in this study. This data is not publicly available due to the sensitive nature of the information collected in the survey (United States Department of Health and Human Services, 2019).