Time-varying Determinants of Changes in E-cigarette Relative Harm Perception among US Young Adults

doi:10.21203/rs.3.rs-2873479/v2

Background

Current e-cigarette use has increased among young adults in the United States (US) despite a consistent decrease in perceiving e-cigarettes as less harmful than cigarettes over time. This study examined time-varying predictors associated with the changes in e-cigarette relative harm perception over time among US young adults.

Methods

Data were from the 2013-2018 Population Assessment of Tobacco and Health (PATH) Study for young adults (18-24 years). A time-varying effect model (TVEM) was applied to examine the association between the relative harm perception change and the associated time-varying predictors.

Results

Of the 8,427 young adults, the prevalence of those who perceived e-cigarettes as less harmful than cigarettes decreased from 50.3% in Wave 1 (2013-2014) to 27.7% in Wave 4 (2016-2018). Young adults who were male were more likely to perceive e-cigarettes as less harmful than cigarettes over time (OR=1.58; 95%CI: 1.53-1.64). In addition, the changes in e-cigarette relative harm perception were less noticeable among those with advanced degrees, who had non-combustible smoke-free home rules, who held negative tobacco-related attitudes, and those who were current e-cigarette users or ever used alcohol (all p-values<0.05).

Conclusions

A decline was observed in US young adults who perceived e-cigarettes as less harmful than cigarettes from the PATH Study across four waves (2013-2018). The study findings underscore the importance of risk communication that focuses on harm perception profiles and the need for appropriate interventions to balance the considerations of e-cigarette use among young adults.

Relative harm perception

e-cigarette

electronic cigarette

ENDS

young adults

E-cigarettes are battery-powered novel devices that simulate tobacco smoking by delivering nicotine in the form of aerosol to the users. E-cigarette has become the most popular used tobacco product among young people in the United States (US) [1]. The Morbidity and Mortality Weekly Report showed that 3.7% (9.1 million) of US adults reported current (every day or some days) use of e-cigarettes in 2020 [2]. Furthermore, young adults (18-24 years) who reported current use of e-cigarettes considerably increased from 2.4% in 2012 to 9.4% in 2020 [2,3] and had the highest prevalence of e-cigarette use across all adult age groups [4]. There is a consistent interest in understanding how e-cigarette use appeals to young people.

Many factors have been found to drive e-cigarette use for youth, such as flavors, social media, social influences, etc. [5]. In contrast, the most often cited reason for e-cigarette use for adults was smoking cessation, followed by reduced health risks of smoking-related diseases [6]. Literature suggests that e-cigarettes are potentially less harmful than cigarettes because e-cigarette aerosol contains fewer toxic chemicals and e-cigarette use is associated with less short-term effects on human health compared to cigarette smoking [7]. E-cigarettes may have potential benefits of encouraging smoking cessation among young adults, but dual use of e-cigarettes and cigarettes in this population is common [4], which may carry additional risks compared to using either product alone [8]. To date, no e-cigarette device has been approved as a smoking cessation tool in the US, and the long-term side effects of e-cigarette use are still unclear [9].

Harm perception is an important determinant that impacts e-cigarette use and transitions between other tobacco products [11, 12], which may influence the decision to seek e-cigarettes as a smoking cessation tool [6]. Relative harm perception of e-cigarettes refers to the comparison of harms between e-cigarettes and other tobacco products, such as combustible cigarettes [12-14]. Previous literature found that one commonly cited reason for e-cigarette uptakes is the belief that e-cigarettes are less harmful than cigarettes [13,14]. A longitudinal study observed that perceiving e-cigarettes as less harmful than cigarettes predicted e-cigarette initiation among non-smokers [10]. Another study found that many cigarette smokers who want to quit smoking choose e-cigarettes as their goal-directed way to quit because of the reduced harms of e-cigarettes compared to cigarettes [6]. Of great importance, understanding the dynamics of e-cigarette relative harm perception may help better characterize those young adults with different cigarette and e-cigarette use statuses and inform policymakers to facilitate harm awareness strategies in balancing the benefits of e-cigarettes to those smokers who are motivated to quit and the unintended consequences, e.g., dual use of e-cigarettes and cigarettes.

A recent study, using two multiyear cross-sectional nationally representative surveys, found that over time, there was a decrease in the perception of e-cigarettes as less harmful than cigarettes and a substantial increase in the perception of e-cigarettes as same or more harmful than cigarettes among US adults aged 18 years or older [15]. Additionally, existing research has identified individual-level predictors of harm perception, with often conflicting findings [16-20]. However, there remains a dearth of population-level research exploring the predictors of changes in e-cigarette harm perception among young adults. Such studies could help identify groups more likely to use e-cigarettes through long-term monitoring of harm perception profile. A recent work conducted among adolescents from the Population Assessment of Tobacco and Health (PATH) Study assessed the associations between changes in e-cigarette harm perception and predictors using a time-varying effect method [21, 22], which depicted a dynamic picture of how time-varying predictors affected harm perception over time. Building on this line of work, this current study will analyze the PATH data from 2013-2018 for US young adults to examine the time-varying determinants of e-cigarette relative harm perception over a five-year period. The findings from this study may be crucial for tailored interventions (e.g., incorporating modifiable factors and health communication) designed to reduce e-cigarette use without lessening the benefits of e-cigarettes to those smokers who are interested in smoking cessation.

Study Population

Data were from Waves 1-4 (2013-2018) of the Population Assessment of Tobacco and Health (PATH) Study for young adults aged 18-24 years old and followed up annually. The PATH Study is a nationally representative and longitudinal cohort of US youth and adults to understand factors associated with changes in tobacco use and related health outcomes [23]. Self-administered audio computer-assisted self-interview questionnaires were used to collect tobacco-use information and associated health behaviors among adults. The retention rates for Wave 2, Wave 3 and Wave 4 among Wave 1 adult respondents were 82.8%, 78.0% and 72.9%, respectively. The detailed procedures of the PATH Study can be found elsewhere [21,23].

Measures

Outcome

Relative harm perception was assessed prospectively across Waves 1-4. Participants were first asked, "Have you ever seen or heard of an e-cigarette before this study?" at Wave 1. Only those who responded "Yes" were followed up with the question, "Is using e-cigarettes less harmful, about the same, or more harmful than smoking cigarettes?" Literature shows that perceiving lower levels of e-cigarette harm was one of the main drivers for e-cigarette initiation [24], so we dichotomized the responses as less harmful vs. same or more harmful (reference group). This method was also guided by the previous research [21, 25, 26], which provides a simplified model for the analysis and interpretation of results.

Time-invariant and time-variant predictors

The selection of variables was guided by the Host, Agent, Vector, Environment (HAVE) conceptual model of the PATH study [23] as well as a review of previous literature [16, 17-21, 26, 27]. According to the HAVE model, the interactions among these four domains impact behavioral and health outcomes. In our study, host factors refer to individuals who are tobacco users or at risk of being tobacco users (demographics, tobacco-related attitudes, current cigarette/e-cigarette use, ever other tobacco/drug/alcohol use, and mental/physical health). Agent factors are the tobacco products' design, formulation, packaging, and promotions (media use and health warning exposure). Environmental factors include current policies, social, cultural, and geographic influences (smoke-free home rules and environmental/peer influence). The vector facilitates interaction between host, agent, and environments. The variables were further categorized into time-invariant and time-variant predictors.

Time-invariant variables included sex and race/ethnicity. Race/ethnicity was coded as non-Hispanic White, non-Hispanic Black, Hispanic, and non-Hispanic Other.

The rest of the predictors were measured as time-variant variables. Specifically, education was categorized into four levels: less than high school, high school, some college or college, and advanced degree. The assessment of environmental/peer influence used the question, "Does anyone who lives with you use any form of tobacco product?" Smoke-free home rules were measured by asking if participants were allowed to smoke a combustible/non-combustible tobacco product inside their homes. To assess tobacco-related attitudes, young adult participants were asked, "Thinking about the people who are important to you, how would you describe their opinion on using tobacco?". The answers were recoded as positive/neutral or negative. Current cigarette/e-cigarette use was measured by asking if participants had used them at least once during the past month. Ever other tobacco use was evaluated by asking, "Have you ever used cigars, smokeless tobacco, traditional cigars, filtered cigars, pipe tobacco, snus pouches, dissolvable tobacco, bids, kreteks, hookah or cigarillos?" Ever drug use was assessed by asking if they had ever used marijuana, prescription drugs, cocaine or crack, stimulants, heroin, inhalants, solvents, and hallucinogens. Ever alcohol use was assessed by asking if they had ever used alcohol. Mental/physical health was evaluated by asking, "In general, would you say your mental/physical health is excellent, very good, good, fair, and poor?" Health warning exposure was assessed by asking if they had ever noticed the health warning messages on packages of e-cigarettes and cigarettes.

For media use, participants were asked whether they had ever signed up for email alerts about tobacco products, read articles online about tobacco products, or watched a video online about tobacco products in the past 6 months. The poverty level was guided by the Department of Health and Human Services poverty guidelines and coded as below/at or above poverty levels. Since the questions of these two variables were only asked in Wave 1, for the analysis, we treated media use and poverty level as time-invariant predictors.

Data analysis

The analysis plan was conducted in four steps, similar to a previous study [21]. First, descriptive statistics for sample characteristics were reported as numbers and percentages for Waves 1-4. Second, four waves were combined by the PERSONID, and the missing outcome at Wave 1 was excluded. Third, except for race, sex, poverty level, and media use, we coded the other variables as time-variant predictors, meaning that these variables could vary across the time (wave) metric and were not constrained to be equal across all four waves. Fourth, the time-variant and time-invariant predictors were examined using multilevel modeling as it is the most widely used analytic approach for intensive longitudinal data with time-variant and time-invariant covariates [28]. We applied a time-varying effect model (TVEM) to our study by using the macro %TVEM (available from http://methodology.psu.edu) [28]. The macro %TVEM used all data for each individual across four waves and automatically excluded the missing time-specific observations in the analysis [29]. We reported adjusted odds ratios (ORs) (95% confidence intervals; CIs), and the statistical significance was set at α=0.05 for 2-sided tests for all analyses. Data analyses were conducted using SAS version 9.4 statistical software.

Sample characteristics

Among 8,427 US young adults, 50.3% perceived e-cigarettes as less harmful than cigarettes in Wave 1 (2013-2014), and this percentage decreased in Wave 2 (2014-2015; 39.3%), Wave 3 (2015-2016; 31.3%) and Wave 4 (2016-2018; 27.7%). At Wave 1, most participants were male (50.9%), Non-Hispanic White (53.0%), not current cigarette (59.5%) and e-cigarette users (82.2%), did not have combustible (74.1%) and non-combustible smoker-free home rules (63.3%), did not have peers using tobacco (62.8%), had negative tobacco-related attitudes (58.8%), ever used drug (55.5%) and alcohol (82.2%), never used other tobacco products (88.2%). Detailed information regarding sample characteristics from Waves 1-4 are presented in Table 1.

Predictors of harm perception

Young adults who were male (vs. female) (OR=1.58; 95%CI: 1.53-1.64) were more likely to perceive e-cigarettes as less harmful than cigarettes over time. Compared to non-Hispanic Whites, Hispanics (OR=0.72; 95%CI: 0.68-0.75) were less likely to report e-cigarettes as less harmful than cigarettes over time. Moreover, young adults above the poverty level (vs. below poverty) (OR=0.90; 95%CI: 0.87-0.94) were less likely to report e-cigarettes as less harmful than cigarettes over time. The detailed results are presented in Table 2.

Young adults who held an advanced degree (vs. less than high school) were more likely to perceive e-cigarettes as less harmful than cigarettes from Wave 1 (OR=1.28; 95%CI:1.01-1.63) to Wave 4 (OR=1.96; 95%CI:1.46-2.64). Current cigarette users were less likely to perceive e-cigarettes as less harmful than cigarettes from Wave 2 (OR=0.84; 95%CI:0.72-0.97) to Wave 4 (OR=0.83; 95%CI:0.69-0.99), while current e-cigarette users were more likely to perceive e-cigarettes as less harmful than cigarettes across four waves. Combustible smoke-free home rules were associated with a lower likelihood of perceiving e-cigarettes as less harmful than cigarettes across Waves 2-4, while non-combustible smoke-free home rules were associated with a higher likelihood of perceiving e-cigarettes as less harmful than cigarettes across four waves. Ever alcohol use was associated with a higher likelihood of perceiving e-cigarettes as less harmful than cigarettes from Waves 2-4. The detailed results are presented in Table 3 and Supplemental figure 1.

This longitudinal analysis of a nationally representative study of US young adults revealed a consistent decline in the perception of e-cigarettes as less harmful than cigarettes from 2013 to 2018. In other words, over time, young adults in the study perceived increased harms associated with e-cigarette use. These changes in relative harm perception were less pronounced among males, those with advanced degrees, those with non-combustible smoke-free home rules, those holding negative tobacco-related attitudes, and those who were current e-cigarette users or had ever used alcohol. This information can help us better understand how the specific modifiable factors impact the evolution of e-cigarette harm perception over time and plan appropriate risk communication interventions to influence the choice of e-cigarette use among young adults.

This study has observed a decline in the perception of e-cigarettes as less harmful than cigarettes among young adults over time, consistent with previous studies conducted in the US [16, 21] and the UK [17, 30]. The former two studies conducted in the US focused on the adult population across all age groups or adolescents, while the latter two studies conducted in the UK investigated harm perception among both adolescents and adults. Unlike those studies, our study focused on a nationally representative sample of young adults who are more likely to use e-cigarettes than other age groups in the US [4]. The observed changes in harm perception may be attributed to emerging evidence of acute respiratory symptoms and lung injury associated with e-cigarette use, substance use (e.g., marijuana), as well as concerns about the risk of nicotine dependence related to e-cigarettes [31, 32]. The decline in relative harm perception may also be due to the regulations and e-cigarette prevention campaigns aimed at discouraging e-cigarette use among young people [33]. Furthermore, we observed that the decrease in relative harm perception of e-cigarettes was accompanied by a slight decline in current cigarette use among young adults. In addition, although the odds decreased over time, current e-cigarette users were more likely, while current cigarette users were less likely, to perceive lower e-cigarette relative harms. This is consistent with previous research [34], demonstrating that the relative harm perception of e-cigarettes may influence the decisions of non-users on e-cigarette use initiation or smokers on whether to use e-cigarettes as a cessation tool. Moreover, the decrease in e-cigarette harm perception may also indicate a transition to more cigarette use or dual use of e-cigarettes and cigarettes. Future studies should monitor changes in e-cigarette harm perception and their impact on young people with different tobacco use patterns.

In line with a study conducted among US adolescents [21], our study found that young adult males were more likely than females to perceive relative harm perception of e-cigarettes over time. This is also similar to the findings of another study, which indicates that sex differences in harm perception change may be influenced by industry advertising [35]. Males may also be more likely to use social media to seek health information, attitudes, and behaviors related to health than females [5]. In contrast, minority populations such as Hispanics were less likely to report e-cigarettes as less harmful than cigarettes over time. Minorities may have a different perception of the harm caused by e-cigarettes because they were targeted by aggressive advertising campaigns that may have modified their harm perception [36].

We also observed that higher education is associated with perceiving lower relative harms of e-cigarettes, which is consistent with a previous study that focused on adolescents and found that educated individuals may be more likely to have knowledge about harm reduction regarding e-cigarette use due to the availability of information [21]. Additionally, allowing non-combustible tobacco use inside the home was associated with perceiving lower relative harms of e-cigarettes. Although the US Food and Drug Administration regulated e-cigarettes as tobacco products in 2016, many e-cigarette users do not consider them as such in real-world settings [37]. The relationship between non-combustible smoke-free home rules and harm perception change might be related to the common perception of e-cigarettes as a non-tobacco product and a healthier alternative to combustible tobacco products. Those with negative tobacco-related attitudes were more likely to report lower relative harm perception over time, indicating awareness of the harms of traditional tobacco products compared to e-cigarettes. However, many still hold favorable attitudes toward e-cigarettes. It is also worth noting that ever having used alcohol was associated with perceiving e-cigarettes as less harmful than cigarettes over time, which may potentially lead to polysubstance use [38]. With the growing popularity of e-cigarette use among US young adults, the results of this study highlight the importance of appropriate interventions (e.g., health communication of e-cigarette-related harms to the public) that need to be implemented to balance the role of e-cigarette use between smokers and those who are tobacco-naïve.

Limitations

While assessing harm perception evolution and its predictors among representative samples of young adults are clear strengths of our analysis, this study is not without limitations. First, as we mentioned, media use was only assessed at baseline, which may underestimate the true relationship between it and the outcome. Second, we only evaluated the e-cigarette relative harm perception. Harm perception related to other specific aspects (e.g., health effects) is warranted in future studies. Lastly, tobacco use behaviors from the PATH Study were measured using self-reported questions, which may introduce response bias.

We observed a consistent decrease in US young adults who perceived e-cigarettes as less harmful than combustible cigarettes from 2013 to 2018. However, these changes in e-cigarette relative harm perception were less pronounced among males, individuals with advanced degrees, those who had non-combustible smoke-free home rules, those with negative tobacco-related attitudes, and those who were current e-cigarette users or had ever used alcohol. Lower relative harm perception regarding e-cigarettes may encourage initiation of e-cigarette use among tobacco-naïve individuals, while higher harm perception may discourage smokers from quitting and lead to continued cigarette use or dual use of cigarettes and e-cigarettes. To better understand the mechanism behind these changes in harm perception, it is important to continue monitoring these trends and identifying predictors. This information can inform tailored tobacco control efforts to effectively balance the consideration of e-cigarette use among young adults.

Informed Consent Not applicable

Ethical Approval The Institutional Review Boards review and approval was not needed for this study since it used a public dataset available at https://doi.org/10.3886/Series606.

The Welfare of Animals Not Applicable

Funding No funding was secured for this study.

Conflict of Interest The authors declare no competing interests or conflicts.

Gentzke AS, Wang TW, Cornelius M, et al. Tobacco Product Use and Associated Factors Among Middle and High School Students - National Youth Tobacco Survey, United States, 2021. MMWR. 2022; 71(5): 1–29.
Cornelius ME, Loretan CG, Wang TW, Jamal A, Homa DM. Tobacco Product Use Among Adults - United States, 2020. MMWR. 2022; 71(11): 397-405.
Agaku IT, King BA, Husten CG, et al. Tobacco product use among adults--United States, 2012-2013. MMWR. 2014; 63(25): 542-547.
Boakye E, Osuji N, Erhabor J, et al. Assessment of Patterns in e-Cigarette Use Among Adults in the US, 2017-2020. JAMA Netw Open. 2022;5(7):e2223266.
Amin S, Dunn AG, Laranjo L. Social Influence in the Uptake and Use of Electronic Cigarettes: A Systematic Review. Am J Prev Med. 2022; 58(1): 129-141.
Sapru S, Vardhan M, Li Q, Guo Y, Li X, & Saxena D. E-cigarettes use in the United States: reasons for use, perceptions, and effects on health. BMC Public Health. 2020; 20(1): 1518.
Marques P, Piqueras L, Sanz MJ. An updated overview of e-cigarette impact on human health. Respir Res. 2021;22(1): 151.
Alzahrani T, Pena I, Temesgen N, Glantz SA. Association Between Electronic Cigarette Use and Myocardial Infarction. Am J Prev Med. 2018; 55(4): 455-461.
US Food & Drug Administration, 2022. E-Cigarettes, Vapes, and other Electronic Nicotine Delivery Systems (ENDS). https://www.fda.gov/tobacco-products/products-ingredients-components/e-cigarettes-vapes-and-other-electronic-nicotine-delivery-systems-ends#:~:text=To%20date%2C%20no%20e%2Dcigarette,adults%20who%20use%20tobacco%20products. Accessed January 26, 2023.
Cooper M, Loukas A, Case KR, Marti CN, Perry CL. A longitudinal study of risk perceptions and e-cigarette initiation among college students: Interactions with smoking status. Drug Alcohol Depend. 2018; 186: 257-263.
Hammig B, Daniel-Dobbs P, & Blunt-Vinti H. Electronic cigarette initiation among minority youth in the United States. Am J Drug Alcohol Abuse. 2017; 43(3): 306–310.
Persoskie A, O'Brien EK, Poonai K. Perceived relative harm of using e-cigarettes predicts future product switching among US adult cigarette and e-cigarette dual users. Addiction. 2019; 114(12): 2197-2205.
Pepper JK, Ribisl KM, Emery SL, Brewer NT. Reasons for starting and stopping electronic cigarette use. Int J Environ Res Public Health. 2014; 11(10):10345-61.
Rass O, Pacek LR, Johnson PS, Johnson MW. Characterizing use patterns and perceptions of relative harm in dual users of electronic and tobacco cigarettes. Exp Clin Psychopharmacol. 2015; 23(6): 494-503.
Huang J, Feng B, Weaver SR, et al. Changing Perceptions of Harm of e-Cigarette vs Cigarette Use Among Adults in 2 US National Surveys From 2012 to 2017. JAMA Netw Open. 2019; 2(3): e191047.
Majeed BA, Weaver SR, Gregory KR, et al. Changing Perceptions of Harm of E-Cigarettes Among U.S. Adults, 2012-2015. Am J Prev Med. 2017; 52(3): 331-338.
Filippidis FT, Laverty AA, Gerovasili V, & Vardavas CI. Two-year trends and predictors of e-cigarette use in 27 European Union member states. Tob Control. 2017; 26(1): 98–104.
Kimber C, Frings D, Cox S, Albery IP, Dawkins L. Communicating the relative health risks of E-cigarettes: An online experimental study exploring the effects of a comparative health message versus the EU nicotine addiction warnings on smokers’ and non-smokers’ risk perceptions and behavioural intentions. Addict Behav. 2020; 101: 106177.
Pokhrel P, Fagan P, Kehl L, Herzog TA. Receptivity to e-cigarette marketing, harm perceptions, and e-cigarette use. Am J Health BehaV. 2015; 39(1), 121-31.
Yong HH, Borland R, Balmford J, et al. Prevalence and Correlates of the Belief That Electronic Cigarettes are a Lot Less Harmful Than Conventional Cigarettes Under the Different Regulatory Environments of Australia and the United Kingdom. Nicotine & Tob Res. 2017; 19(2): 258-263.
Li W, Osibogun O, Li T, Sutherland MT, Maziak W. Changes in harm perception of ENDS and their predictors among US adolescents: findings from the population assessment of tobacco and health (PATH) study, 2013-2018. Prev Med. 2022; 155: 106957.
Baird R, Maxwell SE. Performance of time-varying predictors in multilevel models under an assumption of fixed or random effects. Psychol Methods. 2016; 21(2): 175-88.
Hyland A, Ambrose BK, Conway KP, et al. Design and methods of the Population Assessment of Tobacco and Health (PATH) Study. Tob Control. 2017; 26(4): 371-378.
Malt L, Verron T, Cahours X, Guo M, Weaver S, Walele T, O'Connell G. Perception of the relative harm of electronic cigarettes compared to cigarettes amongst US adults from 2013 to 2016: analysis of the Population Assessment of Tobacco and Health (PATH) study data. Harm Reduct J. 2020; 17(1): 65.
Amrock SM, Zakhar J, Zhou S, & Weitzman M. Perception of e-cigarette harm and its correlation with use among U.S. adolescents. Nicotine & Tob Res. 2015; 17(3): 330–336.
Li W, Osibogun O, Gautam P, Li T, Cano MA, Maziak W. Effect of harm perception on ENDS initiation among US adolescents and young adults: Longitudinal findings from the population assessment of tobacco and health (PATH) study, 2013–2018. Drug Alcohol Depend. 2023; 244: 109784.
US Department of Health and Human Services (USDHHS). National Institutes of Health. National Institute on Drug Abuse, and United States Department of Healthand Human Services. Food and Drug Administration. Center for Tobacco Products. Population Assessment of Tobacco and Health (PATH) Study [United States] Public-Use Files. Inter-university Consortium for Political and Social Research [distributor], 2021-11-11. https://doi.org/10.3886/ICPSR36498.v15
Shiyko MP, Lanza ST, Tan X, Li R, Shiffman S. Using the time-varying effect model (TVEM) to examine dynamic associations between negative affect and self confidence on smoking urges: differences between successful quitters and relapsers. Prev Sci. 2012; 13(3): 288-299.
Dermody SS, Shiffman S. The time-varying effect of alcohol use on cigarette smoking relapse risk. Addict Behav. 2020; 102: 106192.
Perski O, Beard E, & Brown J. Association between changes in harm perceptions and e-cigarette use among current tobacco smokers in England: a time series analysis. BMC Med. 2020; 18(1): 98.
US Department of Health and Human Services (USDHHS). E-Cigarette Use Among Youth and Young Adults: A Report of the Surgeon General. https://e-cigarettes.surgeongeneral.gov/documents/2016_sgr_full_report_non-508.pdf. Accessed January 26, 2023.
National Academies of Sciences Engineering, Medicine, Health, and Medicine (NASEM). Division Board on Population Health, Public Health Practice Committee on the Review of the Health Effects of Electronic Nicotine Delivery Systems. In D. L. Eaton, L. Y. Kwan, & K. Stratton (Eds.), Public Health Consequences of E-Cigarettes. Washington (DC): National Academies Press (US) Copyright 2018 by the National Academy of Sciences. All rights reserved. 2018.
US Food and Drug Administration (FDA). The Real Cost E-Cigarette Prevention Campaign. https://www.fda.gov/tobacco-products/real-cost-campaign/real-cost-e-cigarette-prevention-campaign. Accessed January 26, 2023.
Wackowski OA, & Delnevo CD. Young Adults' Risk Perceptions of Various Tobacco Products Relative to Cigarettes: Results From the National Young Adult Health Survey. Health Educ Behav. 2016; 43(3): 328–336.
Richardson A, Ganz O, Vallone D. Tobacco on the web: surveillance and characterisation of online tobacco and e-cigarette advertising. Tob Control. 2015; 24(4): 341-347.
Rapp JL, Alpert N, Wilson KM, Flores RM, Taioli E. Changes in E-Cigarette Perceptions Over Time: A National Youth Tobacco Survey Analysis. Am J Prev Med. 2021; 61(2): 174-181.
Munafò M. Are e-Cigarettes Tobacco Products? Nicotine & Tob Res. 2018; 21: 267-67.
Parikh AS, Bhattacharyya N. Patterns of concurrent cigarette, alcohol, and e-cigarette use: Off-setting or additive behaviors? Laryngoscope. 2018; 128(8): 1817-1821.

Table 1. Descriptive statistics among US young adults (18-24 years): Population Assessment of Tobacco and Health Study (n=8,472).

Participant Characteristics	Wave 1 (2013-2014) N (%)	Wave 2 (2014-2015) N (%)		Wave 3 (2015-2016) N (%)		Wave 4 (2016-2018) N (%)
Relative harm perception
Less harmful	4,260 (50.3)		2,669 (39.3)		1,980 (31.3)		1,606 (27.7)
Same or more harmful	4,212 (49.7)		4,130 (60.7)		4,352 (68.7)		4,190 (72.3)
Sex (Male)	4,315 (50.9)		-		-		-
Race/ethnicity
Non-Hispanic White	4,489 (53.0)		-		-		-
Non-Hispanic Black	1,239 (14.6)		-		-		-
Hispanic	1,935 (22.8)		-		-		-
Non-Hispanic Others	809 (9.6)		-		-		-
Education
Less than high school	1,532 (18.2)		1,001 (14.7)		870 (13.7)		783 (13.5)
High school	2,578 (30.6)		1,757 (25.8)		1,522 (24.0)		1,274 (21.9)
Some college or college	3,460 (41.0)		3,050 (44.8)		2,715 (42.8)		2,377 (40.9)
Advanced degree	864 (10.2)		1,008 (14.8)		1,230 (19.4)		1,382 (23.8)
Poverty* (Below poverty level)	3,774 (50.2)		-		-		-
*Media use^#** (Yes)	750 (9.1)		-		-		-
Combustible smoke-free home rules (Yes)	2,189 (25.9)		1,687 (24.7)		1,442 (22.7)		1,229 (21.1)
Non-combustible smoke-free home rules (Yes)	3,087 (36.7)		2,491 (36.6)		2,010 (31.9)		1,774 (30.6)
Environment influence^¶ (Yes)	3,804 (47.2)		2,673 (43.1)		2,345 (41.8)		1,983 (38.4)
Tobacco-related attitudes^§
Positive/Neutral	2,705 (41.2)		2,190 (38.9)		1,919 (40.7)		1,944 (37.8)
Negative	3,855 (58.8)		3,437 (61.1)		2,801 (59.3)		3,206 (62.3)
Current cigarette use (Yes)	3,432 (40.5)		2,762 (40.3)		2,454 (38.6)		2,157 (37.0)
Current e-cigarette use (Yes)	1,505 (17.8)		1,054 (15.8)		1,205 (18.9)		862 (14.8)
Ever used other tobacco products^† (Yes)	997 (11.8)		712 (10.4)		474 (7.5)		417 (7.2)
Ever used drug^‡ (Yes)	4,647 (55.5)		2,108 (30.9)		2,004 (31.6)		1,913 (35.0)
Ever used alcohol (Yes)	6,941 (82.2)		5,007 (73.3)		4,668 (73.5)		4,275 (73.4)
Health warning exposure (Yes)	5,734 (67.8)		4,650 (68.1)		4,125 (65.0)		3,836 (65.8)
Physical health
Excellent, very good or good	7,389 (87.3)		6,000 (87.7)		5,606 (88.2)		5,201 (89.2)
Fair or poor	1,077 (12.7)		844 (12.3)		749 (11.8)		628 (10.8)
Mental health
Excellent, very good or good	6,831 (80.8)		5,604 (81.9)		5,148 (81.0)		4,706 (80.8)
Fair or poor	1,627 (19.2)		1,236 (18.1)		1,208 (19.0)		1,121 (19.2)

Note: Numbers may not sum to the total due to missing data and loss to follow-up in subsequent waves.

*Poverty level and media use were only assessed in Wave 1 because the same questions were not available in the subsequent waves.

^#Media use was measured by asking if participants had ever signed up for email alerts about tobacco products, read articles online about tobacco products, or watched a video online about tobacco products in the past 6 months.

^¶Environmental influence was assessed by asking if anyone lived with participants used any form of tobacco product.

^§Tobacco-related attitudes variable was assessed by asking the opinion on using tobacco among people who were important to the participants. If they responded as “Negative” or “Very negative”, they were considered as negative to tobacco.

^†Other tobacco products refer to cigars, smokeless tobacco, traditional cigars, filtered cigars, pipe tobacco, snus pouches, dissolvable tobacco, bids, kreteks, hookah or cigarillos.

^‡Drugs refers to the ever use of marijuana, prescription drugs, cocaine or crack, stimulants, heroin, inhalants, solvents, and hallucinogens.

Table 2. Time-invariant predictors of e-cigarette harm perception among US young adults (18-24 years): longitudinal analysis of the Population Assessment of Tobacco and Health Study, 2013-2018.

Predictor variables	Relative harm perception^#
Predictor variables	AOR (95%CI)	P-value
Gender (Male vs Female)	1.58 (1.53-1.64)**	<.0001
Race
Non-Hispanic White	Ref.
Non-Hispanic Black	0.91 (0.86-0.96)	0.0762
Hispanic	0.72 (0.68-0.75)**	<.0001
Non-Hispanic Others	1.12 (1.06-1.19)	0.0526
Poverty (At or above poverty level vs below level)*	0.90 (0.87-0.94)*	0.0037
Media use (Yes vs No)*	1.03 (0.98-1.09)	0.5481

AOR= Adjusted Odds Ratio; CI= Confidence Interval; Ref.= Reference category

^#The reference group for relative harm perception is e-cigarettes as same or more harmful than cigarettes.

*Poverty level and media use were originally treated as time-variant predictors. However, due to the unavailable data after Wave 1, we assessed them from baseline (Wave 1) only and reported them together with the time-invariant variables.

NOTE: Bold estimates indicate statistical significance at P<0.05. *P<0.05, **P<0.001

Table 3. Time-variant predictors^† of e-cigarette harm perception among US young adults (18-24 years): longitudinal analysis of the Population Assessment of Tobacco and Health Study, 2013-2018.

Predictor variables	Relative harm perception^#
Predictor variables	Wave 1 AOR (95%CI)	Wave 2 AOR (95%CI)	Wave 3 AOR (95%CI)	Wave 4 AOR (95%CI)
Education (Ref.= Less than high school)
High school	1.11 (0.94-1.32)	1.32 (1.07-1.64)*	1.13 (0.87-1.47)	1.06 (0.79-1.41)
Some college or college	1.10 (0.93-1.29)	1.56 (1.27-1.91)*	1.46 (1.14-1.87)*	1.49 (1.14-1.94)*
Advanced degree	1.28 (1.01-1.63)*	1.47 (1.14-1.88)*	1.59 (1.19-2.13)*	1.96 (1.46-2.64)*
Combustible smoke-free home rules (Ref.=No)	0.88 (0.76-1.02)	0.80 (0.68-0.95)*	0.78 (0.64-0.95)*	0.64 (0.51-0.80)*
Non-combustible smoke-free home rules (Ref.=No)	1.66 (1.45-1.90)*	1.72 (1.47-2.00)*	1.62 (1.36-1.92)*	1.82 (1.52-2.19)*
Environment influence (Ref.=No)	1.14 (1.00-1.29)*	1.02 (0.89-1.18)	1.15 (0.98-1.34)	1.10 (0.93-1.31)
Tobacco-related attitude (Ref.= Positive/Neutral)	1.12 (0.99-1.26)	1.17 (1.02-1.34)*	1.24 (1.06-1.45)*	1.32 (1.11-1.56)*
Current cigarette use (Ref.=No)	0.90 (0.79-1.03)	0.84 (0.72-0.97)*	0.83 (0.70-0.98)*	0.83 (0.69-0.99)*
Current e-cigarette use (Ref.=No)	2.22 (1.91-2.56)*	1.59 (1.36-1.87)*	1.34 (1.13-1.60)*	1.31 (1.08-1.59)*
Ever used other tobacco products (Ref.=No)	0.87 (0.71-1.06)	0.80 (0.65-0.99)*	0.66 (0.50-0.88)*	0.73 (0.54-0.99)*
Ever used drug (Ref.=No)	1.26 (1.11-1.43)*	1.14 (1.00-1.30)	1.09 (0.93-1.27)	1.50 (1.27-1.76)*
Ever used alcohol (Ref.=No)	1.19 (0.99-1.43)	1.21 (1.02-1.43)*	1.80 (1.46-2.22)*	1.61 (1.31-1.98)*
Health warning exposure (Ref.=No)	0.89 (0.78-1.02)	1.00 (0.87-1.16)	0.92 (0.79-1.08)	1.09 (0.92-1.28)
Mental health (Ref.=Fair/poor)	0.89 (0.77-1.03)	0.81 (0.69-0.96)*	0.81 (0.67-0.99)*	0.87 (0.71-1.06)
Physical health (Ref.=Fair/poor)	1.00 (0.84-1.19)	1.06 (0.87-1.29)	1.14 (0.89-1.46)	0.68 (0.53-0.87)*

^†Time-variant predictors were assessed from all four waves.

^#The reference group for relative harm perception is e-cigarettes as same or more harmful than cigarettes.

AOR= Adjusted Odds Ratio; CI= Confidence Interval; Ref.= Reference category

NOTE: Bold estimates indicate statistical significance at *P<0.05.

SupplementalFigure104242023.docx
Supplemental figure 1. Significant time-varying predictors of e-cigarette harm perception among US young adults (18-24 years): longitudinal analysis of the Population Assessment of Tobacco and Health Study, 2013-2018.

Time-varying Determinants of Changes in E-cigarette Relative Harm Perception among US Young Adults

Status:

Journal Publication

Version 2

Abstract

Introduction

Methods

Results

Discussion

Conclusions

Declarations

References

Tables

Table 1. Descriptive statistics among US young adults (18-24 years): Population Assessment of Tobacco and Health Study (n=8,472).

Supplementary Files

Status:

Journal Publication

Version 2