INTRODUCTION

For at least the past 5 years, e-cigarettes have been the most commonly used tobacco product among youth1. In 2019, more than a third of US middle and high students had ever used e-cigarettes and one fifth were current users2. Although e-cigarettes have been marketed as healthier alternatives to conventional cigarettes, their long-term health effects are not well documented, especially for youth users, and there is no consensus regarding their utility for smoking cessation3. Preventing youth tobacco uptake and established use are public health concerns that tend to garner more attention than promoting cessation, and e-cigarette use rates demonstrate their appeal to this population47. Because e-cigarettes normally contain nicotine, youth users are at risk for the harms associated with nicotine consumption, such as epigenetic changes to the brain making them prone to future substance use8. Thus, better understanding factors that influence e-cigarette use in youth is important in preventing future use and its associated health implications, especially for youth vulnerable to tobacco use.

Appalachian youth are disproportionately exposed to tobacco. The Appalachian region has a history of tobacco culture from tobacco production to acceptance of tobacco use and celebrating this heritage (e.g. tobacco festivals)9. As a result, youth in Appalachia have higher smoking prevalence than youth residing in other parts of the US10. In addition to culture, rurality and poverty are associated with increased tobacco use11,12, possibly contributing to higher rates of use among Appalachia youth. Accordingly, Appalachian youth are at risk for tobacco use, including the use of e-cigarettes.

Studies have linked youth e-cigarette use to viewing e-cigarettes as less harmful or less addictive than combustible cigarettes57. Additional evidence suggests that youth perceive e-cigarettes as less harmful than other tobacco products, such as cigars and smokeless tobacco13. Although a relationship between perceiving e-cigarettes as less harmful or less addictive and e-cigarette use has been documented with some youth samples57, research on Appalachian youth is limited. One study found that Appalachian youth perceived e-cigarettes as causing fewer health problems and less addiction than conventional tobacco products14. Our study extends prior research by examining specific e-cigarette-related harms, such as whether e-cigarettes cause breathing and oral health problems, as well as whether these perceptions differ by tobacco use.

METHODS

Study design

The Youth Appalachian Tobacco Study (n=1074), a crosssectional survey of tobacco exposures, use patterns, and perceptions and attitudes, was conducted from 2014–2016. The Appalachian states Kentucky, North Carolina, and New York, were selected based on overall tobacco use rates (high, medium, and low, respectively). High school and middle school youth from Appalachian counties in each state were sampled, and participants completed a questionnaire during a regular school day. Detailed information on sampling and participants has been published elsewhere14. The study was approved by the University of Louisville’s Institutional Review Board.

Measures

E-cigarette harm perceptions

In the Youth Appalachian Tobacco Study, youth were asked to indicate whether they strongly disagree, disagree, neither agree nor disagree, agree, or strongly agree (using 1–5 rating scales) with the following six statements indicating that use of e-cigarettes: 1) causes health problems, 2) is healthier than smoking traditional cigarettes, 3) causes breathing problems, 4) causes oral health problems, 5) is addictive, and 6) is difficult to quit.

Tobacco use

Tobacco use was categorized into three groups: never users, ever non-e-cigarette users (i.e. cigarettes, smokeless tobacco), and ever e-cigarette users. Never users were participants who had never used cigarettes, e-cigarettes, or smokeless tobacco. Ever non-e-cigarette users were participants who had ever used cigarettes or smokeless tobacco but not e-cigarettes (referred to below as non-e-cigarette users). Ever e-cigarette users were participants who had ever used e-cigarettes (referred to below as e-cigarette users). The distribution of tobacco use is presented in Supplementary file Table S1.

Covariates

Gender (male, female), race/ethnicity (White/Caucasian, non-White/non-Caucasian), age, school type (middle, high), state (New York, Kentucky, North Carolina), and number of household tobacco users (zero, ≥1) were included in this analysis.

Statistical analysis

The distribution of each e-cigarette harm perception was calculated overall and by tobacco use. Pairwise comparisons were conducted between: 1) never users and e-cigarette users, 2) never users and non-e-cigarette users, and 3) non-e-cigarette users and e-cigarette users. Multivariable multinomial logistic regression models were used to estimate odds ratio (OR) and 95% confidence interval (CI) for associations between each harm perception as a categorical variable (Reference: strongly agree) and e-cigarette use (Reference: never users), adjusted for gender, race/ethnicity, school type, state, and household tobacco users. Because age may modify the association between harm perceptions and tobacco use, analyses were performed for the whole sample, followed by a sensitivity analysis excluding youth aged 11–12 years. Analyses were conducted in Stata 16.1.

RESULTS

About two-thirds (66.1%) of participants were never users, one-tenth (10.5%) were non-e-cigarette users, and one-fourth (23.4%) were e-cigarette users. The distribution of descriptive characteristics by tobacco use is provided in Supplementary file Table S2. More non-e-cigarette users were female (60.2%), and more e-cigarette users were male (53.4%). Both use groups were older than never users, and frequencies of non-e-cigarette and e-cigarette use varied by state.

Prevalence of e-cigarette harm perceptions by tobacco use is presented in Table 1. Never users strongly agreed at higher proportions that e-cigarettes cause health problems (34.5%), breathing problems (34.1%), and oral health problems (32.7%) as well as are addictive (37.3%) and difficult to quit (34.1%), compared to non-e-cigarette users and e-cigarette users. Non-e-cigarette users strongly agreed (25.7%) and agreed (29.2%) at higher proportions than never users that e-cigarettes are healthier than smoking traditional cigarettes; however, e-cigarette users most often strongly agreed (32.3%). Compared to never and non-e-cigarette users, a higher proportion of e-cigarette users strongly disagreed that e-cigarettes cause health problems (10.4%), breathing problems (9.6%), and oral health problems (11.6%) and are difficult to quit (12.0%). The distribution of harm perceptions differed between never users and e-cigarette users (p<0.05). However, when comparing never users to non-e-cigarette users, harm perceptions differed only for e-cigarettes cause health problems (p=0.020) and are difficult to quit (p=0.002). Comparisons between non-ecigarette users and e-cigarettes users were not statistically significant. Additionally, sensitivity analysis results suggest that age did not play a significant role in associations between harm perceptions and e-cigarette use.

Table 1

Prevalence of e-cigarette harm perceptions by tobacco use (N=1074)

Harm perceptionsTotal n (%)Never users n (%)Ever non-e-cigarette users* n (%)Ever e-cigarette users n (%)p
Population1074 (100)710 (66.1)113 (10.5)251 (23.4)
Causes health problems<0.001a
Strongly disagree65 (6.1)31 (4.4)8 (7.1)26 (10.4)0.020b
Disagree104 (9.7)48 (6.8)15 (13.3)41 (16.3)0.12c
Neither314 (29.2)192 (27.0)37 (32.7)85 (33.9)
Agree278 (25.9)194 (27.3)23 (20.4)61 (24.3)
Strongly agree313 (29.1)245 (34.5)30 (26.6)38 (15.1)
Is healthier than smoking traditional cigarettes0.020a
Strongly disagree105 (9.8)74 (10.4)12 (10.6)19 (7.6)0.78b
Disagree107 (10.0)67 (9.4)12 (10.6)28 (11.2)0.69c
Neither293 (27.3)212 (29.9)27 (23.9)54 (21.5)
Agree288 (26.8)186 (26.2)33 (29.2)69 (27.5)
Strongly agree281 (26.2)171 (24.1)29 (25.7)81 (32.3)
Causes breathing problems<0.001a
Strongly disagree57 (5.3)26 (3.7)7 (6.2)24 (9.6)0.34b
Disagree87 (8.1)45 (6.3)11 (9.7)31 (12.4)0.35c
Neither309 (28.8)196 (27.6)31 (27.4)82 (32.7)
Agree297 (27.7)201 (28.3)33 (29.2)63 (25.1)
Strongly agree324 (30.2)242 (34.1)31 (27.4)51 (20.3)
Causes oral health problems<0.001a
Strongly disagree65 (6.1)27 (3.8)9 (8.0)29 (11.6)0.15b
Disagree99 (9.2)53 (7.5)10 (8.9)36 (14.3)0.13c
Neither350 (32.6)219 (30.9)37 (32.7)94 (37.5)
Agree254 (23.7)179 (25.2)30 (26.6)45 (17.9)
Strongly agree306 (28.5)232 (32.7)27 (23.9)47 (18.7)
Is addictive<0.001a
Strongly disagree60 (5.6)33 (4.7)9 (8.0)18 (7.2)0.09b
Disagree86 (8.0)46 (6.5)12 (10.6)28 (11.2)0.89c
Neither268 (25.0)162 (22.8)29 (25.7)77 (30.7)
Agree302 (28.1)204 (28.7)33 (29.2)65 (25.9)
Strongly agree358 (33.3)265 (37.3)30 (26.6)63 (25.1)
Is difficult to quit<0.001a
Strongly disagree77 (7.2)36 (5.1)11 (9.7)30 (12.0)0.002b
Disagree105 (9.8)51 (7.2)17 (15.0)37 (14.7)0.97c
Neither315 (29.3)209 (29.4)35 (31.0)71 (28.3)
Agree259 (24.1)172 (24.2)27 (23.9)60 (23.9)
Strongly agree318 (29.6)242 (34.1)23 (20.4)53 (21.1)

* Ever non-e-cigarette users include cigarette and smokeless tobacco users who have not tried e-cigarettes.

a Chi-squared test comparing never users to ever e-cigarette users.

b Chi-squared test comparing never users to ever non-e-cigarette users.

c Chi-squared test comparing ever non-e-cigarette users to ever e-cigarette users.

Results from regression analyses can be found in Table 2, and values of OR with 95% CI are reported below. Compared to never users, e-cigarette users had greater odds of strongly disagreeing (OR=6.12; 95% CI: 3.16–11.90) and disagreeing (OR=4.78; 95% CI: 2.71–8.41) that e-cigarettes cause health problems. Also, compared to never users, non-e-cigarette users had greater odds of disagreeing (OR=2.30; 95% CI: 1.11–4.75) that e-cigarettes cause health problems. E-cigarette users also had more than fourfold odds of strongly disagreeing (OR=4.40; 95% CI: 2.26–8.56) and over threefold odds of disagreeing (OR=3.14; 95% CI: 1.76–5.60) that e-cigarettes cause breathing problems, compared to never users; no associations were found for non-e-cigarette users. Similarly, e-cigarette users (OR=6.01; 95% CI: 3.16–11.44) and non-e-cigarette users (OR=3.34; 95% CI: 1.37–8.15) had greater odds of strongly disagreeing that e-cigarettes cause oral health problems, compared to never users. Both e-cigarette and non-e-cigarette users had elevated odds of strongly disagreeing or disagreeing that e-cigarettes are addictive and difficult to quit. Associations for e-cigarettes being difficult to quit were similar with both groups having at least threefold odds of strongly disagreeing or disagreeing compared to never users.

Table 2

Multivariable multinomial logistic regression modeling associations between harm perceptions and ever non-e-cigarette use and harm perceptions and ever e-cigarette use (N=1074)

Harm perceptionsEver non-e-cigarette use* vs never use OR (95% CI)Ever e-cigarette use vs never use OR (95% CI)
Causes health problems
Strongly disagree2.39 (0.97–5.92)6.12 (3.16–11.90)
Disagree2.30 (1.11–4.75)4.78 (2.71–8.41)
Neither1.63 (0.95–2.80)2.72 (1.74–4.25)
Agree0.95 (0.52–1.72)1.86 (1.17–2.97)
Strongly agreeRef.Ref.
Is healthier than smoking traditional cigarettes
Strongly disagree0.99 (0.47–2.12)0.58 (0.32–1.05)
Disagree0.96 (0.45–2.05)0.86 (0.50–1.49)
Neither0.79 (0.44–1.41)0.54 (0.36–0.82)
Agree1.04 (0.59–1.83)0.74 (0.50–1.11)
Strongly agreeRef.Ref.
Causes breathing problems
Strongly disagree2.13 (0.82–5.50)4.40 (2.26–8.56)
Disagree1.94 (0.88–4.28)3.14 (1.76–5.60)
Neither1.21 (0.69–2.11)1.85 (1.22–2.81)
Agree1.24 (0.72–2.15)1.34 (0.87–2.06)
Strongly agreeRef.Ref.
Causes oral health problems
Strongly disagree3.34 (1.37–8.15)6.01 (3.16–11.44)
Disagree1.69 (0.75–3.82)3.15 (1.80–5.51)
Neither1.51 (0.87–2.62)2.00 (1.32–3.03)
Agree1.41 (0.79–2.52)1.1 (0.69–1.78)
Strongly agreeRef.Ref.
Is addictive
Strongly disagree2.26 (0.95–5.42)2.09 (1.07–4.08)
Disagree2.31 (1.07–5.02)2.31 (1.30–4.09)
Neither1.34 (0.75–2.37)1.67 (1.11–2.50)
Agree1.30 (0.75–2.25)1.14 (0.76–1.72)
Strongly agreeRef.Ref.
Is difficult to quit
Strongly disagree3.21 (1.39–7.41)3.61 (1.98–6.58)
Disagree3.77 (1.82–7.82)3.21 (1.86–5.54)
Neither1.52 (0.85–2.72)1.27 (0.83–1.93)
Agree1.61 (0.88–2.97)1.45 (0.94–2.25)
Strongly agreeRef.Ref.

* Ever non-e-cigarette users include cigarette and smokeless tobacco users who have not tried e-cigarettes. Models adjusted for gender, race/ethnicity, school type, state, and number of household tobacco users. Ref.: reference group.

DISCUSSION

Compared to never users, Appalachian youth e-cigarette users and non-e-cigarette users perceive e-cigarettes to be less harmful, less addictive and less difficult to quit, with harm-related associations stronger for e-cigarette users. E-cigarette users disagreed that problems with breathing and oral health were caused by e-cigarette use, despite recent evidence of a potential link between e-cigarette use and respiratory disease15. Associations were less clear for non-e-cigarette users, illustrating ways that perceptions vary by type of health issue and use pattern. Among e-cigarette users, associations appear to be stronger for health-related harms, and addiction-related harms were similar among non-e-cigarette and e-cigarette users.

Our findings corroborate previous research that reports associations between perceiving e-cigarettes as less harmful or less addictive and tobacco use among youth5,6. Specifically, among youth never cigarette users, perceiving e-cigarettes as less harmful was associated with e-cigarette use5. Similarly, among a sample of Florida youth, perceiving e-cigarettes as easy to quit was associated with more e-cigarette use including concurrent use of e-cigarettes and other tobacco products6. In our study, Appalachian youth views on whether e-cigarettes are harmful or addictive varied by tobacco use patterns. Although e-cigarette users indicated less concern about harms than did non-e-cigarette users, results for the latter group, who already use tobacco, suggest risk for e-cigarette uptake.

To combat e-cigarette use, health campaigns must inform youth of the associated harms and potential for addiction. Given evidence that perceiving e-cigarettes as less harmful predicts subsequent use among never users7, efforts must better explicate factors that predict youth susceptibility to trying e-cigarettes. Further, targeted messaging to e-cigarette users, who tend to perceive e-cigarettes as less dangerous, is needed to raise awareness and increase understanding, as these youth are especially vulnerable to continued use.

Limitations

Our study has limitations. First, responses were self-reported and thus are susceptible to associated biases. Second, the study is cross-sectional; thus, directionality cannot be determined. Third, the study was conducted prior to the COVID-19 pandemic, which may have altered harm perceptions related to tobacco use. Despite these limitations, our study is among the first to observe associations between perceiving e-cigarettes as less harmful or less addictive and e-cigarette use in a sample of Appalachian youth.

CONCLUSIONS

Compared to never users, Appalachian youth non-ecigarette and e-cigarette users perceive e-cigarettes as less harmful, with associations greater in magnitude for e-cigarette users. These users disagree that e-cigarettes are harmful, cause oral health problems, are addictive, and are difficult to quit. Our findings, which support previous work indicating that perceiving e-cigarettes as less harmful or less addictive is associated with e-cigarette use, provide insights on Appalachian youth tobacco views and use. Health communication messaging should address increasing youth understanding of the health dangers of these products, especially for vulnerable youth.