Advertisement

Cigarette‒E-cigarette Transitions and Respiratory Symptom Development

Open AccessPublished:December 02, 2022DOI:https://doi.org/10.1016/j.amepre.2022.10.006

      Introduction

      E-cigarette use is associated with pulmonary inflammation, functional respiratory changes, and chronic lung disease. Most population-level E-cigarette research has utilized point-in-time measures of E-cigarette exposures, which may not generalize to adults who transition between cigarettes and E-cigarettes.

      Methods

      Data obtained from the Population Assessment of Tobacco and Health study were collected from 2013 to 2019 and analyzed in 2022. Three observations were created per respondent, with exposure intervals assessed over Waves 1–2, 2–3, and 3–4. Each wave of the exposure interval was classified as nonuse, exclusive E-cigarette use, exclusive smoking, or dual use, producing 16 possible cigarette‒E-cigarette transitions. The association between transitions and both dry nighttime cough and wheeze symptom development during follow-up were assessed using mixed-effects Poisson models.

      Results

      Among 33,231 observations from 13,528 unique participants, transitioning from nonuse to exclusive E-cigarette use was associated with 1.62 times higher incidence rate of wheeze (incident rate ratio=1.62; 95% CI=1.12, 2.34) than persistent nonuse. There was no change in reported dry nighttime cough (incident rate ratio=0.84; 95% CI=0.52, 1.35) or wheeze (incident rate ratio=0.87; 95% CI=0.52, 1.46) in individuals who switched from cigarettes to E-cigarettes, whereas transitioning from dual use to E-cigarette use was associated with large reductions in both symptoms (incident rate ratio=0.58; 95% CI=0.39, 0.87 and incident rate ratio=0.36; 95% CI=0.20, 0.63, respectively).

      Conclusions

      E-cigarette initiation among nonusers is associated with increased respiratory morbidity. Further research should assess the risks and benefits of E-cigarette‒assisted cigarette cessation given the reduction in symptom development rates among dual use to E-cigarette switchers.

      INTRODUCTION

      E-cigarettes have become a leading means of nicotine delivery. In humans and animal models, acute E-cigarette use has been associated with pulmonary inflammation and functional changes.
      • Gotts JE
      • Jordt SE
      • McConnell R
      • Tarran R.
      What are the respiratory effects of e-cigarettes?.
      ,
      • Masso-Silva JA
      • Byun MK
      • Crotty Alexander LE
      Acute and chronic effects of vaping electronic devices on lung physiology and inflammation.
      Population-based studies have found that E-cigarette use is associated with respiratory symptoms and disease.
      • Gotts JE
      • Jordt SE
      • McConnell R
      • Tarran R.
      What are the respiratory effects of e-cigarettes?.
      ,
      • Xie W
      • Kathuria H
      • Galiatsatos P
      • et al.
      Association of electronic cigarette use with incident respiratory conditions among U.S. adults from 2013 to 2018.
      ,
      • Wills TA
      • Soneji SS
      • Choi K
      • Jaspers I
      • Tam EK.
      E-cigarette use and respiratory disorders: an integrative review of converging evidence from epidemiological and laboratory studies.
      However, previous research has relied on point-in-time measures of E-cigarette exposures, which may not accurately capture real-world E-cigarette use patterns because individuals frequently transition between tobacco products.
      • Coleman B
      • Rostron B
      • Johnson SE
      • et al.
      Transitions in electronic cigarette use among adults in the Population Assessment of Tobacco and Health (PATH) Study, Waves 1 and 2 (2013–2015).
      This study examined the association between cigarette‒E-cigarette transitions and respiratory symptoms to assess potential pulmonary risks and benefits of switching products.

      METHODS

      Public-use data were obtained from the Population Assessment of Tobacco and Health (PATH) Study, a nationally representative cohort study of 32,320 adults.

      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] Public-Use Files. Inter-university Consortium for Political and Social Research [distributor], 2022-10-07. https://www.icpsr.umich.edu/web/NAHDAP/studies/36498. Accessed November 30, 2022.

      Data collection occurred over 5 annual waves from 2013 to 2019. Groups oversampled in selection were those using tobacco, young adults, and African Americans. Respondents provided written informed consent, and the PATH Study protocol was approved by the Westat IRB.
      The study sample included PATH respondents who completed all the 5 waves of surveys. For each respondent, 3 observations were created, with respective exposure intervals assessed over PATH Waves 1–2, 2–3, and 3–4. In each wave of the exposure interval, observations were classified as nonuse, exclusive E-cigarette use, exclusive smoking, or dual use. Sixteen cigarette‒E-cigarette transitions were then designed on the basis of the possible combinations of product use over the exposure interval (Appendix Figure 1, available online). The baseline for each observation was defined as the second wave of the exposure interval. Follow-up began after baseline and continued through PATH Wave 5. Observations were excluded if they had a respiratory disease diagnosis, including chronic obstructive pulmonary disorder, chronic bronchitis, asthma, or emphysema (19,891), or had a past 12-month cough or wheeze symptom (13,759) at baseline. Other excluded observations were those who did not participate in all the 5 waves (29,507) or missed data for baseline respiratory disease (302), baseline respiratory symptoms (96), cigarette use (60), E-cigarette use (37), or outcome in the first wave of follow-up (77), leaving 33,231 observations (13,528 unique) in the analytical sample (Appendix Figure 2, available online). Eligibility for each observation was determined independently of the eligibility of the respondent's other observations.
      Respondents were considered currently smoking cigarettes for a given wave if they reported smoking at least some days and had smoked 100 or more lifetime cigarettes. They were characterized as using E-cigarettes if they reported using E-cigarettes for at least some days. Outcomes included past 12-month (1) dry nighttime cough not associated with a cold or chest infection or (2) wheeze or whistling in their chest. Each symptom was assessed at baseline for exclusion and again during each wave of follow-up. Respondents were considered to have developed a cough or wheeze if at any point over follow-up they reported experiencing the symptom in the past 12 months.
      The associations between transition group and both cough and wheeze symptom development were assessed using mixed-effects Poisson models adjusted for baseline covariates. Persistent nonuse (nonuse to nonuse transition), persistent E-cigarette use, persistent smoking, and persistent dual use were the reference groups, and each transition group was only compared with the reference with the same cigarette‒E-cigarette use in the first wave of the exposure interval (i.e., nonuse to E-cigarette transitions was only compared with the persistent nonuse reference group). Covariates were collected at baseline (exposure Wave 2) and included all the variables listed in Table 1, excluding exposure Wave 2 use pattern and including a pack-years squared term. An individual-level random effect was incorporated to account for within-person correlation. Person-years were totaled until the individual reached PATH Wave 5 or developed a respiratory symptom. PATH Study all-wave longitudinal weights were added to adjust for the complex sample design. Multiple imputation by chained equations (5 imputations) was utilized to handle missing covariate data. Results were presented as incident rate ratios (IRRs) along with 95% CIs. Incident rates were calculated for age- and sex-adjusted Poisson models.
      Table 1Descriptive Statistics of Observations
      CharacteristicsTotal (N=33,231)Nonuse (n=24,054)Exclusive E-cigarette use (n=1,051)Exclusive smoking (n=6,624)Dual use (n=1,502)
      Pack-years,
      Pack-years were calculated at Wave 1 on the basis of current and past smoking intensity and smoking duration. Pack-years were calculated for Waves 2–4 by taking the pack-years in the previous wave and adding either the current packs smoked per (daily smokers) or packs smoked per day smoking times day smoked in the past 30 days divided by 30 days (nondaily smokers). More detailed information on calculations can be found in the Appendix Methods (available online).
      mean (SD)
      5.2 (22.9)3.8 (22.5)10.3 (18.1)15.4 (25.8)12.6 (15.8)
      Female, (%)16,777 (51)12,478 (52)489 (42)3,097 (41)713 (42)
      Age, years (%)
       18–246,815 (10)5,330 (9)301 (21)863 (9)321 (16)
       25–348,248 (19)5,707 (18)287 (29)1,796 (27)458 (32)
       35–445,492 (18)3,681 (17)189 (18)1,321 (21)301 (22)
       45–545,035 (18)3,452 (18)149 (16)1,192 (18)242 (15)
       55–644,317 (17)3,069 (18)90 (9)1,009 (16)149 (11)
       ≥653,320 (18)2,811 (20)35 (6)443 (8)31 (2)
      Race/ethnicity, (%)
       Non-Hispanic White19,299 (65)13,727 (65)639 (66)3,892 (64)1,041 (74)
       Non-Hispanic Black4,772 (10)3,329 (10)111 (10)1,189 (16)143 (9)
       Non-Hispanic Other2,357 (8)1,759 (8)77 (8)401 (5)120 (8)
       Hispanic6,394 (17)4,945 (17)215 (16)1,044 (14)190 (10)
      Education, (%)
       Less than high school3,229 (9)1,893 (9)97 (8)1,070 (15)169 (11)
       High school/GED8,285 (25)5,103 (23)270 (28)2,430 (38)482 (31)
       Some college11,747 (31)8,359 (31)506 (46)2,234 (32)648 (42)
       Bachelor or above9,970 (34)8,699 (37)178 (18)890 (14)203 (15)
      Household annual income ≥$50,000, (%)14,060 (52)11,396 (55)403 (43)1,818 (32)443 (35)
      Ever used, (%)
       Cigars14,242 (33)8,917 (29)627 (59)3,690 (57)1,008 (67)
       Cigarillos13,945 (28)8,268 (23)663 (63)3,941 (60)1,073 (72)
       Pipes6,785 (17)4,010 (14)340 (34)1,863 (30)572 (39)
       Hookah11,736 (18)8,121 (14)575 (50)2,259 (33)781 (50)
       Snus2,933 (5)1,498 (3)175 (18)884 (14)376 (25)
       Other smokeless tobacco6,357 (15)3,883 (13)252 (26)1,695 (27)527 (36)
       Marijuana16,329 (37)10,278 (33)715 (68)4,244 (65)1,092 (74)
      Current noncigarette tobacco use,
      At least some days use of any of cigars, cigarillos, pipes, hookah, snus, or other smokeless tobacco.
      (%)
      4,336 (7)2,786 (5)253 (23)937 (14)360 (24)
      Cardiac-linked condition,
      Includes at least 1 of high cholesterol, high blood pressure, diabetes, stroke, heart failure, heart attack, and other heart conditions.
      (%)
      12,632 (49)9,168 (49)354 (41)2,588 (45)522 (39)
      BMI, mean (SD)27.8 (6.2)27.9 (6.2)28.3 (6.7)27.5 (6.1)27.4 (6.2)
      Exposure Wave 2 use pattern
       Nonuse22,853 (98)393 (35)744 (11)113 (7)
       Exclusive E-cigarette use351 (1)472 (47)99 (2)111 (7)
       Exclusive smoking755 (1)83 (7)5,255 (80)718 (49)
       Dual use95 (0.2)103 (10)526 (8)560 (37)
      Note: Unweighted n (weighted %) and unweighted n (SD) are presented as noted. Observations include those with exposure intervals over Waves 1–2, 2–3, and 3–4. Descriptive statistics were collected at baseline, which is the second wave of the given exposure interval.
      a Pack-years were calculated at Wave 1 on the basis of current and past smoking intensity and smoking duration. Pack-years were calculated for Waves 2–4 by taking the pack-years in the previous wave and adding either the current packs smoked per (daily smokers) or packs smoked per day smoking times day smoked in the past 30 days divided by 30 days (nondaily smokers). More detailed information on calculations can be found in the Appendix Methods (available online).
      b At least some days use of any of cigars, cigarillos, pipes, hookah, snus, or other smokeless tobacco.
      c Includes at least 1 of high cholesterol, high blood pressure, diabetes, stroke, heart failure, heart attack, and other heart conditions.
      Statistical analyses were conducted in Stata, Version 16.1 (StataCorp, College Station, TX). For significance, a 2-sided threshold of p≤0.05 was used.

      RESULTS

      Among 33,231 observations (51% female, 65% non-Hispanic White), there were 6,208 (19%) cases of cough development and 3,258 (10%) cases of wheeze development (Table 1 and Figure 1 and Appendix Figure 3, available online). The cough development analysis totaled 61,789 years of follow-up (mean=1.86 years), whereas the wheeze analysis totaled 64,359 years of follow-up (mean=1.94 years). Transitioning from nonuse to exclusive E-cigarette use was associated with a 1.62 times higher incidence rate of wheeze development (IRR=1.62; 95% CI=1.12, 2.34) but no change in cough development compared with persistent nonuse. Switching from E-cigarettes to cigarettes was associated with higher rates of both cough (IRR=1.84; 95% CI=1.19, 2.83) and wheeze (IRR=1.96; 95% CI=1.14, 3.38) development than persistent E-cigarette use, whereas switching from cigarettes to E-cigarettes was not associated with a significant change in symptom development compared with persistent smoking (IRR=0.84; 95% CI=0.52, 1.35 and IRR=0.87; 95% CI=0.52, 1.46, respectively). Cigarette to dual-use transitions was also not associated with a change in cough or wheeze development rates; however, transitions from dual use to E-cigarette use were associated with reductions in the rates of both symptoms (IRR=0.58; 95% CI=0.39, 0.87 and IRR=0.36; 95% CI=0.20, 0.63, respectively).
      Figure 1
      Figure 1Association between cigarette‒E-cigarette transitions and respiratory symptom development.
      aDevelopment of a dry nighttime cough not associated with a cold or chest infection.
      bDevelopment of a wheeze or whistling in the chest.
      cTransition groups are defined by cigarette and E-cigarette use status over the 2-wave exposure interval.
      dPer 1,000 person-years, adjusted for age and sex and derived from Poisson models that did not incorporate a person-level random effect.
      eAdjusted for age (years), sex (male or female), and race/ethnicity (non-Hispanic White, non-Hispanic Black, Hispanic, non-Hispanic other); education (less than high school, high school/GED, some college, bachelor or above); household annual income ≥$50,000 (yes or no); cigarette smoking pack-years and its square (both continuous); ever smoked cigars (yes or no); ever smoked cigarillos (yes or no); ever smoked pipes (yes or no); ever smoked hookah (yes or no); ever used snus (yes or no); ever used other smokeless tobacco (yes or no); past 30-day noncigarette tobacco use (yes or no); ever used marijuana (yes or no); ever diagnosed with a cardiovascular-linked condition (yes or no); and BMI (kg/m2).
      IRR, incident rate ratio.

      DISCUSSION

      This study observed that both E-cigarette initiation among nonusers and switching to smoking among E-cigarette users were associated with increased rates of respiratory symptom development. These findings are concerning given that these transitions are becoming increasingly frequent and that those who initiate E-cigarette use are subsequently more likely to smoke cigarettes.
      • McMillen R
      • Klein JD
      • Wilson K
      • Winickoff JP
      • Tanski S.
      E-cigarette use and future cigarette initiation among never smokers and relapse among former smokers in the PATH study.
      ,
      • Brouwer AF
      • Jeon J
      • Hirschtick JL
      • et al.
      Transitions between cigarette, ENDS and dual use in adults in the PATH study (waves 1–4): multistate transition modelling accounting for complex survey design.
      Although switching from cigarettes to E-cigarettes was not observed to be associated with respiratory symptom development, dual use to E-cigarette transitions were associated with reduced rates of cough and wheeze development. Given that relatively few individuals switch from cigarettes to E-cigarettes and that it is far more common for people who smoke to first transition to dual use before exclusively using E-cigarettes, further research is needed to determine whether E-cigarettes could play a role in reducing morbidity among those unable to quit cigarettes using conventional methods.
      • Brouwer AF
      • Jeon J
      • Hirschtick JL
      • et al.
      Transitions between cigarette, ENDS and dual use in adults in the PATH study (waves 1–4): multistate transition modelling accounting for complex survey design.
      However, this work found no evidence that E-cigarette initiation without complete cigarette cessation was associated with changes in respiratory symptom development.
      Strengths of this study include the use of multiple observations per respondent to increase statistical efficiency, nationally representative estimates that increase generalizability, and longitudinal assessments of cigarette and E-cigarette use exposures. In addition, comparison of transition groups only with the same initial use patterns helped to ensure that the comparisons were balanced on baseline characteristics, thus reducing the potential for residual confounding.

      Limitations

      Limitations were the use of self-reported measures that may result in misclassification, the potential for survivorship bias because transitions could occur up to 1 year before follow-up began, and the recent evolution in E-cigarette devices that make it unclear whether these results are generalizable to newer E-cigarette brands.

      CONCLUSIONS

      The present findings suggest that E-cigarette initiation among nonusers and subsequent cigarette smoking may translate to a significant avoidable burden of respiratory morbidity, highlighting the urgency for robust E-cigarette regulations among the nonsmoking population. Transitioning from dual use to exclusive E-cigarette use may reduce respiratory symptom development, suggesting that further research is needed to assess the potential risks and benefits of E-cigarette‒assisted cigarette cessation.

      CRediT AUTHOR STATEMENT

      Jonathan B. Berlowitz: Conceptualization, Formal analysis, Funding acquisition, Methodology, Software, Visualization, Writing – original draft, Writing – review & editing. Wubin Xie: Methodology, Formal analysis, Software, Visualization, Writing – original draft, Writing – review & editing. Alyssa F. Harlow: Methodology, Writing – original draft, Writing – review & editing. Michael J. Blaha: Supervision, Writing – review & editing. Aruni Bhatnagar: Funding acquisition, Supervision, Writing – review & editing. Emelia J. Benjamin: Funding acquisition, Supervision, Writing – review & editing. Andrew C. Stokes: Conceptualization, Funding acquisition, Methodology, Project administration, Writing – original draft, Writing – review & editing.

      ACKNOWLEDGMENTS

      The content of this study is solely the responsibility of the authors and does not necessarily represent the official views of the sponsors.
      Research reported in this publication was supported by the National Heart, Lung, and Blood Institute of the NIH and the U.S. Food and Drug Administration's Center for Tobacco Products (Grants P50HL120163, U54HL120163, T35HL139444, and 1K01HL154130-01). This work was also supported by a Public Policy Award from the American Lung Association.
      Parts of this work were presented at the American Thoracic Society 2022 International Congress in San Francisco on May 16, 2022.
      No financial disclosures were reported by the authors of this paper.

      Appendix. SUPPLEMENTAL MATERIAL

      REFERENCES

        • Gotts JE
        • Jordt SE
        • McConnell R
        • Tarran R.
        What are the respiratory effects of e-cigarettes?.
        BMJ. 2019; 366: l5275https://doi.org/10.1136/bmj.l5275
        • Masso-Silva JA
        • Byun MK
        • Crotty Alexander LE
        Acute and chronic effects of vaping electronic devices on lung physiology and inflammation.
        Curr Opin Physiol. 2021; 22100447https://doi.org/10.1016/j.cophys.2021.06.001
        • Xie W
        • Kathuria H
        • Galiatsatos P
        • et al.
        Association of electronic cigarette use with incident respiratory conditions among U.S. adults from 2013 to 2018.
        JAMA Netw Open. 2020; 3e2020816https://doi.org/10.1001/jamanetworkopen.2020.20816
        • Wills TA
        • Soneji SS
        • Choi K
        • Jaspers I
        • Tam EK.
        E-cigarette use and respiratory disorders: an integrative review of converging evidence from epidemiological and laboratory studies.
        Eur Respir J. 2021; 571901815https://doi.org/10.1183/13993003.01815-2019
        • Coleman B
        • Rostron B
        • Johnson SE
        • et al.
        Transitions in electronic cigarette use among adults in the Population Assessment of Tobacco and Health (PATH) Study, Waves 1 and 2 (2013–2015).
        Tob Control. 2019; 28: 50-59https://doi.org/10.1136/tobaccocontrol-2017-054174
      1. 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] Public-Use Files. Inter-university Consortium for Political and Social Research [distributor], 2022-10-07. https://www.icpsr.umich.edu/web/NAHDAP/studies/36498. Accessed November 30, 2022.

        • McMillen R
        • Klein JD
        • Wilson K
        • Winickoff JP
        • Tanski S.
        E-cigarette use and future cigarette initiation among never smokers and relapse among former smokers in the PATH study.
        Public Health Rep. 2019; 134: 528-536https://doi.org/10.1177/0033354919864369
        • Brouwer AF
        • Jeon J
        • Hirschtick JL
        • et al.
        Transitions between cigarette, ENDS and dual use in adults in the PATH study (waves 1–4): multistate transition modelling accounting for complex survey design.
        Tob Control. 2020; 31: 424-431https://doi.org/10.1136/tobaccocontrol-2020-055967