Abstract
The harm caused by cigarette smoking is overwhelmingly due to byproducts of tobacco combustion. Electronic Nicotine Delivery Systems (ENDS) provide nicotine to users without combustion, and may support tobacco harm reduction among cigarette smokers who would not otherwise quit in the near term. Analyses of Wave 5 of the Population Assessment of Tobacco and Health (PATH) Study compared biomarkers of exposure (BOE) levels for nicotine, 3 metals, 2 tobacco-specific nitrosamines and 14 smoking-related volatile organic compounds in 151 exclusive ENDS users, 1341 exclusive cigarette smokers, 115 dual users (cigarettes and ENDS), and 1846 past 30-day nonusers of tobacco, adjusting for demographics. Nicotine exposure in ENDS users and dual users did not significantly differ from smokers. Among ENDS users, 16 of 18 other BOEs were significantly lower than smokers’; 9 BOEs were not significantly different from nonusers. Among dual users smoking < 10 cigarettes/day, 15 of 18 non-nicotine BOEs were significantly lower than smokers’, whereas in dual users smoking ≥ 10 cigarettes per day none of the BOEs significantly differed from smokers’. In this representative sample of US adults, exclusive use of ENDS (vs. cigarette smoking) was associated with much lower exposures to many harmful chemicals associated with smoking-related disease. BOE levels in dual users were directly related to their cigarette consumption. These BOE data provide further evidence that ENDS expose users to substantially lower levels of toxicants than combustible cigarettes, confirming their potential for harm reduction.
Similar content being viewed by others
Introduction
Cigarette smoking is responsible for more than seven million deaths annually worldwide1,2,3. The harm caused by cigarette smoking is primarily due to exposure to byproducts of tobacco combustion4: cigarette smoke contains more than 7000 harmful chemicals including nearly 70 carcinogens. The US Food and Drug Administration has identified harmful and potentially harmful constituents (HPHCs) in cigarette smoke and in other tobacco products that are known to cause cancer and cardiovascular and respiratory diseases5,6. The risk and severity of many smoking-induced diseases are directly related to the level of exposure to the HPHCs in cigarette smoke5; hence reductions in exposure to HPHCs are expected to reduce risk for these diseases and many other adverse health effects7,8.
Many HPHCs are not directly measurable in the bodies of smokers; however, exposures can be assessed and quantified by measuring biomarkers of exposure (BOEs)—metabolites of the HPHC that are detectable in the urine and bloodstream7,8. BOEs provide a measure of the actual human HPHC absorption associated with tobacco use, and thus can help quantify the potential health risks of tobacco products7,8.
In contrast to cigarettes, electronic nicotine delivery systems (ENDS) deliver nicotine without combusting tobacco9, and thus do not expose users to combustion byproducts such as carbon monoxide (CO) and polycyclic aromatic hydrocarbons9,10,11. Clinical trials demonstrate that smokers who switch to exclusive ENDS use experience significant reductions in tobacco specific nitrosamines (TSNAs), such as NNAL—a biomarker for NNK, and known carcinogen12.
However, there is a need to establish whether, like cigarettes, ENDS expose users to volatile organic compounds (VOCs)—HPHCs that exhibit dose–response relationships with cancer and cardiovascular disease risk13. For example, acrolein (whose exposure can be assessed using 3-hydroxypropyl-mercapturic acid [3-HPMA]) is a respiratory irritant that is also associated with increased risk of cardiovascular disease14, and 1,3-butadiene (whose exposure can be assessed using monohydroxylbutenyl-mercapturic acid [MHBMA]) is a known human carcinogen and is also associated with respiratory and reproductive toxicity15,16,17. Additionally, some studies suggest metals can leech from heating coils into ENDS aerosols18,19, potentially causing adverse health effects given their high level of toxicity and carcinogenic effects20,21.
Several recent analyses have utilized BOE data from the Population Assessment of Tobacco and Health (PATH) Study, a longitudinal cohort study of tobacco use among a nationally representative sample of US adults, to assess variation in HPHC exposure. Cross-sectional analyses of PATH Wave 1 (2013–2014) demonstrated that exclusive ENDS users have lower levels of many of the measured BOEs compared to cigarette smokers and dual users22,23, and that among dual users, it is cigarette smoking, and not ENDS use, that is the primary driver of HPHC exposures24,25,26. Consistent with these findings, two longitudinal studies using data from PATH Wave 1 (2013–2014) to Wave 2 (2014–2015) found that smokers who switch from cigarette smoking to exclusive ENDS use experienced substantial decreases in BOEs, and smokers who transition to dual use and reduce their cigarette consumption by at least 50% also experience significant reductions in BOEs27,28.
Although these studies offer valuable insights into HPHC exposures among ENDS users, the findings are limited to early-generation ENDS products marketed when the data was collected between 2013–2015. ENDS products have rapidly evolved since 2015, with open system modular devices and fourth-generation pod-based nicotine-salt ENDS products becoming increasingly prevalent29,30,31. These later-generation products deliver nicotine more efficiently than earlier-generation products32,33, and some implement better control of the temperature to which the e-liquid is heated, a major factor in production of toxic compounds34; hence there is a need to assess differences in BOEs among ENDS users in more recently-collected data.
The primary aim of the current study was to assess differences in levels of BOEs to TSNAs, VOCs, metals and nicotine, comparing: (i) exclusive ENDS users; (ii) exclusive smokers; (iii) dual users; and (iv) past 30-day tobacco non-users using data from Wave 5 (2018–2019) of the PATH Study. Further, given heterogeneity in cigarette smoking behaviors among dual users26,35, exposures among dual users were also evaluated by their level of cigarette consumption.
Materials and methods
Study design
The PATH Study is a nationally-representative, longitudinal cohort study of adults and youth in the US36. Recruitment used a stratified address-based, area-probability sampling design; survey weights are used to produce national estimates36. The study was conducted by Westat under a contract with the Food and Drug Administration (FDA) Center for Tobacco Products (CTP), National Institute on Drug Abuse (NIDA) and National Institutes of Health (NIH). Westat’s Institutional Review Board (IRB) approved the study design and data collection protocol36. All participants in the PATH Study provided informed consent36. This secondary analysis of deidentified PATH Study data was reviewed by Advarra IRB Number IRB00000971 and deemed exempt on the basis of 45 CFR 46.104(d)(4). Use of these data was approved by the Inter-university Consortium for Political and Social Research (ICPSR), which reviewed analytic results and cleared these results for public dissemination on the basis that they did not present a risk of identifying individual participants. All procedures were performed in accordance with relevant guidelines.
The current analyses used data from Wave 5 (December 2018 to November 2019) of the PATH Study Biomarker Restricted-Use Files (BRUF)37 and Restricted-Use Files (RUF)38. Urine specimens were analyzed by the Centers for Disease Control and Prevention (CDC), National Center of Environmental Health. Because they contain more granular data which may pose a risk to respondent identification, the PATH Study BRUF and PATH Study RUF are only available to researchers that apply for access and complete a data use agreement, which includes review of statistical outputs before any results can be published. At this time, PATH Study biomarker data are available only in restricted-use files. Additional information is provided by36 and at https://doi.org/10.3886/ICPSR36840.v18 and https://doi.org/10.3886/ICPSR36231.v31.
Participants
The analytic sample for the current manuscript is comprised of PATH Study Wave 5 adult respondents who were at least 21 years old at time of interview, provided urine samples and also had (i) valid data on sex and race/ethnicity; (ii) body mass index (BMI) between 18–40; (iii) urinary creatinine in the range 10–370 mg/dL; and (iv) valid BOE data on one or more analyte (including imputed data below the limit of detection [LOD]). To avoid confounding of BOEs by other exposures, participants were excluded if they reported past 30-day use of other tobacco, nicotine, or marijuana products (traditional or filtered cigars, cigarillos, pipe tobacco, hookah, smokeless tobacco, snus, nicotine replacement products, marijuana and hashish) at interview or past 3-day use at urine collection (Table S1). Further, respondents were excluded if they reported past 3-day product use behaviors at the time of urine collection inconsistent with their prior interview reports (e.g., reported no past-30-day smoking at interview but reported smoking in the past 3 days at urine collection). Smokers were excluded if their reported average daily cigarette consumption (i.e., cigarettes per day) was missing, zero, or greater than 100 (an unrealistically high number).
Tobacco product use groups
Participants were initially classified into one of four groups on the basis of past 30-day ENDS use and past 30-day cigarette smoking: (i) exclusive ENDS users (“ENDS Users”; past-30-day ENDS use and no past-30-day smoking); (ii) exclusive cigarette smokers (“Smokers”; past-30-day cigarette smoking, no past-30-day ENDS use); (iii) dual users (“Dual Users”; cigarette smoking and ENDS use in the past 30 days); and (iv) past 30-day tobacco product nonusers (“Nonusers”; neither smoking cigarettes nor using ENDS in the past 30 days). Preliminary analyses showed BOE levels were more heterogeneous in Dual Users than in other tobacco use groups (Figs. S1, S2, S3, and S4). Therefore, for the primary analyses reported here Dual Users were partitioned into two subgroups by the median daily cigarette consumption: those smoking < 10 cigarettes/day vs. those smoking ≥ 10 cigarettes/day (Table S2).
Measures
BOE outcomes
In Version 18 of the PATH Study Biomarker Restricted-Use Files, available Wave 5 urine panels include BOE data for nicotine, metals, TSNAs, and VOCs. All 14 available VOC markers, two available TSNAs, and three available metals, as well as two aggregate measures of nicotine and its metabolites (total nicotine equivalents-2 [TNE2] and total nicotine equivalents-6 [TNE6]) were analyzed (Table 1). Table 1 displays the BOEs, noting their chemical name, the name of the parent compound, and the disease-related toxicities with which they are associated6. Altogether, 15 BOEs were considered to be relevant to cancer, 9 to respiratory disease, 8 to reproductive or developmental problems, and 4 to cardiovascular disease, with most BOEs relevant to multiple disease classes.
Data analysis
Urinary biomarker measurements were adjusted for urinary creatinine by expressing the concentration of the BOE as a ratio to the concentration of creatinine. This is an often-used adjustment for dilution of the spot urine samples by water27. Measurements below the LOD were imputed with the value LOD/\(\sqrt {2}\) and then adjusted for urinary creatinine (https://doi.org/10.3886/ICPSR36840.userguide). (Thus, levels are never considered to be zero.) Weighted geometric means and 95% confidence intervals were computed for each creatinine-adjusted BOE by tobacco use group (Table S4). Pairwise group differences in log-transformed creatinine-adjusted BOEs, adjusted for age, sex, race/ethnicity, and BMI, were tested in a weighted regression analysis. Adjusted geometric means were computed by exponentiating predicted population margins with covariate values fixed at the observed means. Contrast t tests assessed the null hypothesis that the adjusted geometric mean ratio (“GMR”) was equal to one.
Adjusted analyses of geometric mean creatinine-adjusted BOE levels with two Dual User strata are presented in Figs. 1, 2, 3, and 4. Unadjusted and adjusted analyses, and analyses with one Dual User stratum are presented in the Supplemental Materials.
Nicotine equivalents among smokers, ENDS users, dual users stratified by cigarettes/day and tobacco nonusers (weighted adjusted geometric mean and 95% confidence interval). Note. Groups whose bars do not share a letter above the bar are significantly different (p < 0.05). Groups whose bars share a letter do not significantly differ from each other (p ≥ 0.05). Geometric mean ratios, and more exact p-values are shown in Table S6 in the Supplemental Materials. Adjusted geometric means and confidence interval endpoints were derived from a weighted regression analysis with covariates for age, sex, race/ethnicity, and BMI and were computed by exponentiating predicted population margins with covariate values fixed at the observed means. The analysis was weighted to represent the US adult civilian, noninstitutionalized population of never, current, and recent (within 1-year) former tobacco users. Current Smokers: TNE2, N = 1341; TNE6, N = 1327. 10 + CPD Dual Users: TNE2, N = 61; TNE6, N = 61. < 10 CPD Dual Users: TNE2, N = 54; TNE6, N = 53. Current ENDS Users: TNE2, N = 151; TNE6, N = 146. Past 30-day Nonusers: TNE2, N = 1842; TNE6, N = 152.
BOEs of tobacco-specific nitrosamines among smokers, ENDS users, dual users stratified by cigarettes/day and past 30-day nonusers (weighted adjusted geometric mean and 95% confidence interval). Note. Groups whose bars do not share a letter above the bar are significantly different (p < 0.05). Groups whose bars share a letter do not significantly differ from each other (p ≥ 0.05). Geometric mean ratios, and more exact p-values are shown in Table S6 in the Supplemental Materials. Adjusted geometric means and confidence interval endpoints were derived from a weighted regression analysis with covariates for age, sex, race/ethnicity, and BMI and were computed by exponentiating predicted population margins with covariate values fixed at the observed margins. The analysis was weighted to represent the US adult civilian, noninstitutionalized population of never, current, and recent (within 1-year) former tobacco users. Current Smokers: NNAL, N = 1338; NNN, N = 1309. 10 + CPD Dual Users: NNAL, N = 60; NNN, N = 58. < 10 CPD Dual Users: NNAL, N = 54; NNN, N = 53. Current ENDS Users: NNAL, N = 149; NNN, N = 148. Past 30-day Nonusers: NNAL, N = 1828; NNN, N = 1836.
BOEs of VOCs among smokers, ENDS users, dual users stratified by CPD and past 30-day nonusers (weighted adjusted geometric mean and 95% confidence interval). Note. All values represent μg/g creatinine. Groups whose bars do not share a letter above the bar are significantly different (p < 0.05). Groups whose bars share a letter do not significantly differ from each other (p ≥ 0.05). Geometric mean ratios, and more exact p-values are shown in Table S6 in the Supplemental Materials. Adjusted geometric means and confidence interval endpoints were derived from a weighted regression analysis with covariates for age, sex, race/ethnicity, and BMI and were computed by exponentiating predicted population margins with covariate values fixed at the observed means. The analysis was weighted to represent the US adult civilian, noninstitutionalized population of never, current, and recent (within 1-year) former tobacco users. Current Smokers, N = 1341; 10 + CPD Dual Users, N = 61; < 10 CPD Dual Users, N = 54; ENDS Users, N = 151; Past 30-day Nonusers, N = 1846.
BOEs of metals among smokers, ENDS users, dual users stratified by cigarettes/day and past 30-day nonusers (weighted adjusted geometric mean and 95% confidence interval). Note. Groups whose bars do not share a letter above the bar are significantly different (p < 0.05). Groups whose bars share a letter do not significantly differ from each other (p ≥ 0.05). Geometric mean ratios, and more exact p-values are shown in Table S6 in the Supplemental Materials. Adjusted geometric means and confidence interval endpoints were derived from a weighted regression analysis with covariates for age, sex, race/ethnicity, and BMI and were computed by exponentiating predicted population margins with covariate values fixed at the observed means. The analysis was weighted to represent the US adult civilian, noninstitutionalized population of never, current, and recent (within 1-year) former tobacco users. Current Smokers, N = 1341; 10 + CPD Dual Users, N = 61; < 10 CPD Dual Users, N = 54; ENDS Users, N = 151; Past 30-day Nonusers, N = 1845.
All analyses were conducted using SAS© software (version 9.4, SAS Institute, Cary, NC, USA) and were weighted to represent the 2013–2014 U.S. adult civilian, noninstitutionalized population of never, current, and recent former (within 1-year) tobacco users with the Wave 5 single-wave weights for the Wave 1 Biomarker Core. Hypothesis tests used the 0.05 significance level without adjustment for multiple comparisons. Variance estimation used Fay’s Balanced Repeated Replication method with Fay’s factor set to 0.3 and 100 replicate weights to account for the complex survey design structure of the PATH Study. Estimates are reported for which the relative standard error (RSE) is greater than 30% or for which the proportion of subjects with measurements below the LOD exceeded 40%.
Results
Sociodemographic characteristics
On average, Smokers and Nonusers were older than Dual Users and ENDS Users (Table 2). Nearly three-quarters of Dual Users were female. Approximately 90% of ENDS Users had a history of established smoking (i.e., smoked more than 100 cigarettes). Dual Users smoked significantly fewer cigarettes/day (mean = 10.3 [SE = 0.93]) than Smokers (mean = 13.8 [SE = 0.39]). When stratified by median cigarettes/day, Dual Users formed two distinct groups: those who smoked less than 10 cigarettes per day (N = 54; “< 10 CPD Dual Users”) averaged 3.5 cigarettes/day (SE = 0.5) and those who smoked at least 10 cigarettes per day (N = 61; “10 + CPD Dual Users”) averaged 14.5 cigarettes/day (SE = 1.5). Past 30-day smoking and ENDS use behaviors differed in the two Dual User strata; the average 10 + CPD Dual User smoked nearly every day (29.5 of the past 30 days) and used ENDS on about half as many days (18.2 of the past 30 days), whereas < 10 CPD Dual Users used ENDS (24.0 of the past 30 days) more frequently than cigarettes (20.6 days of the past 30 days).
About three-quarters of ENDS Users used open systems and used ENDS on approximately 28 of the last 30 days, on average (Table 2). Over half of Dual Users and ENDS Users reported using non-Tobacco/Menthol/Mint flavors.
Nicotine equivalents
Geometric mean levels of nicotine equivalents were significantly higher in in all nicotine-using groups than in Nonusers, and did not significantly differ between Smokers, ENDS Users and both Dual Users groups (Fig. 1). Levels of nicotine exposure in 10 + CPD Dual Users were more variable than levels in other nicotine-using groups.
TSNAs
Geometric mean levels of the two available TSNA BOEs—NNAL and NNN—were significantly lower in ENDS Users than in Smokers (Fig. 2). Levels of NNAL were significantly higher in ENDS Users than in Nonusers, but levels of NNN did not significantly differ between ENDS Users and Nonusers. Levels of NNAL and NNN in < 10 CPD Dual Users were significantly lower than levels in Smokers and significantly greater than levels in ENDS Users. Dual Users who smoked 10 or more cigarettes/day did not significantly differ from Smokers. The magnitude of differences in geometric mean NNAL levels are substantial: NNAL levels in all smoking groups were at least 22 times greater than levels in ENDS Users, with levels in Smokers nearly 86 times greater than levels in ENDS Users (Tables S5 and S6). In contrast, NNAL levels in ENDS Users were 3.5 times greater than levels in Nonusers.
BOEs of VOCs
Geometric mean levels of PMA, a marker for benzene exposure, did not significantly differ in any of the study groups (Fig. 3); therefore PMA was not considered further, leaving 13 VOC BOEs for subsequent comparisons.
Levels of all 13 VOC BOEs were significantly lower in ENDS Users than in Smokers, and for 8 of 13 VOC BOEs, levels in ENDS Users did not significantly differ from Nonusers (Fig. 3). VOC BOE levels in < 10 CPD Dual Users were intermediate between Smokers’ and ENDS Users’; for 12 of 13 VOC BOEs, levels in < 10 CPD Dual Users were significantly lower than in Smokers, but significantly greater than in ENDS users (Fig. 3). VOC BOE levels in 10 + CPD Dual Users did not significantly differ from levels in Smokers (Fig. 3).
Metals
Geometric mean levels for all three metals were significantly higher among Smokers and ENDS Users than Nonusers (Fig. 4). Levels of one metal—cadmium—were significantly lower in ENDS Users and Dual Users who smoked < 10 cigarettes/day than in Smokers (Fig. 4). In contrast, levels in Dual Users who smoked ≥ 10 cigarettes/day did not significantly differ from levels in Smokers.
Discussion
In this nationally-representative observational study of US adults, exclusive users of ENDS showed equivalent levels of nicotine, but substantially lower levels of TSNAs, VOCs, and one metal compared to cigarette smokers. Exclusive ENDS users did not significantly differ from past 30-day tobacco nonusers in inferred exposure to many TSNAs and VOCs, and in many cases the observed BOE values were very similar in magnitude. Overall, the BOE data demonstrate that exclusive ENDS users were able to obtain similar levels of nicotine as cigarette smokers but with significantly lower levels of exposure to numerous toxicants, including carcinogens.
The results of this analysis concord with previous observational studies using data from the PATH Study24,25,26 and other studies39 that were conducted when earlier-generation ENDS products were predominant, and extend these findings to the more recent tobacco product marketplace, specifically recent-generation ENDS products. The levels of BOEs observed among ENDS users relative to cigarette smokers are also consistent with controlled confinement studies, randomized trials and longitudinal observational studies in which smokers who switch to ENDS experience substantial reductions in BOEs10,40,41, and suggest that smokers who switch to exclusive ENDS use can reduce exposure to many HPHCs for which biomarker data are available to a similar extent as abstinence from tobacco products or smoking cessation. Although the dose–response relationship between HPHC exposures and subsequent disease are complex (e.g., subject to threshold effects), lower exposures are likely to be associated with lower disease risk, implying that smokers who switch to ENDS would likely experience decreased disease risk5,6.
Dual Users were extremely heterogeneous in exposure, and partitioning by cigarette consumption showed that the key driver of toxic exposure was the quantity of cigarette consumption. Some analyses have reported that dual users of both ENDS and cigarettes have exposures as high or even higher than those seen in exclusive smokers42,43,44. In the present analysis, levels of both TSNA and 12 of 13 VOC BOEs were significantly lower in dual users reporting smoking < 10 cigarettes/day than in exclusive cigarette smokers, though levels of all BOEs assessed were significantly greater in < 10 CPD Dual Users than in Nonusers.
Conversely, Dual Users smoking 10 or more cigarettes/day—whose average cigarette consumption was similar to that seen in Smokers (mean CPD = 14.5 vs. 13.8), showed levels of BOEs that were not significantly lower than Smokers’ levels but higher than ENDS Users and Nonusers. Thus, the results indicate that the BOE levels observed in Dual Users are largely attributable to Dual Users’ cigarette smoking. This dose–exposure association is consistent with previous analyses of the PATH Study showing that cigarette smoking is the primary driver of HPHC exposure among dual users24,25, with HPHC exposure increasing with increasing cigarette consumption27,45.
The findings regarding dual use are also consistent with controlled clinical confinement studies and randomized trials demonstrating that dual users who reduce their daily cigarette consumption by at least 50% significantly reduce their HPHC exposure compared to smokers who continue smoking as usual40,46,47,48,49,50,51,52,53. These data regarding HPHC exposure among dual users are especially pertinent given that dual use is a common initial use pattern among smokers who adopt ENDS, although it often serves as a transitionary state on the pathway to complete switching away from smoking54, ultimately reducing exposures further.
These findings regarding actual human exposure are consistent with chemical analyses of aerosols produced by ENDS under conditions that mimic real-world use, which show that these TSNAs and VOCs are absent or present at levels much lower than in cigarette smoke9. It is logical that the chemical composition of the ENDS aerosol would be predictive of the users’ ultimate exposure, as users only inhale chemicals present in the aerosol matrix. This suggests that chemical analyses of the aerosols can be used to infer human exposures, which is particularly important since some constituents that can be measured in the aerosols do not have validated biomarkers.
Chemical analyses of ENDS aerosols have found levels of metals above established toxicological standards18,19,55,56. However, many of these experiments have been criticized for using procedures that do not replicate real world patterns of ENDS use and other methodological limitations (e.g., over-powering or overheating the coil)57. The results of the current study demonstrate that, on average, ENDS Users are exposed to lower levels of cadmium than Smokers. Future naturalistic research is needed to assess levels of other metals beyond the three included in the PATH Study. Exposure to many VOCs can occur from sources other than tobacco use, and thus nonusers represent the benchmark for baseline levels of environmental exposures to such VOCs8. For example, acrylamide exposure can occur from many foods as well as from tobacco use58,59, and the levels observed in ENDS users were almost exactly equivalent to those in nonusers of tobacco. PMA, a marker for benzene exposure, was not elevated in Smokers (vs. Nonusers), and showed similar levels across all groups, suggesting that benzene exposure was occurring primarily from environmental sources, rather than tobacco product use.
All the nicotine-using groups (i.e., Smokers, ENDS Users and both Dual User strata) demonstrated similar levels of nicotine exposure, all significantly higher than those in tobacco product nonusers. This is consistent with the concept that ENDS are intended as alternative nicotine sources to allow smokers who are not quitting to maintain nicotine intake, while reducing exposure to HPHCs associated with cigarette smoking60,61. Used as a harm-reduction strategy, ENDS are intended to draw smokers away from combustible cigarettes, which are the most harmful and addictive nicotine-delivery product. Consistent with these conceptual models of switching and nicotine delivery, data from observational studies62 and randomized trials63,64 support the concept that adequate levels of nicotine are likely necessary to facilitate switching away from smoking.
Strengths of the study include the representative sample of US adults, large overall sample size and use of a dataset that includes more recent-generation ENDS devices. A limitation of the current analysis was that its measures of exposure were limited to select TSNAs, VOCs, three heavy metals, and nicotine. Other HPHCs such as polycyclic aromatic hydrocarbons (PAHs) were not available from PATH Study Wave 5 at the time of analysis (Version 18). The BOEs currently available from the PATH Study also did not include markers for exposures that might be higher in ENDS users. Finally, the analysis was a cross-sectional, between-groups comparison, so it does not directly show a within-participants reduction in HPHCs among smokers who switch to ENDS.
In summary, this nationally-representative analysis of a large sample of US adults found that exclusive users of ENDS have significantly and substantially lower exposures to a range of TSNAs, VOCs and one metal compared to smokers, with exposures among ENDS users often comparable to those seen in individuals not using any tobacco products at all. Such reductions in exposure have been taken as evidence of reductions in risk of smoking-induced disease9,10,11. In concert with the results of randomized clinical trials and longitudinal studies, the current results indicate that smokers who switch to ENDS likely experience reductions in health risk. The data also indicate that exposure to TSNAs, VOCs and metals among dual users was primarily a function of their cigarette smoking, as dual users who were smoking fewer than 10 cigarettes a day also consistently showed lower exposures compared to smokers. However, the data make clear that complete switching, with no smoking at all, results in reduced exposures compared to dual use with low levels of smoking. Further research on other toxicants, and also on biomarkers of potential harm, such as markers of inflammation, can inform reductions in health risks likely to accompany switching from smoking cigarettes to ENDS.
Data availability
The PATH Study datasets analyzed in this study are publicly available via application to the Inter-university Consortium for Political and Social Research (ICPSR) Virtual Data Enclave. Details are provided for the BRUF and RUF at https://doi.org/10.3886/ICPSR36840.v18 and https://doi.org/10.3886/ICPSR36231.v31, respectively.
Code availability
SAS code used to generate results for the current study are available in the Supplemental Code file.
References
World Health Organization. WHO Report on the Global Tobacco Epidemic, 2017: Monitoring Tobacco Use and Prevention Policies (World Health Organization, 2017).
Jha, P. & Peto, R. Global effects of smoking, of quitting, and of taxing tobacco. N. Engl. J. Med. 370(1), 60–68 (2014).
Jha, P. Avoidable global cancer deaths and total deaths from smoking. Nat. Rev. Cancer 9(9), 655–664 (2009).
USDHHS. Reports of the Surgeon General, in The Health Consequences of Smoking-50 Years of Progress: A Report of the Surgeon General (Centers for Disease Control and Prevention (US), 2014).
USDHHS. Publications and Reports of the Surgeon General, in How Tobacco Smoke Causes Disease: The Biology and Behavioral Basis for Smoking-Attributable Disease: A Report of the Surgeon General (Centers for Disease Control and Prevention (US), 2010).
US Food and Drug Administration. Harmful and potentially harmful constituents in tobacco products and tobacco smoke: Established list. Fed. Regist. 77(64), 20034–20037 (2012).
Chang, C. et al. Biomarkers of tobacco exposure: Summary of an FDA-sponsored public workshop. Cancer Epidemiol. Biomark. Prev. 26(3), 291–302 (2017).
Institute of Medicine. Scientific Standards for Studies on Modified Risk Tobacco Products (The National Academies Press, 2012).
NASEM In Public Health Consequences of E-Cigarettes (eds Eaton, D. L. et al.) (National Academies Press, 2018). Copyright 2018 by the National Academy of Sciences. All rights reserved, Washington (DC).
Akiyama, Y. & Sherwood, N. Systematic review of biomarker findings from clinical studies of electronic cigarettes and heated tobacco products. Toxicol. Rep. 8, 282–294 (2021).
Hiler, M. et al. Systemic biomarkers of exposure associated with ENDS use: A scoping review. Tob. Control https://doi.org/10.1136/tobaccocontrol-2021-056896 (2021).
Hartmann-Boyce, J. et al. Biomarkers of potential harm in people switching from smoking tobacco to exclusive e-cigarette use, dual use or abstinence: Secondary analysis of Cochrane systematic review of trials of e-cigarettes for smoking cessation. Addiction 118, 539–545 (2022).
Ward, A. M., Yaman, R. & Ebbert, J. O. Electronic nicotine delivery system design and aerosol toxicants: A systematic review. PLoS One 15(6), e0234189 (2020).
Henning, R. et al. Acrolein can cause cardiovascular disease: A review. Cardiovasc. Toxicol. 17(3), 227–236 (2017).
Dean, W. L. et al. Effects of the butadiene metabolite diepoxybutane on endothelial cell and platelet function. FASEB J. 23, 739.7 (2009).
Soeteman-Hernández, L. G., Bos, P. M. & Talhout, R. Tobacco smoke-related health effects induced by 1,3-butadiene and strategies for risk reduction. Toxicol. Sci. 136(2), 566–580 (2013).
Ashizawa, A. et al. Toxicological profile for 1, 3-butadiene (2012).
Olmedo, P. et al. Metal concentrations in e-cigarette liquid and aerosol samples: The contribution of metallic coils. Environ. Health Perspect. 126(2), 027010 (2018).
Williams, M. et al. Metal and silicate particles including nanoparticles are present in electronic cigarette cartomizer fluid and aerosol. PLoS One 8(3), e57987 (2013).
Potter, N. A. et al. Particulate matter and associated metals: A link with neurotoxicity and mental health. Atmosphere 12(4), 425 (2021).
Jaishankar, M. et al. Toxicity, mechanism and health effects of some heavy metals. Interdiscip. Toxicol. 7(2), 60–72 (2014).
Goniewicz, M. L. et al. Comparison of nicotine and toxicant exposure in users of electronic cigarettes and combustible cigarettes. JAMA Netw. Open 1(8), e185937 (2018).
De Jesús, V. R. et al. Urinary biomarkers of exposure to volatile organic compounds from the population assessment of tobacco and health study wave 1 (2013–2014). Int. J. Environ. Res. Public Health 17(15), 5408 (2020).
Smith, D. M. et al. Exposure to nicotine and toxicants among dual users of tobacco cigarettes and e-cigarettes: Population Assessment of Tobacco and Health (PATH) Study, 2013–2014. Nicotine Tob. Res. 23(5), 790–797 (2021).
Majeed, B. et al. Cluster analysis of urinary tobacco biomarkers among U.S. adults: Population Assessment of Tobacco and Health (PATH) biomarker study (2013–2014). Prev. Med. 140, 106218 (2020).
Lizhnyak, P. N. et al. Understanding heterogeneity among individuals who smoke cigarettes and vape: Assessment of biomarkers of exposure and potential harm among subpopulations from the PATH Wave 1 Data. Harm. Reduct. J. 19(1), 90 (2022).
Anic, G. M. et al. Changes in biomarkers of tobacco exposure among cigarette smokers transitioning to ENDS use: The population assessment of tobacco and health study, 2013–2015. Int. J. Environ. Res. Public Health 19(3), 1462 (2022).
Dai, H. et al. Exposure to toxicants associated with use and transitions between cigarettes, e-cigarettes, and no tobacco. JAMA Netw. Open 5(2), e2147891 (2022).
Cantrell, J. et al. History and current trends in the electronic cigarette retail marketplace in the United States: 2010–2016. Nicotine Tob. Res. 22(5), 843–847 (2018).
Coleman, B. et al. An Examination of device types and features used by adult electronic nicotine delivery system (ENDS) users in the PATH study, 2015–2016. Int. J. Environ. Res. Public Health 16(13), 2329 (2019).
Ali, F. R. M. et al. E-cigarette unit sales, by product and flavor type—United States, 2014–2020. Morb. Mortal. Wkly. Rep. 69(37), 1313 (2020).
Feng, J. et al. Urinary nicotine metabolites and self-reported tobacco use among adults in the population assessment of tobacco and health (PATH) study, 2013–2014. Nicotine Tob. Res. 24(5), 768–777 (2022).
Romberg, A. R. et al. Patterns of nicotine concentrations in electronic cigarettes sold in the United States, 2013–2018. Drug Alcohol Depend. 203, 1–7 (2019).
Talih, S. et al. Characteristics and toxicant emissions of JUUL electronic cigarettes. Tob. Control 28(6), 678–680 (2019).
Baig, S. A. & Giovenco, D. P. Behavioral heterogeneity among cigarette and e-cigarette dual-users and associations with future tobacco use: Findings from the Population Assessment of Tobacco and Health Study. Addict. Behav. 104, 106263 (2020).
Hyland, A. et al. Design and methods of the Population Assessment of Tobacco and Health (PATH) Study. Tob. Control 26(4), 371–378 (2017).
United States Department of Health and Human Services. National Institutes of Health. National Institute on Drug Abuse. Population Assessment of Tobacco and Health (PATH) Study [United States] Biomarker Restricted-Use Files. https://www.icpsr.umich.edu/web/NAHDAP/studies/36840/versions/V18 (cited 2023 April 10) (2022).
United States Department of Health and Human Services. National Institutes of Health. National Institute on Drug Abuse. Population Assessment of Tobacco and Health (PATH) Study [United States] Restricted-Use Files. https://www.icpsr.umich.edu/web/NAHDAP/studies/36231/versions/V31 (cited 2023 April 10) (2022).
Oliveri, D., Liang, Q. & Sarkar, M. Real-world evidence of differences in biomarkers of exposure to select harmful and potentially harmful constituents and biomarkers of potential harm between adult e-vapor users and adult cigarette smokers. Nicotine Tob. Res. 22(7), 1114–1122 (2020).
Cohen, G. et al. Changes in biomarkers of cigarette smoke exposure after 6 days of switching exclusively or partially to use of the JUUL system with two nicotine concentrations: A randomized controlled confinement study in adult smokers. Nicotine Tob. Res. 23, 2153–2161 (2021).
Jay, J. et al. Five-day changes in biomarkers of exposure among adult smokers after completely switching from combustible cigarettes to a nicotine-salt pod system. Nicotine Tob. Res. 22(8), 1285–1293 (2020).
Shahab, L. et al. Nicotine, carcinogen, and toxin exposure in long-term e-cigarette and nicotine replacement therapy users: A cross-sectional study. Ann. Intern. Med. 166(6), 390–400 (2017).
Wang, J. B. et al. Cigarette and e-cigarette dual use and risk of cardiopulmonary symptoms in the Health eHeart Study. PLoS One 13(7), e0198681 (2018).
Bhatta, D. N. & Glantz, S. A. Association of e-cigarette use with respiratory disease among adults: A longitudinal analysis. Am. J. Prev. Med. 58(2), 182–190 (2020).
Rostron, B. L. et al. Associations of cigarettes smoked per day with biomarkers of exposure among U.S. adult cigarette smokers in the population assessment of tobacco and health (PATH) study wave 1 (2013–2014). Cancer Epidemiol. Biomark. Prev. 28(9), 1443–1453 (2019).
Pulvers, K. et al. Effect of pod e-cigarettes vs cigarettes on carcinogen exposure among African American and Latinx smokers: A randomized clinical trial. JAMA Netw. Open 3(11), e2026324 (2020).
Cobb, C. O. et al. Tobacco-use behavior and toxicant exposure among current dual users of electronic cigarettes and tobacco cigarettes. Exp. Clin. Psychopharmacol. 29, 625–635 (2021).
Czoli, C. D. et al. Biomarkers of exposure among “dual users” of tobacco cigarettes and electronic cigarettes in Canada. Nicotine Tob. Res. 21(9), 1259–1266 (2019).
D’Ruiz, C. D., Graff, D. W. & Robinson, E. Reductions in biomarkers of exposure, impacts on smoking urge and assessment of product use and tolerability in adult smokers following partial or complete substitution of cigarettes with electronic cigarettes. BMC Public Health 16(1), 543 (2016).
O’Connell, G., Graff, D. W. & D’Ruiz, C. D. Reductions in biomarkers of exposure (BoE) to harmful or potentially harmful constituents (HPHCs) following partial or complete substitution of cigarettes with electronic cigarettes in adult smokers. Toxicol. Mech. Methods 26(6), 443–454 (2016).
Jacob, P. et al. Biomarkers of exposure for dual use of electronic cigarettes and combustible cigarettes: Nicotelline, NNAL, and total nicotine equivalents. Nicotine Tob. Res. 22(7), 1107–1113 (2020).
St Helen, G. et al. Comparison of systemic exposure to toxic and/or carcinogenic volatile organic compounds (VOC) during vaping, smoking, and abstention. Cancer Prev. Res. (Phila) 13(2), 153–162 (2020).
Piper, M. E. et al. Dual users compared to smokers: Demographics, dependence, and biomarkers. Nicotine Tob. Res. 21(9), 1279–1284 (2019).
Selya, A. S., Shiffman, S., Greenberg, M. & Augustson, E. M. Dual use of cigarettes and JUUL: Trajectory and cigarette consumption. Am. J. Health Behav. 45(3), 464–485 (2021).
Zervas, E. et al. Transfer of metals in the liquids of electronic cigarettes. Inhal. Toxicol. 32(6), 240–248 (2020).
Na, C. J. et al. The transfer characteristics of heavy metals in electronic cigarette liquid. Environ. Res. 174, 152–159 (2019).
Soulet, S. & Sussman, R. A. A critical review of recent literature on metal contents in e-cigarette aerosol. Toxics 10(9), 510 (2022).
Tareke, E. et al. Analysis of acrylamide, a carcinogen formed in heated foodstuffs. J. Agric. Food Chem. 50(17), 4998–5006 (2002).
Jackson, L. S. & Al-Taher, F. Effects of consumer food preparation on acrylamide formation. Adv. Exp. Med. Biol. 561, 447–465 (2005).
Abrams, D. B. et al. Harm minimization and tobacco control: Reframing societal views of nicotine use to rapidly save lives. Annu. Rev. Public Health 39, 193–213 (2018).
Abrams, D. B. et al. Managing nicotine without smoke to save lives now: Evidence for harm minimization. Prev. Med. 117, 88–97 (2018).
Goldenson, N. I. et al. Differences in switching away from smoking among adult smokers using JUUL products in regions with different maximum nicotine concentrations: North America and the United Kingdom. Nicotine Tob. Res. 23, 1821–1830 (2021).
Cobb, C. O. et al. Effect of an electronic nicotine delivery system with 0, 8, or 36 mg/mL liquid nicotine versus a cigarette substitute on tobacco-related toxicant exposure: A four-arm, parallel-group, randomised, controlled trial. Lancet Respir. Med. 9, 840–850 (2021).
Foulds, J. et al. Effect of electronic nicotine delivery systems on cigarette abstinence in smokers with no plans to quit: Exploratory analysis of a randomized placebo-controlled trial. Nicotine Tob. Res. 24(7), 955–961 (2021).
Author information
Authors and Affiliations
Contributions
Authors N.M.H., R.A.B., N.I.G., M.J.O. and S.S. conceived the analysis. N.M.H. and M.A.S. conducted the analysis. N.M.H. generated tables and figures. N.I.G., S.S., N.M.H., and M.J.O. drafted the manuscript. All authors reviewed and approved the manuscript.
Corresponding author
Ethics declarations
Competing interests
This study was funded by Juul Labs, Inc. N.M.H., R.A.B., N.I.G. and M.J.O. are full-time employees of Juul Labs, Inc. S.S. is a senior advisor to Pinney Associates, Inc, through which he provides consulting services on tobacco harm reduction on an exclusive basis to Juul Labs, Inc. M.A.S. is a full-time employee of Pinney Associates, Inc.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Holt, N.M., Shiffman, S., Black, R.A. et al. Comparison of biomarkers of exposure among US adult smokers, users of electronic nicotine delivery systems, dual users and nonusers, 2018–2019. Sci Rep 13, 7297 (2023). https://doi.org/10.1038/s41598-023-34427-x
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-023-34427-x
- Springer Nature Limited