Can E-Cigarettes and Pharmaceutical Aids Increase Smoking Cessation and Reduce Cigarette Consumption?

Findings From a Nationally Representative Cohort of American Smokers

Tarik Benmarhnia, John P. Pierce; Eric Leas; Martha M. White; David R. Strong; Madison L. Noble; Dennis R. Trinidad


Am J Epidemiol. 2018;187(11):2397-2404. 

In This Article

Abstract and Introduction


Many smokers believe that electronic nicotine delivery systems (ENDS) and pharmaceutical cessation aids can help them quit smoking or reduce cigarette consumption, but the evidence for e-cigarettes to aid quitting is limited. Examining 3,093 quit attempters in the nationally representative US Population Assessment of Tobacco and Health (PATH) Study, using data from 2013–2015, we evaluated the influence of ENDS and pharmaceutical cessation aids on persistent abstinence (≥30 days) from cigarettes and reduced cigarette consumption, using propensity score matching to balance comparison groups on potential confounders and multiple imputation to handle missing data. At PATH Wave 2, 25.2% of quit attempters reported using ENDS to quit during the previous year, making it the most popular cessation aid in 2014–2015. More quit attempters were persistently cigarette abstinent than were persistently tobacco abstinent (15.5% (standard error, 0.8) vs. 9.6% (standard error, 0.6)). Using ENDS to quit cigarettes increased the probability of persistent cigarette abstinence at Wave 2 (risk difference (RD) = 6%, 95% confidence interval (CI): 2, 10), but using approved pharmaceutical aids did not (for varenicline, RD = 2%, 95% CI: −6, 13; for bupropion, RD = 4%, 95% CI: −6, 17; for nicotine replacement therapy, RD = −3%, 95% CI: −8, 2). Among quit attempters who relapsed, ENDS did not reduce the average daily cigarette consumption (cigarettes per day, −0.18, 95% CI: −1.87, 1.51).


Electronic nicotine delivery systems (ENDS) became commercially available in the United States in 2007, when the rates of smoking cessation had stalled.[1] A recent comprehensive National Academies of Sciences, Engineering, and Medicine report[2] concluded that e-cigarettes have lower levels of toxicants than is found in smoke from combusted tobacco cigarettes, and experienced adult users of third-generation e-cigarettes (e.g., advanced personal vaporizers) are able to extract similar levels of nicotine to the those amounts obtained from cigarette smoking.[3] The National Academies of Sciences, Engineering, and Medicine report also noted: "For both individuals and for public health, the central potential benefit of e-cigarettes is to promote smoking cessation among established cigarette smokers or at least to reduce a smoker's exposure to combustible tobacco products" ([2]p. 423). At the same time, there is substantial evidence that e-cigarette use in young people increases the risk of future cigarette smoking.[4]

Current prevalence estimates for ENDS use in the United States vary with surveys, with Wave 1 of the US Population Assessment of Tobacco and Health (PATH) Study (2013–2014) indicating that 5.5% of the population were current users[5] and the 2014–15 Tobacco Use Supplement of the Current Population Survey indicating a lower 2.4%.[1] In this latter survey, current ENDS use rates were 19.0% for recent quitters and 11.5% for current smokers, with the majority being nondaily users. In the PATH Study, 83.5% of current and 76% of former cigarettes smokers[5] agreed that e-cigarettes could substitute in places where cigarettes were proscribed. Similarly, approximately 75% agreed that e-cigarettes can help people quit smoking cigarettes.

There is a substantial literature of randomized controlled trials showing the efficacy of nicotine replacement, varenicline, and bupropion as smoking cessation aids[6] but few trials of e-cigarettes. For e-cigarettes, meta-analyses[2] consistently identify only 2 international trials that appropriately address the role of e-cigarettes in cessation. Bullen et al.[7] randomized smokers who wanted to quit to either e-cigarettes, nicotine patches, or placebo e-cigarette. Biochemically verified 6-months continuous abstinence was not significantly higher in the e-cigarette group than in the nicotine patch group or the placebo control (7.3% vs. 5.8% vs. 4.1%). Caponnetto et al.[8] compared e-cigarettes with a placebo e-cigarette, and verified 6-months continuous abstinence was 11% vs. 4%, which reached borderline significance.

In contrast to randomized trials, population-based longitudinal studies reflect how e-cigarettes are actually being used in the target population. Six longitudinal studies of e-cigarettes in cessation[9–14] met the inclusion criteria in both recent systematic reviews that the National Academies of Sciences, Engineering, and Medicine judged to be the most comprehensive and rigorous.[15,16] Two of these[13,14] focused exclusively on patient populations and as such are not generalizable to the general population. All the others were considered to have significant methodological flaws that limited the quality of conclusions, including lack of identification and control of known prognostic factors for successful cessation. In particular, higher nicotine dependence is positively associated with use of a cessation aid as well as negatively associated with cessation success. Another National Academies of Sciences, Engineering, and Medicine criticism focused on the inclusion of participants who were using e-cigarettes for reasons other than as a cessation aid. Additionally, there was an expressed concern that relapsed smokers might not have recalled all of their unsuccessful quit attempts or had degraded their significance by recalling them as "not really a quit attempt." This proposed pattern of differential recall of past quit attempts could lead to a bias toward fewer recalled failures.[17,18] However, the suggestion that this recall bias is differential according to use of cessation aids is not convincing.[19]

Understanding the potential role that ENDS products may play in cessation requires direct comparisons with recommended efficacious pharmaceutical products (varenicline, bupropion, and nicotine replacement therapy (NRT)).[20] The effectiveness of the dissemination of these aids at the population level has been questioned,[21] and these findings need replication.

In this study, we used the first 2 waves from the large, nationally representative PATH Study[22] to evaluate the influence of ENDS and pharmaceutical cessation aids on persistent abstinence (≥30 days) from cigarettes as well as reduced cigarette consumption, using propensity score matching (PSM) to balance comparison groups on potential confounders.