Discussion

We conducted the largest chest physician survey regarding perceptions and practise patterns related specifically to ECs reported to date by querying all members of CHEST through an e-mail survey. Despite a low response rate, we were still able to ascertain opinions of nearly 1,000 clinicians in practise. We attempted to maximise the response rate by offering a cash incentive, keeping the survey length brief, and sending out the survey many times over the course of a year. Despite these interventions, the response rate remained low. We suspect that the primary reason for this is that compared to other groups in the general population, physicians have greater time constraints, a high paperwork and electronic messaging burden, and less need for the financial incentives offered. There may also be variability in physician interest in addressing tobacco dependence in the clinical setting, and survey respondents may represent the most motivated individuals in this regard. Although not the primary focus of this investigation, we included questions related to practise patterns with respect to traditional tobacco cigarettes and note similarity in reported patterns with respect to frequency of asking, advising, and treating tobacco dependence between our respondents and providers recently surveyed from the International Association for the Study of Lung Cancer (Warren et al., Reference Warren, Marshall, Cummings, Toll, Gritz and Hutson2013). This increases the likelihood that our respondent population is representative of prior physician survey groups despite the limited response rate.

We found that among respondents, health care provider perceptions regarding potential harms and benefits of ECs varied substantially. While most believed that regular EC use is not inherently safe, there was an even split as to whether these products are effective in promoting cessation from burned tobacco consumption. Our findings were similar to a recent national physician survey that found 30% of the respondents would recommend ECs for smoking cessation and 37% would recommend them for smoking reduction (Nickels, Warner, Jenkins, Tilburt, & Hays, Reference Nickels, Warner, Jenkins, Tilburt and Hays2016). Similarly, an earlier survey done among physicians of North Carolina found similar perceptions among physicians regarding EC use for smoking cessation (67% believing ECs are helpful for this purpose and 35% recommending them to patients) (Kandra, Ranney, Lee, & Goldstein, Reference Kandra, Ranney, Lee and Goldstein2014).

Although only 6% of our respondents thought ECs were more harmful than tobacco smoking, less than a quarter thought that a switch from regular tobacco smoking to daily EC use could improve their patients’ health. This was an apparent contradiction in the respondent perceptions, and we can only speculate as to why this difference in responses was observed. The results could indicate a belief that while a switch from tobacco to ECs could reduce harm in the short term, the reduction might be too little to have clinical significance on patients’ long-term health. There may also have been concern regarding unknown health risks with long-term EC use. Furthermore, nearly half of respondents (48%) expressed concern that ECs could increase the risk of relapsing into tobacco use. This suggests a view that even if ECs could reduce harm temporarily, it is possible that they will not lead to long-term smoking abstinence and would therefore not meaningfully impact long-term patient outcomes. Respondent comments did not routinely indicate a concern about high dose aerosol exposure from ECs as a potential mechanism preventing harm reduction among users switching from tobacco to EC use, but rather a general discomfort with unknown or unforeseen health risks intrinsic to ECs. In light of the fact that the long-term health risks of ECs remain unknown, we were surprised to find that more than half of respondents reported feeling comfortable discussing health effects of ECs. This proportion was higher than we expected given the current state of the scientific evidence.

Survey respondents indicated that they frequently encounter EC users in the clinical setting, and that users are likely to ask clinicians for advice with regard to these products. However, only 19% of respondents reported always asking about EC use, compared to 84% who always asked about smoking. We suspect that this discrepancy is related primarily to the fact that harms of tobacco smoking are well known and understood, while the nature and magnitude of long-term harms of EC use are still unclear. Although FDA recently deemed regulation of EC products, at present and in the near future much remains unknown regarding EC toxicity and efficacy in promoting cessation from combustible tobacco. Furthermore, because there is significant variation within EC products themselves (i.e. ‘cig-alike’ vs. ‘personal vaporizers’, or ‘tank systems’, different nicotine strengths, additives, and flavouring chemicals), it may be difficult to determine relative harms or benefits among products. Physicians might also be less likely to ask about EC use since there is no clear evidence or treatment guidelines for EC cessation. Since the products are still relatively new, physicians might be hesitant to address the topic or may view them as secondary in importance as compared to smoking.

Despite the overall lack of scientific evidence, previous surveys of smokers and EC users to date reveal that the majority of respondents generally viewed ECs as less harmful compared to tobacco smoking and in some cases perceiving some positive impact with regard to reducing or stopping tobacco smoking (Brose, Brown, Hitchman, & McNeill, Reference Brose, Brown, Hitchman and McNeill2015; Eastwood et al., Reference Eastwood, Dockrell, Arnott, Britton, Cheeseman and Jarvis2015; Mark, Farquhar, Chisolm, Coleman-Cowger, & Terplan, Reference Mark, Farquhar, Chisolm, Coleman-Cowger and Terplan2015; Martinez-Sanchez et al., Reference Martinez-Sanchez, Fu, Martin-Sanchez, Ballbe, Salto and Fernandez2015; Wackowski, Bover Manderski, & Delnevo, Reference Wackowski, Bover Manderski and Delnevo2016). We note the discrepancy in the perception of ECs between the majority of survey respondents and the RCP position, which generally looked more favourably upon ECs for their potential role for tobacco harm reduction.

This discrepancy underscores the importance of generating more high-quality scientific data regarding potential harms and benefits of ECs in order to inform both health professionals and the general public. As previously noted, the EC debate remains complex, but the aggregate evidence is lacking. From a regulatory standpoint, an important question surrounding ECs is how the potential benefits to some groups (i.e. tobacco smokers who might use them to quit smoking) compare to potential harms to other groups (i.e. adolescents, non-smokers, current tobacco smokers who might become dual users, and former smokers who become EC users or relapse onto tobacco smoking who would not have otherwise). FDA recently considered these factors, and released its regulations of ECs on May 5, 2016, which took a generally cautious approach to these products and required EC vendors to provide evidence that the products are safer than regular cigarettes. From a clinician's point of view, several unanswered questions regarding ECs are as follows:

1. (1) Will EC's assist in cessation from smoked tobacco, and if so, will this definitively lead to harm reduction? There have been very few randomised trials studying EC efficacy for tobacco cessation (Bullen et al., Reference Bullen, Howe, Laugesen, McRobbie, Parag and Williman2013; Caponnetto et al., Reference Caponnetto, Campagna, Cibella, Morjaria, Caruso and Russo2013). The largest trial conducted enrolled 657 smokers and randomised them to nicotine EC (16 mg/ml), non-nicotine EC, and nicotine patch (Bullen et al., Reference Bullen, Howe, Laugesen, McRobbie, Parag and Williman2013). Six-month follow-up quit rates were unexpectedly low in all groups and not statistically different (7.3%, 5.8%, and 4.1% in the nicotine e-cigarette, nicotine patch, and placebo e-cigarette groups, respectively). There was a noted difference in the proportion of smokers who reduced cigarette consumption by 50% or more between the e-cigarette and patch users. The study findings were limited by low statistical power and the use of an early generation EC that may not have had reliable nicotine delivery. Smaller studies have suggested that ECs can assist in smoking reduction among those not willing to stop smoking (Adriaens, Van Gucht, Declerck, & Baeyens, Reference Adriaens, Van Gucht, Declerck and Baeyens2014; Polosa, Caponnetto, Maglia, Morjaria, & Russo, Reference Polosa, Caponnetto, Maglia, Morjaria and Russo2014; Polosa et al., Reference Polosa, Caponnetto, Morjaria, Papale, Campagna and Russo2011; Tseng et al., Reference Tseng, Ostroff, Campo, Gerard, Kirchner and Rotrosen2016). Larger studies with more technologically advanced nicotine delivery systems would be helpful evidence for the field.

The most recent data suggest that puff-for-puff, ECs are less toxic compared to smoked tobacco (Goniewicz et al., Reference Goniewicz, Knysak, Gawron, Kosmider, Sobczak and Kurek2014; Hecht et al., Reference Hecht, Carmella, Kotandeniya, Pillsbury, Chen and Ransom2015). Still, toxicity from any inhaled substances is likely to be dose-dependent. If an individual stops smoking tobacco but begins vaping with a much higher rate of aerosol delivery, toxin inhalation could remain high and harm reduction may not occur. Furthermore, production of toxins (i.e. volatile carbonyl compounds) has been shown to increase significantly with higher temperature and electrical power applied to the e-liquid (Bekki et al., Reference Bekki, Uchiyama, Ohta, Inaba, Nakagome and Kunugita2014; Geiss, Bianchi, & Barrero-Moreno, Reference Geiss, Bianchi and Barrero-Moreno2016; Kosmider et al., Reference Kosmider, Sobczak, Fik, Knysak, Zaciera and Kurek2014; Talih, Balhas, Salman, Karaoghlanian, & Shihadeh, Reference Talih, Balhas, Salman, Karaoghlanian and Shihadeh2015). Consequently, since ECs and EC use patterns vary significantly with respect to product characteristics, it is inappropriate to think of ECs as a single entity with a given level of risk, and it would be more appropriate to view this as a class of products with a continuum of risk.

1. (2) Will EC's hinder cessation from smoked tobacco by providing behavioural cues that reinforce smoking without serving as a sufficient tobacco substitute? There is no evidence to date that ECs prevent cessation among motivated smokers seeking abstinence, but long-term studies have not yet been conducted. One reasonable concern is the possibility that ECs may deter smokers from seeking less harmful alternatives first (such as the nicotine patch or varenicline). Such a phenomenon could potentially reduce the degree of harm reduction achieved by a smoker.

2. (3) Will long-term EC use lead to unforeseen health consequences in the future? Although these questions will take many years to answer, it is hopeful that large controlled clinical trials and long-term observational cohort studies of EC and tobacco users will provide much-needed data to more accurately inform opinions both publically and within the medical community.

We note that our study has several limitations. First, our response rate was low, which limits the degree to which we can generalise our findings to account for the perceptions of all chest physicians. Second, we surveyed providers only within CHEST, and opinions and experiences of physicians outside of the organisation and in other specialties may vary. More than three quarters of respondents were from the US, and opinions from physician organisations in other parts of the world may differ. In particular, we had very few respondents from the UK, where public health officials have taken a much more positive view of ECs for tobacco treatment as compared to their US counterparts. Third, since knowledge regarding ECs is rapidly evolving, the perceptions and opinions here are likely subject to change over the next few years as more becomes known. As noted above, the RCP endorsed the use of ECs for tobacco users in the UK (Britton et al., Reference Britton, Arnott, McNeill and Hopkinson2016). Shortly thereafter, FDA issued several regulatory actions on ECs. Both of these events occurred after the administration of our survey, and it is unclear whether this would have affected our results if it occurred during the study period. Nevertheless, this study represents the largest survey of chest physicians’ knowledge, perceptions, and behaviours regarding ECs conducted to date.