NOT PEER REVIEWED In this interesting study by Cheah et al,1 the authors have raised several safety issues concerning electronic cigarettes. The majority of them were based either on the finding that nicotine content was inconsistent or that chemical constitution (for example glycols) may be hazardous to health.

There is some inconsistency in characterizing polypropylene glycol as "a known irritant when inhaled or ingested" in introduction section and as "a non-toxic chemical" in discussion section. This substance has been "generally recognized as safe" (GRAS) by FDA for ingestion, however, only few studies have evaluated the long term inhalation risk. A study by Robertson et al2 found that in experimental animals inhaling large quantities of propylene glycol for 12-18 months no lung, kidney, liver, spleen or bone marrow irritation or disease was observed, while others have used it on a daily basis as a vehicle for drug administration without finding any significant irritant effects on the respiratory tract. There are also no reports from electronic cigarette users that propylene glycol has caused any significant irreversible damage besides some throat irritation and cough, that has been resolved by using liquids not containing propylene glycol.

Concerning the production of carbonyls, it should be mentioned that electronically heated cigarettes that are mentioned in the study are in no way similar to electronic cigarettes. The temperature in electrically heated cigarettes is 600oC.3 Glycerol has a boiling point of 290oC, and diluting it with 10% water (usually, for electronic cigarette liquids glycerol is diluted with 15-20% water) the boiling point is 138oC,4 significantly lower than the 250oC needed to produce acrolein. These theoretical concepts have been backed up by data provided by Schripp et al,5 who found traces of formaldehyde not attributed to electronic cigarette use, and Romagna et al,6 who found no acrolein or formaldehyde in the air after electronic cigarette use for several hours.

An important finding of the study was the absence of nitrosamines. This has been somewhat underestimated by the authors. Nicotine in electronic cigarette liquids (and in other products, including pharmaceutical products) is derived from tobacco. Therefore, there is a possibility that nitrosamines may be present; in fact, they have been detected in approved nicotine gum and patch products.7 We think that the absence of nitrosamines is an important finding of this study and should have been further discussed, since they are major causes of lung disease including cancer.

Quality control during the production process is a major issue in electronic cigarette industry. Unfortunately, the fact that no regulation has been implemented by public health authorities allows low-quality products to be available to the market. This may raise safety issues like the presence of nitrosamines mentioned above. In addition, this is also the reason for the inadequate labeling of these products. We think that regulation standards should include proper labeling not only about the contents but also about the risks of accidental exposure to the liquids, similarly to other consumer products used daily in every home. However, we believe that the nicotine content discrepancies that the authors have found do not represent a major health risk, since it is well known that smoking is a dynamic process and changes in response to the yield characteristics of the cigarette.8 Most probably, the users would have adjusted their smoking pattern to the nicotine levels obtained by the use of these liquids. In fact, this has been a problem of tobacco cigarettes, with FTC protocol levels of nicotine significantly underestimating nicotine doses to smokers.8 This is accompanied by an underestimation of doses of carcinogens obtained by smokers, like nitrosamines, which were not found in electronic cigarettes tested in this study. Thus, it is not the nicotine that poses a health risk but other chemicals that are released during the smoking process.

References 1. Cheah NP, Chong NWL, Tan J, Morsed FA, Yee SK. Electronic nicotine delivery systems: regulatory and safety challenges: Singapore perspective. Tob Control, 2012. Dec 1 [Epub ahead of print]

2. Robertson OH, Loosli CG, Puck TT, Wise H, Lemon WM, Lester W Jr. Tests for the chronic toxicity of propylene glycol and triethylene glycol on monkeys and rats by vapor inhalation and oral administration. J Pharmacol Exper Ther 1947;91:52-76.

3. Patskan G, Reininghaus W. Toxicological evaluation of an electrically heated cigarette. Part 1: overview of technical concepts and summary of findings. J Appl Toxicol 2003;23:323-8

4. Flick EW. Industrial solvents handbook, 5th edition, 1998. ISBN 0- 8155-1413-1 Noyes Data Co, 1998

5. Schripp T, Markewitz D, Uhde E, Salthammer T. Does e-cigarette consumption cause passive vaping? Indoor Air 2012. Jun 2 [Epub ahead of print]

6. Romagna G, Allifranchini E, Bocchieto E, Todeshi S, Esposito M, Farsalinos K. Cytotoxicity of electronic cigarette vapor extract on cultured mammalian fibroblasts (ClearStream-Life project): comparison with tobacco smoke extract [abstract]. [http://www.srnteurope.org/assets/Abstract-Book-Final.pdf] Poster RRP17. 14th Annual Meeting of the Society for Research on Nicotine and Tobacco Europe, Helsinki, 2012. (accessed December 2012).

7. Cahn Z, Siegel M. Electronic cigarettes as a harm reduction strategy for tobacco control: a step forward of a repeat of past mistakes? J Public Health Policy 2011;32:16-31.

8. Djordjevic MV, Stellman SD, Zang E. Doses of nicotine and lung carcinogens delivered to cigarette smokers. J Natl Cancer Inst 2000;92:106 -11.

Conflict of Interest:

None declared

Most of us know the people who control Hollywood. Well, the Movie Industry is controlled in a similar manner, by their Cousins. They assist in the production of the films by, having their cancer causing product portrayed as a natural thing that your favorite stars do, so why aren't you? Films should have NO tobacco products in them whatsoever!!! If I had my way, I'd stop all tobacco production. If You want to smoke, grow it, cut it & wrap it, & you don't sell it to anyone!!

Conflict of Interest:

None declared

Omid Fotuhi,1 Geoffrey T Fong,1,2 Mark P Zanna,1 Ron Borland,3 Hua- Hie Yong,3 K Michael Cummings4

1. Department of Psychology, University of Waterloo, Waterloo, Ontario, Canada 2. Ontario Institute for Cancer Research, Toronto, Ontario, Canada 3. The Cancer Council Victoria, Melbourne, Victoria, Australia 4. Department of Health Behavior, Roswell Park Cancer Institute, Buffalo, New York, USA

Email for lead author, Omid Fotuhi: ofotuhi@uwaterloo.ca

NOT PEER REVIEWED Response to letter:

In our recent study--using a large set of nationally representative samples of smokers from Canada, the US, the UK, and Australia--we reported on the longitudinal patterns of smoking-related beliefs and how these beliefs vary with changes in smoking status. We found a consistent pattern of attitude-behaviour congruence: smokers highly endorsed risk-minimizing beliefs (e.g., "I have the genetic make-up that allows me to smoke without any health problems") and functional beliefs (e.g., "Smoking helps me concentrate"). But the most interesting finding was the longitudinal pattern of how these justifications for smoking changed over time as their smoking status changed: smokers endorsed these beliefs the least when they had quit; and again endorsed these beliefs to their pre-quit levels if they relapsed back to smoking, whereas the levels of endorsement of these beliefs stayed low among those smokers who had quit smoking and were able to stay quit in the long-term. We proposed that the waxing and waning of these smoking-related beliefs as a function of smoking status were driven by motivations to reduce cognitive dissonance (Festinger, 1957)--a fundamental human motivation to maintain consistency between one's attitudes and one's behaviours.

In response to these findings, Gould, Clough, and McEwen have offered a thoughtful commentary. In addition to writing about the importance for public health measures to target smokers' erroneous beliefs that smoking reduces stress, they agreed with our view that smokers are driven to modify their risk-minimizing beliefs because of their motivation to reduce dissonance.

However, Gould et al. suggest that an alternate mechanism is responsible for the longitudinal pattern of functional beliefs that we report in our study. Rather than being driven by dissonance-reducing motivations, they suggest that higher endorsements of functional beliefs among smokers are "representations of smokers' genuine experiences of nicotine withdrawal 'in between' cigarettes or on quitting."

We, on the other hand, do not see a contradiction between their interpretation and ours. Rather, we suggest that the physiological reactions to withdrawal and dependence are the starting point for the cognitive dissonance process. This is a view that has long been shared by dissonance researchers (e.g., Zanna, Cooper, & Taves, 1978; Croyle & Cooper, 1983).

So the Gould et al. account does not, at the core, differ from our account. They are pointing out the nature of the reasons for the justifications, which is the whole point of our argument: the fact that smokers are addicted and that they suffer withdrawal symptoms leads to the search for justifications for their smoking (rather than saying that "I am addicted"). The physiological symptoms of dependence and withdrawal can, therefore, lead to effects far outside the realm of the physiology of the smoker.

Thus, their account is not an alternative explanation--it may well be the starting point for what then become biases in cognitions to justify smoking.

In addition, when looking at the data from our study, we note that non-quitters endorsed both risk-minimising and functional beliefs more, compared to successful and failed quitters, at all three waves--even at times when all three groups were smoking (wave 1). Because it is unlikely that the pattern of risk-minimizing beliefs (e.g., "You've got to die someday, so why not enjoy yourself and smoke") is driven primarily by withdrawal symptoms--and given the strikingly similar pattern for both functional and risk-minimizing beliefs--we suggest that, at least in part, similar dissonance-reducing processes may also be responsible for the shifting of functional beliefs as smokers vacillate between smoking and having quit.

Furthermore, let us be clear that we do not claim that all smokers' smoking-related beliefs are distortions that serve only to reduce dissonance. We fully acknowledge that there may, in fact, be unique and genuine physiological experiences of nicotine consumption and withdrawal. We propose, however, that these experiences can more effectively be captured by specific measures that tap into the visceral aspects of nicotine addiction. For instance, the Hughes (1992) article cited by Gould and colleagues nicely captures these physiological experiences among quitters at various time points (e.g., increased irritability, hunger, restlessness, and cravings to smoke). These items are more directly representative of physiological responses to nicotine consumption and withdrawal than some of our functional beliefs measure (e.g., "Smoking is an important part of your life" or "Smoking makes it easier to socialize").

In fact, we would even argue that in comparison to risk-minimizing beliefs, functional beliefs are more readily employed in the service of dissonance reduction because they are less likely to be countered by reality constraints (Kunda, 1990). Specifically, we think that the functional beliefs in our study [(1) "You enjoy smoking too much to give it up"; (2) "Smoking calms you down when you are stressed or upset"; (3) "Smoking helps you concentrate better"; (4) "Smoking is an important part of your life"; and (5) "Smoking makes it easier for you to socialize"] are exactly the kind of malleable beliefs that smokers commonly employ--more so than the risk-minimizing beliefs which may be countered by rational thought (e.g., "The medical evidence that smoking is harmful is exaggerated")--to rationalize a behaviour that they know is harmful to their health.

Nonetheless, we appreciated the comments by Gould et al. because they encouraged us to take a closer look at our data and, consequently, to further think about our original interpretation of the findings.

We hope that further research continues to explore the role of attitudes in the domain of health behaviour, and specifically addictive behaviours, such as smoking. Experimental studies that more clearly determine causality and studies that examine the taxonomy of rationalizations commonly used by smokers would be especially useful for the advancement of this research topic. These findings would also have the important potential of informing policies to more effectively help save lives.

References

Croyle, R. T., & Cooper, J. Dissonance arousal: Physiological evidence. J Pers Soc Psychol. 1983;45:782-791.

Festinger L. A Theory of Cognitive Dissonance. Evanston, IL: Row, Peterson, 1957.

Hughes JR. Tobacco withdrawal in self-quitters. J Consult Clin Psychol. 1992;60(5):689-97.

Kunda Z. The case for motivated reasoning. Psychol Bull. 1990;108:480e98.

Zanna, M. P., & Cooper, J. Dissonance and the pill: An attribution approach to studying the arousal properties of dissonance. J Pers Soc Psychol 1974;29:703-709.

Conflict of Interest:

None declared

NOT PEER REVIEWED We wish to comment on the findings of Smerecnik et al.1 with respect to significant advances in genetic testing , which are highly relevant to their review. Unlike the early single genetic marker tests analysed by Smerecnik et al.,1 where subjects are dichotomised to positive or negative results, genetic susceptibility tests for lung cancer are now multivariate risk tests.2 These new risk tests incorporate clinical and genetic data to derive a composite gene-based risk score. In doing so they recognise that (1) environmental factors, like how much you smoked, are important and (2) genetic data alone is not sufficiently accurate to assess a person's risk. These distinguishing features of the recently developed lung cancer susceptibility tests are very important in assessing how patients respond for two reasons. First, this approach acknowledges that environment is important and that regardless of level of risk, all smokers can significantly mitigate that risk by quitting smoking (a unique feature of smoking-related lung diseases). This means there is no concern about genetic determinism (or nihilism), risk reduction is always possible. Second, in contrast to these early single marker tests, there are no "positive" and "negative" tests, all smokers tested have some level of risk and importantly, only lifelong non-smokers are "low risk" (more accurately reflecting the real-world situation). We have developed a gene-based lung cancer risk score based on a persons smoking, age, COPD, family history and genetic markers, where these previously validated variables are combined to derive a composite score.2 This score has been prospectively verified and assigns smokers to elevated, high and very high risk according to their total risk profile.3 We have assessed the potential clinical utility of this lung cancer risk score in a feasibility study where randomly selected smokers underwent brief counselling and were offered smoking cessation treatment. We found 84% of the smokers offered the test took the test, of which 52% took NRT and 28% had quit smoking 6 months after testing (2 fold and 5 fold greater than controls respectively).4 We conclude that our lung cancer susceptibility test improved the outcome of brief intervention, by facilitating the use of smoking cessation products (NRT) and subsequent quit rate. Such a finding concurs with the tension-trigger-treatment model proposed by Robert West where our gene-based test increased motivational tension, undermined optimistic bias and, for 30-50% of smokers, triggered a favourable outcome (NRT uptake and/or quitting smoking).

References 1. Smerecnik C, Grispen JE, Quaak M. Effectiveness of testing for genetic susceptibility to smoking-related diseases on smoking cessation outcomes: a systematic review and meta-analysis. Tob Control 2012; 21: 347-354. 2. Young RP, Hopkins RJ, Whittington CF, et al. Individual and cumulative effects of GWAS susceptibility loci in lung cancer: associations after sub -phenotyping for COPD. Plos One 2011; 6: e16467. 3. Young RP, Hopkins RJ, Hay B, Gamble GD. GWAS and candidate SNPs for COPD and lung cancer combine to identify lung cancer susceptibility: validation in a prospective study. Am J Respir Crit Care Med 2010; 181: A3738. 4. Hopkins RJ, Young RP, Hay B, et al. Lung cancer risk testing enhances NRT uptake and quit rates in randomly recruited smokers offered a gene- based risk test. Am J Respir Crit Med 2012; 185: A2590.

Conflict of Interest:

Dr Young has helped to develop a gene-based risk test for lung cancer susceptibility. Patents related to this test are held by Synergenz Bioscience Ltd who helped fund the research underlying the develeopment of this test.