Henley et al’s paper (1) showing worse health outcomes in men switching from cigarettes to smokeless tobacco, compared with men ceasing tobacco use completely, adds to our understanding of the potential risks from smokeless tobacco use. However, it also raises some additional questions:

1. Like the authors’ earlier paper comparing health outcomes in exclusive smokers with those of exclusive smokeless users in CPS-II (2), this paper did not report a comparison with those people who continued to smoke. Papers by other groups examining the health outcomes from potential harm-reducing behavior changes (3,4) have presented the whole picture, comparing the outcomes for continuing smokers, never smokers and those making the potentially less harmful change (e.g. reducing cigarette consumption). Occasionally clinicians are asked by smokers who don’t want to quit tobacco, whether their health risks would be reduced by switching to smokeless tobacco. This question may come up more frequently with some of the major cigarette manufacturers now test-marketing smokeless products. The CPS studies have the data to help answer that question. The public should be informed just how much their chances of premature death from lung cancer, COPD etc are likely to be reduced by using smokeless tobacco rather than smoking. The authors should be encouraged to analyze and publish those data as well. A survival curve comparing never tobacco users, smokers and smokeless tobacco users would be helpful, as would the adjusted risks of each tobacco-related disease for each group.

2. Also like the previous paper (2), this study found raised risks of death from lung cancer and COPD among those who switched to smokeless tobacco. For example, among snuff users (27% of switchers), the adjusted hazard ratios for all cause mortality (1.11, 95% CI=0.94-1.3), coronary heart disease (1.12: 0.82-1.53) and stroke (0.89: 0.49-1.62) were not significantly elevated, and were lower than those for lung cancer (1.75: 1.2-2.5) and COPD (1.68: 0.9-3.3). The authors have pointed out the possibility that the increased lung cancer risk could be caused by circulating carcinogens from the tobacco. However, the authors did not speculate on the possible cause of increased risk of death from COPD among those switching to smokeless tobacco compared with those quitting completely. It is hard to think of mechanisms that do not involve increased exposure to smoke, either from secondhand smoke, or increased smoking (including smoking other substances) before or after recruitment to the study. It would be useful to hear the authors’ thoughts on what caused the smokeless users’ raised COPD risks, and also how that might affect interpretation of the other raised risks found in these studies (i.e. are these effects likely due to confounding with smoke exposure, rather than smokeless use per se?).

3. This excellent study by the American Cancer Society reported increased risks of cancer of the oral cavity and pharynx (HR=2.5, CI=1.2- 5.7), based on 7 deaths in switchers, and the previous paper comparing exclusive smokeless users with never tobacco users in CPS-II found an adjusted hazard ratio for oropharynx cancer of 0.90 (0.12-6.71). The US American Cancer Society website currently states that:

“Smokeless tobacco ("snuff" or chewing tobacco) is associated with cancers of the cheek, gums, and inner surface of the lips. Smokeless tobacco increases the risk of these cancers by about 50 times.” (5)

The authors have previously stated that, “We do believe that there has been inadequate concern about potential adverse risks of spit tobacco use”(6). In fact the available evidence suggests that the public drastically overestimates the relative risks from smokeless tobacco. For example, only 11% of smokers believe that smokeless is less harmful than cigarettes (7). Perhaps the information on the ACS website should be updated to be more consistent with the results of these two ACS studies so as not to add to the public’s biased perception?

Lastly, this paper is important to informing the harm reduction debate as it pertains to smokeless tobacco, but it only contributes to part of the story. It fails to point out that the largest difference in risk is likely to be the one between switchers and continuing smokers, while the difference between switchers and complete quitters is relatively small. It is not surprising that those that switch to another form of tobacco may have elevated health risks compared to those who quit tobacco entirely. But what is sorely needed is analysis of the risks of switching to a potentially less harmful tobacco product (smokeless) versus continuing to smoke the most deadly form of tobacco, the manufactured cigarette.

1. Henley SJ, Connell CJ, Richter P, Husten C, Pechacek T, Calle EE, Thun MJ. Tobacco-related disease mortality among men who switched from cigarette to spit tobacco. Tobacco Control 2007;16:22-28

2. Henley SJ, Thun MJ, Connell C, Calle EE. (2005) Two large prospective studies of mortality among men who use snuff or chewing tobacco (United States). Cancer Causes and Control 16:347-358

3. Godtfredsen NS, Holst C, Prescott E, Vestbo J, Osler M. Smoking reduction, smoking cessation, and mortality: a 16-year follow-up of 19,732 men and women from The Copenhagen Centre for Prospective Population Studies. Am J Epidemiol. 2002 Dec 1;156(11):994-1001.

4. Tverdal A, Bjartveit K. Health consequences of reduced daily cigarette consumption. Tob Control. 2006 Dec;15(6):472-80.

5. www.cancer.org “Detailed Guide: Oral Cavity and Oropharyngeal Cancer What Are The Risk Factors for Oral Cavity and Oropharyngeal Cancer?” (accessed Feb 14, 2007)

6. Henley SJ, Thun MJ. Response to: Foulds J and Ramstrom L letter regarding "Causal effects of smokeless tobacco on mortality in CPS-I and CPS-II". Cancer Causes Control. 2006 Aug;17(6):857-8.

7. O'Connor RJ, Hyland A, Giovino GA, Fong GT, Cummings KM. Smoker awareness of and beliefs about supposedly less-harmful tobacco products.Am J Prev Med. 2005 Aug;29(2):85-90.

NOT PEER REVIEWED To the Editors, In the article entitled, "Weight control belief and its impact on the effectiveness of tobacco control policies on quit attempts: findings from the ITC 4 Country Project" I noticed a problem regarding the measurement of weight control beliefs. This variable (weight control beliefs associated with tobacco use) is measured using only one question. The researchers indicate, "In order to identify weight concerns related to smoking, we exploit a question that measures smokers' level of agreement with the following statement using a 5-point scale (strongly agree, agree, neither agree nor disagree, disagree and strongly disagree): Smoking helps weight control" (Shang et. al, p.2, 2015). This statement illustrates the limited manner in which the aforementioned variable was measured. While the limitation of weight control beliefs being analyzed using self- reporting was addressed, the limitation of using only one question to measure this variable was not. In the study entitled, "Smoking Expectancies, Weight Concerns, and Dietary Behaviors in Adolescence" the authors noted that they used the appetite control factor of the Smoking Consequences Questionnaire (SCQ) to determine weight control beliefs. The author of "Smoking Expectancies, Weight Concerns, and Dietary Behaviors in Adolescence indicates, "Participants who endorsed smoking were given 5 possible consequences of smoking and were asked to rate the likelihood of each consequence on a 10-point scale from 'completely unlikely' to 'completely likely.' The statements included, 'Smoking controls my appetite,' 'Smoking keeps my weight down,' 'Cigarettes keep me from overeating,' 'Cigarettes keep me from eating more than I should,' and, 'Smoking helps me control my weight.' Scores were an average across all items" (Cavallo et. al., p. 68, 2010). This multifaceted approach to measuring a variable is a more thorough and a more accurate measure of the weight control variable. This more detailed measure, as indicted by the author, has been measured by three different criteria: internal consistence, degree of factor loading, and coefficient significance (.72 to .97). This measure starkly compares to the measure used in the article, "Weight control belief and its impact on the effectiveness of tobacco control policies on quit attempts: findings from the ITC 4 Country Project", which was only measured in its degree of sensitivity.

References

Cavallo, D. A., Smith, A. E., Schepis, T. S., Desai, R., Potenza, M. N., & Krishnan-Sarin, S. (2010). Smoking expectancies, weight concerns, and dietary behaviors in adolescence. Pediatrics, 126(1), e66-e72.

Shang, C., Chaloupka, F. J., Fong, G. T., Thompson, M., Siahpush, M., & Ridgeway, W. (2015). Weight control belief and its impact on the effectiveness of tobacco control policies on quit attempts: findings from the ITC 4 Country Project. Tobacco control. doi:10.1136/tobaccocontrol- 2014-051886

Conflict of Interest:

None declared

Foulds and Ramström raise important questions regarding a direct comparison of mortality rates among smokers, smokeless tobacco (ST) users, persons with mixed or former use, and non-users. They urge officials from the Centers for Disease Control and Prevention (CDC) and from the American Cancer Society (ACS) to make these comparisons and report the results, so that Americans are fully informed about the health risks related to tobacco use. But there is a simpler and more compelling solution: The CDC must release publicly all data it uses to estimate the relative risks and mortality rates among tobacco users.

Every year the CDC publishes statistics concerning how many Americans smoke, and how many Americans die as a consequence (1,2). These statistics form the raison d’être for current tobacco policies at all levels of American government – and for the massive regulatory scheme currently under consideration by the U.S. Congress.

The data from which the CDC estimates prevalence of tobacco use are publicly available from the National Health Interview Surveys. In stark contrast, the data from which the CDC estimates deaths from tobacco use are not available to researchers outside the agency or its collaborator, the ACS. Instead, the CDC takes a black-box approach of filtering information on mortality through its online program called Smoking- Attributable Mortality, Morbidity, and Economic Costs (SAMMEC) (3).

But SAMMEC is marginally informative, and utterly unsatisfactory. It does not provide any information comparing the mortality experience of smokers and ST users. It cannot even provide simple statistics like the number of deaths among current and former smokers. In 2006 I submitted a request for these data through the SAMMEC web site. I received this unsigned response from the CDC Office on Smoking and Health: “Data are not available for current or former smokers separately.”

The public release by the CDC of data relating to tobacco-related mortality will also place the agency in compliance with the intention of the NIH Data Sharing Policy (4), which states that “data sharing is essential for expedited translation of research results into knowledge, products, and procedures to improve human health.”

Brad Rodu

Professor of Medicine

Endowed Chair, Tobacco Harm Reduction Research

University of Louisville

Competing Interests: Dr. Rodu's research is supported by unrestricted grants from two smokeless tobacco manufacturers to the University of Louisville. More information is available at www.smokersonly.org

References

1. Centers for Disease Control and Prevention 2005: Cigarette smoking among adults – United States, 2004. MMWR 54:1121-1124.

2. Centers for Disease Control and Prevention, 2005: Annual smoking- attributable mortality and years of potential life lost, and productivity losses – United States. MMWR 54:625-628. 1997–2001.

3. Smoking-Attributable Mortality, Morbidity, and Economic Costs (SAMMEC). Available at: http://apps.nccd.cdc.gov/sammec/login.asp

4. NIH Data Sharing Policy. Available at: http://grants.nih.gov/grants/policy/data_sharing/

On March 15 2007, my attention was drawn to a patent for a tobacco smoking device, filed with the U.S. Patent and Trademark Office (USPTO) for a "Hookah with simplified lighting" on June 9 2005. One of the authors of the device being patented was Kamal Chaouachi, who on December 2 2004, had a rapid response published in Tobacco Control [1] which was critical of a paper by Masiak et al [2]. The submission process for rapid responses asks authors to “Please declare any competing interests”. I noted that Dr Chaouachi’s letter contained no competing interest statement and so wrote to him requesting that he submit a further rapid response which would clarify his competing interest.

He duly submitted two rapid responses. The first, which I have not published but retained, simply said “no competing interests”. I replied that this brief response was unacceptable and that in the circumstances of the revelation about his ostensible interest in the smoking device, he should elaborate in a further rapid response.

In the second response Dr Chaouachi stated that he had signed away his rights “in the presence of a State Attorney” to his tobacco smoking invention patent on June 15 2005, some six days after the US patent was filed. Dr Chaouachi reiterated that “I had no competing interest at the time my Letter to the Editor entitled 'Serious Effors in this Study" was sent to the Tobacco Control Journal.”

This last statement, in fact, is false. I am in possession of a report from the French patent office (Bulletin Officiel De La Propriete Industrielle Brevets D'Invention). At page 18 a patent in the names of Billard, Chaouachi (Kamal), and De La Giraudiere is described. The patent number is 04 06287, the company filing the patent is "Shishamania International", the title of the patent is "NARGUILLE A ALLUMAGE SIMPLIFIE". The date of filing is June 10 2004.

Thus, the US patent, filed at the US PTO on June 9, 2005, was first filed in France on June 10, 2004. Dr Chaouachi’s e-letter was submitted on December 2 2004. Therefore, the e-letter was submitted after the French patent was filed, and before the date on which he reports that he relinquished his rights to the patent.

In further correspondence during March 17, I confronted Dr Chaouachi with the fact that his device had been registered with the French patent authorities in June 2004 and that therefore he had made a false statement in his declaration of no competing interests. He replied “This is not "false statement" (!). … The French patent was filed at the date you said. You are probably right as I cannot say myself so far when it was. I have been informed of procedure. This is all.”

In summary, at the date Dr Chaouachi submitted his rapid response, he had a commercial interest in the subject of that submission. He did not declare this interest. When later given the opportunity to do so, he maintained that he had no competing interest at the time of writing his rapid response. He did not voluntarily declare that the patent had in fact been filed in France in June 2004.

In such circumstances, it is Tobacco Control’s policy to inform offending authors’ institutions of such conduct. Dr Chaouachi would appear to not be currently working for any institution. Tobacco Control is unwilling to accept any further submissions from Dr Chaouachi.

Simon Chapman Editor

1. Chaouachi K. Serious Errors in this Study. http://tc.bmj.com/cgi/eletters/13/4/327 rapid response

2. Maziak W, Ward KD, Afifi Soweid RA, Eissenberg T. Tobacco smoking using a waterpipe: a re-emerging strain in a global epidemic Tob Control 2004; 13: 327-333