Jump to comment:

• A Belated Reply to Foulds et al. and Bates
  Scott L. Tomar, DMD, DrPH
  Published on: 3 March 2009
• Still no response - but there's still time
  Clive D Bates
  Published on: 20 February 2009
• The subsequent response?
  Clive D Bates
  Published on: 9 April 2004
• Evidence and argument over smokeless tobacco – another response to Tomar et al
  Clive Bates
  Published on: 31 December 2003
• A Reply to Bates et al. and Foulds et al.
  Scott L. Tomar
  Published on: 19 December 2003
• Another simple 'quit or die' statement
  Clive Bates
  Published on: 14 December 2003
• "Another simple issue"?
  John R. Polito
  Published on: 7 December 2003
• A reply to Tomar et al's flat earth commentary
  Clive Bates
  Published on: 6 December 2003
• Effects of smokeless tobacco in Sweden: a reply to Tomar et al.
  Jonathan Foulds
  Published on: 5 December 2003

• Published on: 3 March 2009

  A Belated Reply to Foulds et al. and Bates

  • Scott L. Tomar, DMD, DrPH, Professor and Department Chair
  • Other Contributors:
    • Greg N. Connolly

  First, an apology is in order for taking so long to respond to the online discussion surrounding the review by Foulds et al. [1] and the opinion piece by Bates et al. [2]. As we had promised in our earlier reply to Foulds et al. (19 December 2003) and have been reminded by Bates, we are belatedly responding to the specific points raised by Foulds et al. in their e-letter dated 5 December 2003:

  1. “Misrepresentation of...

  Show More

  First, an apology is in order for taking so long to respond to the online discussion surrounding the review by Foulds et al. [1] and the opinion piece by Bates et al. [2]. As we had promised in our earlier reply to Foulds et al. (19 December 2003) and have been reminded by Bates, we are belatedly responding to the specific points raised by Foulds et al. in their e-letter dated 5 December 2003:

  1. “Misrepresentation of our review.” Our commentary did not misrepresent the conclusions reached by Foulds et al. [1]. We cited their direct quote that snus had “...a direct effect on the changes in male smoking and health” and made the observation that their review added little additional evidence to support that conclusion beyond the spotty evidence cited by Bates et al. (and those two papers had several co-authors in common). Yes, we read their 11 journal pages, 8 figures, 2 table, and 66 references, as well as the 8 journal pages Tobacco Control generously devoted to Bates et al. [2]. No one in the mainstream scientific community questions the underlying premise that exclusive use of snus conveys lower risks for death and disease than does exclusive cigarette smoking. The primary question is whether snus was responsible for the decline in smoking in Sweden and related disease patterns. In support of that hypothesis, Foulds et al. cite sales data from Swedish Match, trend data for tobacco use among men and women age 18–70 years that was unadjusted for age, and cross-sectional data from two northern Swedish counties. That evidence for the role of snus in improving public health does, in fact, provide little additional evidence to what was cited by Bates et al. Foulds et al. make much of the sex differences in use of tobacco products in Sweden to support their hypothesis, yet their reliance on crude (unadjusted) patterns actually masks recent trends in tobacco usage in Sweden and undermines their conclusions. In reality, true age-adjusted smoking initiation rates and cessation rates for males and females in Sweden are essentially equal [3].

  2. “Selective reporting of findings”. Foulds et al. acknowledge that they omitted several studies because they were from a part of the country where snus usage was low, claiming that “(b)asing conclusions about snus use in Sweden on a study based exclusively in Malmo is like basing conclusions on smoking and smokeless use in the USA on studies in Utah.” Fair enough, although Foulds et al. had fewer concerns with drawing conclusions about the role of snus in Sweden as a whole based on patterns in northern Sweden.

  We stand corrected on our statement that “between 1981 and 2001 daily smoking declined more rapidly for 15-16 year old girls (23% to 16%) than boys (13% to 10%), snus use remained rare among girls, and the sex difference in smoking prevalence decreased.” As the figure presented by Foulds et al. in their electronic letter indicates, the rate of decline in smoking among 15–16-year-olds was about the same for boys and girls. However, that pattern, coupled with a high and increasing level of snus usage among adolescent males in Sweden and very low levels of usage among adolescent girls, provides little support for the conclusion reached by Ramstrom and Foulds [4], two of the authors of the 2003 review paper, that use of snus in Sweden is “associated with a reduced risk of becoming a daily smoker.” If that were truly a causal association, we would expect the initiation rate to be declining more rapidly among boys than among girls due to the much greater growth is snus usage among boys, but it is not. Although there was a 10-fold difference in snus usage between 15–16 year-old boys and girls (20% vs. 2%), smoking initiation exhibited a rather constant and much more modest 3–6 percentage point difference during that time period. Foulds et al. did not respond in their e-letter to broader national trends in Sweden for young males and females that we mentioned in our commentary. We present here more recent data for 16–24-year-olds in Sweden, the age range during which nearly all smoking initiation occurs [3]:

  Figure 1. Trends in proportion of persons age 16–24 who used snuff daily or currently smoked (daily or occasionally), by sex. Sweden, 1988–1989 to 2004–2005. Data from Statistics Sweden ULF Surveys.

  

  Secular trends in tobacco use among adolescents and young adults in Sweden (or Norway and the United States, for that matter) do not support a preventive effect of smokeless tobacco use for cigarette smoking. Official national data from Statistics Sweden indicate that daily snuff use among 16-24 year-olds has increased over the past 15 years, from 23.0% in 1988–1989 to 26.5% in 2005 among males and from 0.6% to 3.9% among females (Figure 1). Current smoking (i.e., daily or occasional) in that age group exhibited a flat trend line for males during that time period and a declining trend line for females over the same time period. In 2004–2005, 33.4% of males and 30.2% of females aged 16–24 years in Sweden were current smokers. However, the prevalence of daily smoking was lower for males (9.3%) than for females (13.3%) in that age group. The secular patterns in tobacco use among Swedes aged 16–24 years suggest that snuff may have served as a partial substitute for smoking among males, but had a negligible effect, if any, on smoking initiation rates for either sex.

  3. “Tomar et al’s errors in critical appraisal of health effects of snus”. We stand by the original comments regarding the interpretation of the studies by Lewin et al. [5] and Schildt et al. [6]. In the study by Lewin et al., there was an elevated risk for head and neck cancers among an important subgroup (lifetime non-smokers); Foulds et al. only cite the non-significant multivariate relative risk estimate, ignoring that subgroup analysis or the obvious colinearity that occurs in multivariate modeling when nearly all snus users also have a history of smoking. [Since that time that study was published, its lead author, Freddi Lewin, has gone on to a career with Swedish Match]. In contrast, Foulds et al. chose to report only the univariate analyses in Schildt et al., when the multivariate analysis found that neither smoking, alcohol consumption, nor snus were associated with oral cancer. We may be “out on a limb” regarding the carcinogenicity of snus, but we appeared to be joined on that limb by the International Agency for Research on Cancer [7] and the European Commission’s Scientific Committee on Emerging and Newly Identified Health Risks [8].

  4. “Birth-cohort patterns relating smoking and snus use.” The birth cohort data presented by Foulds et al. do not support a role for snus in smoking cessation or refute our contention that most snus uptake occurred among young people because they present no data on snus usage. All birth cohorts of men and women experienced declines in daily smoking during that 17-year period; for most birth cohorts the difference between men and women was 2–7 percentage points. Differences were larger for older adults (8–11 percentage points), but women aged 50 years and older began the cohort study with much lower prevalence of smoking. Non-daily smoking remains unreported, even though it accounts for a large proportion of current smoking in Sweden. Most smoking cessation trials would not consider reduction from daily smoking to less than daily smoking as cessation, but perhaps the authors’ clinics and trials use different criteria. The primary point is that evidence such as Figure 1 in the review by Foulds et al., which presents trend data from Swedish Match on consumption of snus and cigarettes, or the cohort data on smoking they posted in their e-letter cannot determine whether the groups taking up snus are necessarily the same ones driving the decline in cigarette consumption; we continue to contend that they largely are not.

  5. “Is the sex difference in smoking prevalence due to fewer women in the smoke-free workplace?” We acknowledge that our hypothesis was speculative and lacked direct supporting evidence. However, in their e-letter, Foulds et al. stated: “One thing that doesn’t seem to fit with that is the data on the older age groups presented in the table above. Those aged 50+ in 1980–1 in that (sic) data would mostly have retired from the workforce by 1996–7 and so might be less affected by workplace smoking bans. However, despite that, the sex-difference in cessation is actually stronger in that age group than any other.” That conclusion is not necessarily true; the prevalence of smoking was substantially higher among men age 50 or older than among same-aged women, men were more likely than women to be lost to follow-up by the end of the study, and a greater proportion of men (9.2%) than women (6.2%) died between the second interview and the time of the third follow-up interview [9]. Some of that “sex-difference in cessation” was very likely to have been differential mortality between older men and women, driven heavily by higher smoking-attributable mortality rates among older men. One sure way to quit smoking is to die.

  6. “Use of snus as a smoking cessation aid”. Foulds et al. claim that we tried to “brush this important piece of evidence (of snus as a smoking cessation aid) under the carpet.” In fact, in our commentary we stated, “…the large majority of men (71%) and women (97%) who quit smoking did not use snus at their last quit attempt, with modest effectiveness for snus as a cessation strategy in that observational study. That is hardly compelling evidence for snus as "an important explanation" for the decline in smoking in Sweden.” We did not claim that no smokers in Sweden quit smoking by using snus, but we do maintain that the role of snus in reducing smoking has been substantially overstated. When half of the adult population (women) has never used a purported smoking cessation method but still achieved a greater prevalence of complete smoking cessation than the half that has widely adopted that method, it suggests that there are other, more important factors that explain the decline in smoking in Sweden.

  7. “This is about Sweden, not the USA.” While that may be true, USA remains the world’s largest market for commercial moist snuff products and reviews such as those by Foulds et al. and an opinion piece such as that of Bates et al. can have significant effects on U.S. tobacco policy, its tobacco industry, and the usage of tobacco products. A great deal has happened since those papers were published six years ago. Nearly the entire moist snuff market in the USA is now controlled by cigarette manufacturers, who are developing and test-marketing new smokeless tobacco products at a furious pace. New products include Marlboro Snus, Camel Snus, and Camel Dissolvables that include Camel Orbs, Strips and Sticks. Those products are largely being positioned as complements to cigarette smoking, not substitutes, and we are likely to see a growth in dual product usage [10]. The primary target audiences in test marketing appear to be young people, not middle-age smokers looking to reduce their risks from smoking. Those companies also continue to heavily promote their traditional moist snuff products to young males, as evidenced by a 12-page advertising insert the January 2009 issue of Playboy magazine.

  It remains to be seen whether snus or more traditional U.S. types of moist snuff will be adopted by smokers as substitutes for cigarettes, but so far they have not gained much traction [11]. While it is true that we may have a relatively U.S.-centric focus, we do note that nearly all U.S. states have achieved a lower prevalence of smoking than has Sweden. Even using the more stringent definition of daily smoking among persons age 15 years or older, about one-half of U.S. states have levels of smoking equal to or lower than Sweden's, with relatively little use of moist snuff despite its widespread availability.

  8. “Both snus and Swedish tobacco control deserve some of the credit”. We obviously have a different interpretation of the situation in Sweden. Our skepticism and concerns are fueled, in part, by the lack of an evident public health benefit in neighboring Norway, which has seen a skyrocketing prevalence of snus usage among young males with no apparent impact on smoking initiation or cessation rates [3].

  Disclosures
  Scott Tomar is currently serving as an expert witness for plaintiffs in product liability law suits brought against a smokeless tobacco manufacturer and against a cigarette manufacturer.
  Greg Connolly has no conflicts to disclose.
  
  Scott L. Tomar
  University of Florida College of Dentistry
  
  Greg N. Connolly
  Harvard School of Public Health
  

  References
  [1] Foulds J, Ramstrom L, Burke M, Fagerstrom K. Effect of smokeless tobacco (snus) on smoking and public health in Sweden. Tob Control. 2003;12(4):349-59.
  [2] Bates C, Fagerstrom K, Jarvis MJ, Kunze M, McNeill A, Ramstrom L. European Union policy on smokeless tobacco: a statement in favour of evidence based regulation for public health. Tob Control. 2003;12(4):360-7.
  [3] Tomar SL. Epidemiologic perspectives on smokeless tobacco marketing and population harm. Am J Prev Med. 2007;33(6 Suppl):S387-97.
  [4] Ramstrom LM, Foulds J. Role of snus in initiation and cessation of tobacco smoking in Sweden. Tob Control. 2006;15(3):210-4.
  [5] Lewin F, Norell SE, Johansson H, Gustavsson P, Wennerberg J, Biorklund A, et al. Smoking tobacco, oral snuff, and alcohol in the etiology of squamous cell carcinoma of the head and neck: a population-based case-referent study in Sweden. Cancer. 1998;82(7):1367-75.
  [6] Schildt EB, Eriksson M, Hardell L, Magnuson A. Oral snuff, smoking habits and alcohol consumption in relation to oral cancer in a Swedish case-control study. Int J Cancer. 1998;77(3):341-6.
  [7] International Agency for Research on C. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans. Vol. 89. Smokeless Tobacco and Related Nitrosamines. Lyon, France: IARC; 2007.
  [8] Scientific Committee on Emerging and Newly Identified Health Risks (SCENIHR). Health Effects of Smokeless Tobacco Products. Brussels: European Commission, Health & Consumer Protection Directorate-General; 2008.
  [9] Rasmussen F, Tynelius P, Kark M. Importance of smoking habits for longitudinal and age-matched changes in body mass index: a cohort study of Swedish men and women. Prev Med. 2003;37(1):1-9.
  [10] Carpenter CM, Connolly GN, Ayo-Yusuf OA, Wayne GF. Developing smokeless tobacco products for smokers: an examination of tobacco industry documents. Tob Control. 2009;18(1):54-9.
  [11] Zhu SH, Wang JB, Hartman A, Zhuang Y, Gamst A, Gibson JT, et al. Quitting Cigarettes Completely or Switching to Smokeless: Do U.S. Data Replicate the Swedish Results? Tob Control. 2009 Jan 23.

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top
• Published on: 20 February 2009

  Still no response - but there's still time

  • Clive D Bates, NA

  I thought I would revisit this debate some five years on, only to find that the promised response (19 December 2003) has not yet been done.

  None of the facts have changed much - those that wish to intervene to prevent smokers choosing tobacco products that are many times less hazardous still have the upper hand - not in argument or evidence, but in dominant public health approach and (in Europe) in the most...

  Show More

  I thought I would revisit this debate some five years on, only to find that the promised response (19 December 2003) has not yet been done.

  None of the facts have changed much - those that wish to intervene to prevent smokers choosing tobacco products that are many times less hazardous still have the upper hand - not in argument or evidence, but in dominant public health approach and (in Europe) in the most extraordinary and perverse legislation.

  They might not like the to be labelled appropriately with the blunt but accurate epithet "quit or die", but that is the price of taking a position so strongly at odds with evidence and ethics.

  Come on.... it's never too late to put the record straight and defend your work... or admit you were wrong. I suspect this contribution will still be on the internet in 100 years time.

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top
• Published on: 9 April 2004

  The subsequent response?

  • Clive D Bates, Personal

  In their e-letter of 19 December 2003, Tomar et al promised that "Many of the specific comments of Foulds et al. will be addressed in a subsequent response". No response has since been forthcoming.

  Given that Tomar et al's contribution managed to avoid peer review and to appear in the paper edition of Tobacco Control as apparently the last word on the subject, I think it is beholden upon them to say what they...

  Show More

  In their e-letter of 19 December 2003, Tomar et al promised that "Many of the specific comments of Foulds et al. will be addressed in a subsequent response". No response has since been forthcoming.

  Given that Tomar et al's contribution managed to avoid peer review and to appear in the paper edition of Tobacco Control as apparently the last word on the subject, I think it is beholden upon them to say what they accept and don't accept in the criticisms of their article that followed on e-TC. Obviously, if competing interests could also be declared that would be a bonus. I have none.

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top
• Published on: 31 December 2003

  Evidence and argument over smokeless tobacco – another response to Tomar et al

  • Clive Bates, Personal

  I think the most important point to address in Tomar et al’s e-response [1] is their call for more evidence before any change to the status quo (the status quo is a ban on oral tobacco in the EU, and public health disinformation in the US). They say that “neither we nor the IOM Report are ready to accept extant data as sufficient for endorsing smokeless tobacco for harm reduction”. This stance does not reflect the real...

  Show More

  I think the most important point to address in Tomar et al’s e-response [1] is their call for more evidence before any change to the status quo (the status quo is a ban on oral tobacco in the EU, and public health disinformation in the US). They say that “neither we nor the IOM Report are ready to accept extant data as sufficient for endorsing smokeless tobacco for harm reduction”. This stance does not reflect the real-world policy choices and inverts the natural burden of proof.

  First, “endorsing smokeless tobacco for harm reduction” in Europe means unbanning and regulating the product as part of the tobacco market. In the US, it would involve health advocates no longer misleading the public about it – as regrettably the Surgeon General recently did. It does not involve doctors prescribing it, health advocates advocating it, or guest slots for UST at the Washington World Conference. What does ‘endorsing’ actually mean in Europe? In reality, there are some quite concrete decisions in Europe which we must take on the available evidence. These include: continue the oral tobacco ban or lift it? Tell the truth or mislead smokers about it? Whether to regulate oral tobacco, and if so, in what way? These are questions that cannot be ducked on the back of too little evidence – the ‘do nothing’ philosophy is simply a decision in favour of the status quo. This amounts to an active endorsement of the ”quit or die” approach, in which smokers are denied less hazardous alternatives to cigarettes, which themselves remain practically unregulated. There is no supporting evidence for leaving the market to the most dangerous products, and it seems illogical to me.

  Second, let us be clear about the burden of proof. The 'intervention' in Europe is to ban oral tobacco while not banning cigarettes. It is that intervention that requires an evidence base – a point not considered by the IOM. The burden of proof naturally falls on those making and sustaining this intervention (the EU), but three reasons make the demand for supporting evidence more pressing:

  • It is highly irregular to ban a less hazardous variant of a product, thereby deliberately denying users of the more hazardous product the choice of a switch and so preventing an individual risk reduction response. There are no precedents for this, and for that reason alone I would expect good evidence to support such an unorthodox approach to consumer safety.
  • Common sense suggests that it is plausible that addicted smokers would use oral tobacco as an alternative to smoking and to quit, thus reducing harm. Given there is a reasonable hypothesis that smokeless tobacco will reduce harm, there is a commensurate need to show that hypothesis to be flawed if one is to support a ban on oral tobacco without banning smoking tobacco.
  • The evidence from Sweden shows a compelling public health case against banning the product and no-one in their right mind would seek to extend the EU ban to Sweden. Sweden sends a powerful cautionary signal to the rest of the EU about perverse consequences of clumsy interventions. If it is not right to ban it in Sweden, how can we be so confident that it is right to ban it in the rest of the EU? Given the hypothetical opportunity, would American campaigners really ban smokeless tobacco in the US, while leaving cigarettes under the current weak regulatory regime?

  If this was a controlled trial and the intervention (rest of EU, ban on oral tobacco) had consistently worse results than the control (Sweden, no intervention), we would stop the trial and abandon the intervention. To continue with the trial and intervention, the burden of proof would clearly fall on those supporting the intervention to prove that Sweden was not representative or that some other factors are at work in the rest of the EU. Tomar et al haven't any evidence and don't even seem to think they should supply it – perhaps relying on an a priori argument that any ban on any tobacco product represents some sort of progress. Sweden is not a controlled trial of course, but snus is not a medicine and long term changes in market structure do not lend themselves to such trials. Sweden shows us what is possible, and we forego that potential at our peril and at the expense of unfortunate people addicted to smoking and nicotine.

  Sweden almost certainly is different to the rest of Europe – if only by virtue of history and the length of time this market has been established. But that is a reason to expect the benefits in the rest of Europe to be less and slower, not a reason to prevent that change in market structure ever occurring or a basis for believing that the effects would be the opposite in the rest of Europe. The EU justified its ban on the basis of a assumption that smokeless tobacco would unleash a plague of oral cancer and be a major gateway into smoking. Though this was muddled thinking from the outset, the evidence is clear from Sweden, and the opposite appears to be the case - there is no gateway and apparently little risk (and minimal compared to smoking). As the economist John Maynard Keynes famously said: “When the facts change, I change my mind. What do you do?”. In Europe, many of us are changing our minds.

  So, let me restate the question – where is the evidence base that justifies this highly irregular intervention, especially in the face of evidence from Sweden that not banning the product contributes to a public health gain there? How do Tomar et al know their implicit backing for the highly irregular oral tobacco intervention in Europe isn’t simply killing more Europeans? Given what we know about how it works in Sweden, and given that the evidence shows relatively low risk of serious harm from using this product compared to smoking, don’t they think it would be wise to have some evidence to back the case for maintaining this intervention in the rest of the EU? I say they have implicit support for the EU ban because they raise difficult to impossible evidential hurdles to justify a move away from the ban, while offering no evidence to support it.

  Tomar et al worry that people that would otherwise never use tobacco might use oral tobacco if it was unbanned in Europe. The question should be inverted – especially because of what we know in Sweden. How do they know that there are not people that would use snus instead of smoking or quit smoking using snus had it not been banned? Tomar et al pose this question as if it is a ‘red line’ and that it must be shown that no-one who would not otherwise use tobacco would use an oral tobacco product if it was unbanned. What if, more realistically, there was less smoking as smokers switched to snus or quit but some people that did use oral tobacco that would otherwise not have been tobacco users? The balance of risk and benefit (and civil liberties) is on the side of unbanning oral tobacco because oral tobacco is not especially harmful and there is little sign of a gateway to smoking. On the other hand, anyone displacing smoking or quitting with oral tobacco use experiences a considerable reduction in risk. Tomar et al offer no thoughts on this balance, though it is a central concern and has been discussed in Tobacco Control [2].

  Tomar et al also resort to setting impossible evidential hurdles – "If smokeless tobacco has played any overall positive role in Swedish health, what is the relevance of the experience to other countries?" I have discussed above how the burden of proof is really with them, but how could anyone prove to their satisfaction how the tobacco market would respond to the unbanning of snus in Europe without unbanning it and conducting market surveillance? But they want the proof as a pre-condition for unbanning it. The most sensible way forward is to unban the product and conduct market surveillance, adjusting the regulatory regime if needed - and, yes, banning it again if it all goes wrong or the manufacturers behave badly. One could even envisage a 'sunset' clause on the lifting of the ban, requiring it to be reaffirmed in 10 years time.

  Tomar et al use contradictory arguments at different times. For example they raise the (evidence-free) theory that smokeless tobacco may help smokers deflect the pressure to quit arising from smoke-free policies. In their e-response, they worry about "deadly delays in quitting smoking with the support of smokeless tobacco to manage smoking restrictions." But in their initial commentary [3] they attribute the low smoking prevalence in Sweden to the effectiveness of smoke-free policies specifically amongst men despite their high use of snus, suggesting that: "therefore, men would be more likely than women to be impacted by smoke-free workplace regulations". Hmmm… now which is it?

  Let us also address the question of regulation. Yes it is true that a regulatory regime for smokeless tobacco and all tobacco products would be ideal. But the real issue is what should happen in what could be a long (or indefinite) interim period before such a regulatory regime is in place. Should the absence of a comprehensive regulatory regime justify the European ban or American disinformation? I believe not, because even without regulation these products are far less hazardous than cigarettes. The difference between smokeless tobacco products is small compared to the difference between smokeless products and cigarettes. Regulation is highly desirable, but not essential to justify a change from the status quo in the EU. If the EU simply moved to the same position as the US, that would represent progress in the EU because smokers would no longer be denied this option and would at least have the choice available in Sweden, if not the ‘endorsement’ of elements of the public health community.

  However, there is a very good opportunity in unbanning snus to introduce world-leading regulation in the EU. There is a risk that the opportunity will be bungled, either because the public health community remains in denial and slumbers through the opportunity, or because there will be excessive zeal in applying regulation. In the latter case, the danger is that over-strong regulation would leave in place counter-productive asymmetries in regulation between smoking and smokeless tobacco. Regulation shouldn’t be so exacting and one-sided that it prevents the market functioning for public health in the way it has so far succeeded in Sweden. The same argument applies to pharmaceutical nicotine, where the excessive caution of regulators and manufacturers and over-zealous regulation is a barrier to clean nicotine maintenance products and competitors to tobacco and so, paradoxically, works against the wider public health interest [4].

  The EU already has a regulatory regime for marketing – tobacco advertising, sponsorship and promotion is not regarded as legitimate free speech, and is banned. The more challenging regulatory question is whether smokeless companies should have some tightly defined partial exemption from that in order target smokers to switch.

  Finally, please note that this response and my others are only from me, not ‘Bates et al’, and therefore do not include my fellow authors of the printed article.

  [1] Tomar et al, A Reply to Bates et al. and Foulds et al. Tobacco Control e-response, 19 December 2003.

  [2] Kozlowski L. et al. Applying the risk/use equilibrium: use medicinal nicotine now for harm reduction Tob Control 2001;10:201 -203

  [3] Tomar SL, Connolly GN, Wilkenfeld J, Henningfield JE. Declining smoking in Sweden: Is Swedish Match getting the credit for Swedish tobacco control’s efforts? Tobacco Control2003; 12:368-59

  [4] McNeill A, Foulds J, Bates C. Regulation of nicotine replacement therapies (NRT): a critique of current practice. Addiction 2001; 96: 1757-1768.

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top
• Published on: 19 December 2003

  A Reply to Bates et al. and Foulds et al.

  • Scott L. Tomar, Associate Professor
  • Other Contributors:
    • Greg N. Connolly, Judith Wilkenfeld, Jack E. Henningfield

  An important discussion of issues is being missed in a rash of name calling. Let’s back up, recognize our common goals and see if we can discuss issues and skip the personalities. We believe that the letters of Foulds et al and Bates et al badly mangled our comments and took statements out of context. Foulds et al. and Bates et al. obviously feel the same about our article. This issue has precipitated name calling, qu...

  Show More

  An important discussion of issues is being missed in a rash of name calling. Let’s back up, recognize our common goals and see if we can discuss issues and skip the personalities. We believe that the letters of Foulds et al and Bates et al badly mangled our comments and took statements out of context. Foulds et al. and Bates et al. obviously feel the same about our article. This issue has precipitated name calling, questioning of integrity, even rather defamatory slogans used to attack those with differing opinions ranging from “quit or die” to “flat earth believers”. Clearly, we disagree on several aspects of the issue, even while agreeing on others (such as the importance of efforts to reduce death and disease in those who continue to use nicotine without undermining prevention and cessation). This core point of agreement is shared by many of us engaged in this debate. It is unfortunate, and frankly very discouraging, that the differences seem to be resulting in more vitriol than are the common values resulting in constructive dialog. Many of the specific comments of Foulds et al. will be addressed in a subsequent response. Here we would like to briefly address some of the main issues. Our commentary takes issue with the position of Bates et al. and Foulds et al. which calls for a role for smokeless tobacco in tobacco control efforts to reduce smoking. We believe that Bates et al. and Foulds et al. overstate the benefits and generality of the Swedish snus experience and that they understate the risks and areas of uncertainty. Not surprisingly, they imply that we have overstated the risks and understated the benefits in our urge for caution and a regulated playing field. As we note in our commentary:

  “If there is a role for oral tobacco in a comprehensive effort to reduce the death toll from tobacco use, then its manufacture and marketing must be overseen by an agency with comprehensive regulatory authority. A regulatory agency should be open to all strategies that are scientifically based and that will save lives. However, the decision about what role oral tobacco plays in that overall scheme is a decision that can only be made by an agency that has all of the relevant information.”

  The fact that a population of persons using exclusively smokeless tobacco is at overall lower risk of most forms of smoking caused disease is not disputed by us. We do assert that smokeless tobacco has been repeatedly demonstrated to be a deadly addictive product that causes a variety of serious and life-threatening diseases as documented in the Institute of Medicine Report (IOM) (see pages 426-429 and 563-564 for a review of widely accepted oral diseases and attributable cancer). With regard to Swedish smokeless tobacco, at least as marketed in Sweden, we concur with the conclusion of that report on page 167 as follows: “It may be considered that such products could be used as PREPs [potentially reduced exposure products] for persons addicted to nicotine, but these product should undergo testing as PREPs using the guideline and research agenda contained herein.” Neither we nor the IOM Report are ready to accept extant data as sufficient for endorsing smokeless tobacco for harm reduction. Furthermore, we support the general strategies recommended by the IOM Report to develop the data that would enable such and endorsement. We apparently differ from Foulds et al. and Bates et al. in our call for such data. We stand by our position.

  Perhaps the most important issue implicit in our concerns about our colleague’s characterizations of the Swedish experience is their explicit or implicit judgment that it is relevant outside of Sweden. Bates et al. do not recite their analysis of the Swedish experience solely for rhetorical purposes or to support its continuation in Sweden. They cite it in support of its application outside of Sweden to greater Europe and beyond. Similarly, because we were asked to provide a commentary for the Tobacco Control issue as Americans, we responded to this analysis with our concerns about the application of the Swedish experience to the United States and elsewhere. We noted that the U.S. Smokeless Tobacco Company (UST) did not cite the Swedish snus experience in its pleadings before the Federal Trade Commission in order to gratuitously commend Sweden: UST was making a pitch to expand its marketing of smokeless tobacco in the United States, armed with government endorsed harm reduction claims which it argued could then be made for all if its products, including those known to be gateways to tobacco use among youth and those that are among the highest in cancer-causing nitrosamines. In fact, neither the Foulds et al. nor Bates et al. reviews noted that outside of Sweden, even Swedish Match products are higher in cancer-causing nitrosamines than they sell in Sweden.

  Thus, a fundamental question for all tobacco control experts is as follows: If smokeless tobacco has played any overall positive role in Swedish health, what is the relevance of the experience to other countries? Moreover, what questions should tobacco control leaders consider before endorsing smokeless tobacco as a component of comprehensive tobacco control strategies? We have many questions about the benefits attributable to snus in Sweden. Moreover, we are very concerned about the potential for expanded smokeless tobacco marketing in the absence of comprehensive tobacco and nicotine product regulation, particularly if endorsed by public health advocates, to cause more damage to public health than it will to improve public health. The potential risks are numerous and include: Deadly delays in quitting smoking with the support of smokeless tobacco to manage smoking restrictions. Uptake of smokeless tobacco by persons who would not have otherwise used any tobacco product. Substitution of the most popular high nitrosamine smokeless products in the U.S. and as available most everywhere but Sweden for cigarettes as an alternative to complete tobacco cessation. Graduation from smokeless tobacco use to cigarette smoking as has already been well documented in the United States.

  Finally, it is not as if there are no alternatives to reducing smoking prevalence. As demonstrated in California, Florida, Mississippi, and Massachusetts, dramatic reductions in youth smoking initiation and adult smoking prevalence can occur, as compared to surrounding states, without advocating snuff use or trading cigarettes for snuff, but rather on the basis of comprehensive tobacco control efforts.

  Other tobacco control experts will need to carefully examine the data, weigh their concerns, and come to their own conclusions. We remain concerned about the place of smokeless tobacco in tobacco control and will continue to insist that that comprehensive tobacco regulation should be the prerequisite for its further consideration.

  Scott L. Tomar, University of Florida, Division of Public Health Services and Research, Gainesville, Florida, USA

  Greg N. Connolly, Massachusetts Tobacco Control Program, Boston, Massachusetts, USA

  Judith Wilkenfeld, Campaign for Tobacco-Free Kids, Washington, DC, USA

  Jack E. Henningfield, Pinney Associates, Bethesda, Maryland; and Johns Hopkins University School of Medicine, Baltimore, Maryland, USA

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top
• Published on: 14 December 2003

  Another simple 'quit or die' statement

  • Clive Bates, Personal

  Like many others, John Polito [1] misses the point about smokeless tobacco. It is not a health strategy to be widely recommended by doctors, nor is it a medical smoking cessation treatment, nor should it be part of a community-based health programme. It is, or should be, part of a market for nicotine products in which the world will go on allowing the sale of cigarettes - the most hazardous form of nicotine. It should b...

  Show More

  Like many others, John Polito [1] misses the point about smokeless tobacco. It is not a health strategy to be widely recommended by doctors, nor is it a medical smoking cessation treatment, nor should it be part of a community-based health programme. It is, or should be, part of a market for nicotine products in which the world will go on allowing the sale of cigarettes - the most hazardous form of nicotine. It should be a real- world assumption of all those working in public health that cigarettes will continue to be sold legally for the foreseeable future, and that their decline (which I believe is inevitable) will come as people, communities and wider society turn away by choice and through development of new norms. The availability of much less hazardous forms of the product makes sense in that context.

  Polito's method may well work for some people some of the time - though I notice he cites only himself, self-published on his own web site as evidence for this. But even on the generous assumption that there is something in his approach, it cannot be the only possible route for reducing tobacco-related harm. What if people just don't want what he has to offer? What if people can’t or won’t quit using nicotine? What if they need to come off nicotine more gradually? Do you just deny them an alternative to reduce their risks because they are not doing or cannot do what you think they should? Frankly, that is an authoritarian “quit or die” mindset that belongs to an earlier century (not even the last one).

  Tomar and his colleagues [2] clearly do not like to be referred to as advocates of “quit or die” - but that is exactly the choice they offer. However much they dislike it, the label will stick because it is accurate. Polito takes that prescription a step further into "quit my way or die".

  On the subject of disclosure of competing interests, I believe Tobacco Control does require this (though Polito and Tomar et al make none). None of the authors of the discussion on smokeless tobacco (Bates et al [3]) is or was in any way supported by any part of the tobacco industry. We are not wounded by the charge of 'tobacco industry stooges' (as Tomar et al imply we might be) because it is an inaccurate and empty rejoinder to the accurate labelling of Tomar et al as advocates of 'quit or die'. As we explain in our discussion paper, our concern is about reducing cancer, lung and heart disease, and the rest of the consequences of tobacco use as effectively as possible. That's all. We worry that some prominent advocates have uncritically conflated these real goals with the common tactical aim of attacking the tobacco industry in every possible situation. In doing so, they have become confused about priorities and disorientated in the debate on harm reduction.

  The "quit or die" philosophy would be easier to explain if it was straightforward influence-peddling and greed at work on behalf of Big Pharma, which has clear interests in this area. But I think that is the wrong (or only partial) explanation. I suggest the real explanation lies in the realm of ideology, personal views on the relation between the state and the citizen, vanity about anti-tobacco industry credentials, and on the pillars of orthodoxy on which tobacco control currently rests with inadequate critical scrutiny. Talk to people outside the public health field and the idea that governments ban, or advocates lie about, far less hazardous forms of the market-leading product and they find it as absurd as it is unacceptable.

  [1] Polito J. "Another simple issue"?. Electronic response to Tobacco Control, 7 December 2003.

  [2] Tomar SL, Connolly GN, Wilkenfeld J, et al. Declining smoking in Sweden: is Swedish Match getting the credit for Swedish tobacco control’s efforts? Tobacco Control 2003;12: 368-371

  [3] Bates C, Fagerstrom K, Jarvis MJ, et al. European Union policy on smokeless tobacco: a statement in favour of evidence based regulation for public health. Tobacco Control 2003;12 360-367

  Clive Bates [Former Director, ASH UK. Writing in a personal capacity]

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top
• Published on: 7 December 2003

  "Another simple issue"?

  • John R. Polito, Nicotine Cessation Educator

  Watching this first salvo in the battle over whose nicotine is safer and which side eventually makes the big nicotine maintenance bucks, Big Pharm or Big Oral Tobacco, is sad yet understandable? Even for those few without any financial stake in the debate, imagine the natural frustrations born from having turned the wrong research or policy corner and dedicated two decades of your life to having chosen to fight nicotine...

  Show More

  Watching this first salvo in the battle over whose nicotine is safer and which side eventually makes the big nicotine maintenance bucks, Big Pharm or Big Oral Tobacco, is sad yet understandable? Even for those few without any financial stake in the debate, imagine the natural frustrations born from having turned the wrong research or policy corner and dedicated two decades of your life to having chosen to fight nicotine addiction by feeding nicotine addicts more nicotine.

  We've already watched as half-baked pharmaceutical financed science undertook the intentional destruction of the credibility of earth's most productive means of nicotine dependency recovery. We watched as Big Pharm bought the policy door keys and embarked upon a massive campaign to erase earth's most productive tool from cessation literature around the globe.[1] In exchange for what, the Swedish experience or NRT?

  Before throwing in the towel wouldn't a bit of reflection upon where your last campaign took us be in order? You threw out the baby with the bathwater in declaring the life's work, and the daily dependency recovery programs of thousands, to be unscientific. Why grab hold of a shark when drowning, when the water is just five feet deep?

  The March 2003 OTC NRT meta-analysis published here in TC found that only 7% succeeded in remaining smoke-free at six months.[2]

  A November 2003 persistent NRT use study, also published here in TC, suggests that as many as 7% of gum users may still be chewing nicotine at six months.[3] If true, who actually broke free from nicotine while using it? Unlike the one puff lesson that can eventually flow from repeated attempts at abrupt cessation, we've known since 1993 that the only lesson flowing from repeated NRT use is that the odds of relapse increase to nearly 100%,[4] but that too has been kept a secret from those who needed to know.

  Overzealous public health officials must be held accountable for the demise of highly effective community-based abrupt nicotine cessation programs, many of which were achieving 40% midyear nicotine cessation.[5] Imagine a mind so convinced its right that its willing to pervert the term "science based" and use it as a weapon in order to destroy the credibility of superior performance, so that it can claim market share and carry out its own grand insane nicotine weaning experiment.

  Now it's almost as if many of those same so called "experts" who so badly damaged worldwide cessation (some of whom have never personally conducted a single cessation clinic program themselves)have given-up on dependency recovery and embarked upon a massive new social experiment to try and transfer their failure to "safer" forms of delivery.

  Worldwide cessation is in shambles and now we must watch as those who've made the mess argue whether sloshing nicotine-rich tobacco juices around in the mouth or allowing NRT to at last live up to its name - "replacement" not "cessation" - is the answer to all our problems.

  No one here argues with the logic of cleaner delivery but we should all be deeply troubled by the knee-jerk cattle herding tactics and priorities already employed by those now pushing transfer to "safer" delivery. Many teach at institutions whose graduates mold society yet somehow they seem unable to comprehend that, to one degree or another, every graduate of effective community-based recovery programs became recovery teachers themselves. High quality short-term abrupt cessation education, skills development and support programs have now all but vanished, having lost funding and favor after having been declared non science-based, and overrun by those toying with months of weaning.

  I submit that all nicotine dependent humans are entirely capable of quitting. I submit that any attempt by science to put a positive spin on any form of nicotine dependency should cast science in the same mold as any other drug pusher, as more humans, not less, will become dependent. I'm not talking about true harm reduction efforts but marketing spin and easy access that will inevitably snare the curiosity and lives of untold thousands of youth.

  We need only look to Nicorette's current nicotine gum marketing spin to begin to imagine just how out-of-hand a license to push daily maintenance will quickly become.

  Its website asserts that "Once in your brain, nicotine begins working. It stimulates the secretion of neurotransmitters (chemicals in the brain), which appear to enhance awareness and judgment. Nicotine also increases dopamine levels, improving your mood. The substance has also been known to even enhance memory and reduce aggression." ... "Heightened awareness. Enhanced judgment. Better moods. Adrenaline boosts. No wonder cigarette smoking is hard to quit."[5]

  Imagine the tactics that will be employed by the tobacco industry once Pandora's box is fully opened. Just one question, why would you again demand the entire world as your stage when any damage could have been limited to small test communities? It's probably a good thing that TC does not require disclosure of financial interests. Profits or science?

  [1] Polito, Is cold turkey quitting more productive and effective than NRT? WhyQuit, July 2003 - http://whyquit.com/whyquit/A_Cold_Turkey.html

  [2] Hughes, JR, Shiffman, S, et al., A meta-analysis of the efficacy of over-the-counter nicotine replacement . Tobacco Control, March 2003;12:21-27 - http://tc.bmjjournals.com/cgi/content/full/12/1/21

  [3] Shiffman S, Hughes JR, et al, Persistent use of nicotine replacement therapy: an analysis of actual purchase patterns in a population based sample, Tobacco Control 2003 November; 12: 310-316 - http://tc.bmjjournals.com/cgi/content/abstract/12/3/310

  [4] Tonnesen P, et al., Recycling with nicotine patches in smoking cessation. Addiction. 1993 Apr;88(4):533 - http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=8485431&dopt=Abstract

  [5] CDC Sept. 4, 1992 MMWR, Public Health Focus: Effectiveness of Smoking-Control Strategies, United States - http://www.cdc.gov/mmwr/preview/mmwrhtml/00017511.htm

  [6] Nicorette website, How smoking affects your body. http://nicorette.quit.com/nicr_internal/nrt1.asp

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top
• Published on: 6 December 2003

  A reply to Tomar et al's flat earth commentary

  • Clive Bates, Personal

  A reply to Tomar et al’s flat earth commentary

  Foulds et al‘s e-response [1] provides an excellent and scathing critique of the commentary contributed by Tomar et al [2]. Though Foulds et al are far too modest to point this out, it is important that readers understand that their original review [3] is a substantial and careful piece of work, properly edited and peer-reviewed. In contra...

  Show More

  A reply to Tomar et al’s flat earth commentary

  Foulds et al‘s e-response [1] provides an excellent and scathing critique of the commentary contributed by Tomar et al [2]. Though Foulds et al are far too modest to point this out, it is important that readers understand that their original review [3] is a substantial and careful piece of work, properly edited and peer-reviewed. In contrast, while Tomar et al response [2] has the appearance of an evidence review, it is a commentary that has not been peer-reviewed. As Foulds et al [1] very ably demonstrate, their commentary is little more than a catalogue of misinterpretation, misunderstandings and non sequiturs that would not withstand even the most cursory peer review. It is troubling that it stands published in the paper edition of Tobacco Control, appearing to the untrained eye to have equal status to and, worse still, the last word on the excellent work by Foulds et al [3].

  The paper by me and colleagues, Bates et al [4], is not an attempt to review the literature, as this has been done elsewhere and anyone wishing to have an objective appraisal of the evidence can do so. Our attempt was to develop a conceptual and ethical framework for discussing and managing harm reduction, and to suggest what implications the science has for policy in Europe. Kozlowski et al [5] provide a welcome continuation of that effort by discussing what sort of evidence is required to justify action (or a move from the evidence-free support of the status quo) and remind us that real people are involved by invoking the ‘what-to-tell-your-brother’ thought experiment – which they address with a sane and humane argument.

  A disturbing aspect of Tomar et al’s contribution is the way that convoluted argument and a blizzard of mostly meaningless statistics have been deployed to make some sort of case against harm reduction. The style is reminiscent of the ‘keep the controversy alive’ strategies of the tobacco industry – the sort of thing we have seen for years purporting to demonstrate that there is no link between smoking and lung cancer or that passive smoking is harmless. Indeed if distraction was the aim, they will probably be successful. Such obfuscation is a gift to timid law-makers and regulators seeking an excuse for continuing inaction.. The question is, what are they trying to achieve? And who’s interests do they think they are protecting? At least with the tobacco industry, that bit is obvious.

  Equally disturbing is the profound bias against the use of common sense and “reality checking” that Tomar et al display. Many of the findings in Foulds et al [3], and drawn upon in Bates et al [4], are not at all surprising. In case anyone is disorientated by the arguments here are a few pointers back to the real world:

  • It is hardly surprising that use of snus is much less hazardous than smoking tobacco. The latter involves drawing a toxic mixture of volatile organics, oxidants and super-heated reactive particulates into the delicate tissues of the lungs. The former doesn’t – there are no products of combustion or inhalation.

  • It is hardly surprising that snus would be useful in smoking cessation. NRT is agreed to be effective but in several respects snus has superior characteristic as a substitute for smoking. It offers a stronger bolus of nicotine and some of the other sensory aspects of tobacco found in cigarettes but not in NRT. Why wouldn’t it be a better cessation aid for smoking than NRT?

  • It is hardly surprising that Swedish smokers might instinctively know snus is a safer option for continuing nicotine use because of the physical processes involved and therefore switch to it or use it from the outset to control their health risks.

  • It is hardly surprising that the desperate efforts to find a gateway effect have failed. Sweden has the lowest male smoking prevalence in the world but with high levels of snus use. If snus is a gateway then where are all the smokers emerging the other side of the snus gateway? Snus is a gateway out of smoking and an alternative to it for some that would otherwise smoke – that’s why male smoking prevalence is so low.

  • It is hardly surprising that male smoking in Sweden is the lowest in the developed world, and I think the only place where male prevalence is below female. The obvious unique factor is that many men in Sweden use their tobacco and get their nicotine in a different way – through snus. The effort that has gone into denying this simple truth is astonishing. But the corollary of claiming that snus doesn’t contribute to the low male prevalence is that some other reason must be found, and that the use of snus is in effect additional to smoking. It means male smoking is held at a record low level by some other factor that no-one can convincingly identify. In fact it would be very surprising if the widespread use of tobacco in another form didn’t reduce smoking, as these are substitute products.

  • It is hardly surprising that female smoking is also low, even though there is little snus use among women. Admittedly, some of this may be due to Sweden’s tobacco control efforts – but there are no plane-loads of experts from California, Atlanta or Massachusetts trying to discover the secrets of Sweden’s amazing results on per capita programme spend of about one tenth of the top US programmes. There is a less surprising explanation. Doesn’t the tobacco control community see ‘de-normalisation’ as one of the most important approaches. This justifies tireless campaigning for smoke-free environments in the hope that it will de-normalise smoking, remove sensory cues to smoke and provide a temptation-free environment that supports quitting. But this is exactly what snus use does. And furthermore, it takes the denormalisation of smoke into the home. It would be very surprising if snus use among men in Sweden didn’t drag down female smoking through de-normalisation.

  • It is hardly surprising that sane public health advocates should call for snus to be un-banned in Europe and no longer lied about in the US. There are no precedents we can think of for banning a many-times less hazardous variant of the market leader (in this case cigarettes). In fact, if it was attempted in any other area of public life, we would think the perpetrators were acting immorally, and probably illegally. Banning a much less hazardous version of a product that causes a great deal of harm is just plain dumb.

  • It is not surprising that the widespread use of a much less hazardous alternative to the market leader tobacco product reduces harm and has a net public health benefit both through reducing active smoking and passive smoking exposure. What is surprising is that legislators in Europe have decided to deny smokers outside Sweden the option to use products like this to take control of their risks, and have thereby stopped a market in harm reduction products developing. The equivalent American approach appears to be to do this by misleading smokers about smokeless tobacco. Sadly, this credibility-busting tactic seems to have spread to the once-authoritative US Surgeon General in his recent testimony to Congress, thus supporting those that wish to devalue the science base underpinning tobacco control.

  Of course, the fact that something isn’t surprising doesn’t make it so – but common sense is a good starting point for reality-checking very convoluted, tenuous or unconvincing arguments [2] and gives extra confidence when careful assessment of the evidence converges with our understanding of how the real world and real people work [3].

  Finally, I would like to say that I think this is actually quite a simple issue, not the great complex challenge that some suggest. It is beyond doubt that smokeless tobacco products are much less hazardous and can substitute for the market leader, cigarettes. No-one has the right to stop nicotine users taking responsibility for their health by switching to such products if they choose to or cannot or will not give up tobacco or nicotine. Where did Tomar et al, the US Surgeon General, the European Union and others acquire the authority and astonishing high-handedness to sit in their smoke-free citadels and deny smokers these potentially life-saving choices?

  I think a strong regulatory framework is a good idea and that there is an opportunity to achieve it in Europe, if the tobacco control community is clear that its goal is to reduce death and disease, not just have fights with the tobacco industry. But should a new regulatory framework be a pre-requisite for moving from the status quo in which this option is banned outright in the EU? Well, I also think that the status quo is unacceptable - banning a product that is so much less hazardous than the market leader is such an egregious violation of the right of smokers (and potential smokers) to contain the severe risks they face, that it should be reversed even under the current regulatory environment.

  Should American campaigners be honest with the public about relative risks of smokeless and smoking? In my view there is not even an overwheening paternalist case for misleading people about this, let alone a respectful, honest, citizen-focussed public health justification.

  Clive Bates
  [Former director of Action on Smoking and Health UK, writing in a personal capacity]

  [1] Foulds J, Ramstrom L, Fagerstrom K. Effects of smokeless tobacco in Sweden: a reply to Tomar et al. Tobacco Control Online, 5 Dec 2003

  [2] Tomar SL, Connolly GN, Wilkenfeld J, et al. Declining smoking in Sweden: is Swedish Match getting the credit for Swedish tobacco control’s efforts? Tobacco Control 2003;12: 368-371

  [3] Foulds J, Ramstrom L, Burke M, et al. Effect of smokeless tobacco (snus) on smoking and public health in Sweden. Tobacco Control 2003;12: 349-359

  [4] Bates C, Fagerstrom K, Jarvis MJ, et al. European Union policy on smokeless tobacco: a statement in favour of evidence based regulation for public health. Tobacco Control2003;12 360-367

  [5] Kozlowski LT, O’Connor RJ, Quinio Edwards B. Some practical points on harm reduction: what to tell your lawmaker and what to tell your brother about Swedish snus. Tobacco Control 2003;12:372-373

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top
• Published on: 5 December 2003

  Effects of smokeless tobacco in Sweden: a reply to Tomar et al.

  • Jonathan Foulds
  • Other Contributors:
    • Lars Ramstrom, Michael Burke, Karl Fagerstrom

  Dear Editor

  The earth is flat, ABBA couldn’t sing a song, Scotland is going to win the soccer World Cup sometime soon, and snus has played no part in the reduction in smoking prevalence among Swedish men – or so Tomar et al. [1]would have us believe. Of all of these issues not remotely supported by the evidence, the last one is a little more serious in that it may influence tobacco control...

  Show More

  Dear Editor

  The earth is flat, ABBA couldn’t sing a song, Scotland is going to win the soccer World Cup sometime soon, and snus has played no part in the reduction in smoking prevalence among Swedish men – or so Tomar et al. [1]^{would have us believe. Of all of these issues not remotely supported by the evidence, the last one is a little more serious in that it may influence tobacco control policies that will affect the lives (and premature deaths) of millions of people. We therefore feel the need to respond to the plethora of inaccuracies contained in Tomar et al’s commentary on our review of the effects of snus in Sweden.[2]}

  1. Misrepresentation of our review
  Tomar et al’s commentary misrepresents our paper throughout. For example, it states in the second paragraph that a section of Bates et al’s [3]^{article cites only three reports and that our review adds "little additional evidence". We can only assume that Tomar et al. were missing some of the 11 journal pages, 8 figures, 2 tables and 66 references of evidence. Our concern that Tomar and colleagues may have been missing some pages was strengthened when they accused us of ignoring a recently published critical review by Critchley et al. [4] on the health effects of smokeless tobacco. On the contrary, our review not only cited the Critchley article, but quoted its main conclusion verbatim:}

  "Chewing betel quid and tobacco is associated with a substantial risk of oral cancers in India. Most recent studies from the US and Scandanavia are not statistically significant, but moderate positive associations cannot be ruled out due to lack of statistical power."[4] (quoted on p351)

  Similarly, the accusation that "Foulds et al. pay little attention to those other plausible determinants of patterns of tobacco use in Sweden" seems rather strange as we stated plainly that:

  "Both within and outside Sweden, smoking is primarily influenced by factors other than availability of smokeless tobacco (for example, real price of cigarettes, health education, smoke-free air policies, industry marketing etc)." (p357)

  Tomar et al asked, "Could any health professional seriously advocate taking up oral tobacco as a means of preventing cigarette smoking? This seems dangerously close to advocating oral opiod narcotics such as codeine as a means of avoiding heroin use." However, as neither our nor Bates et al’s articles mentioned a word about health professionals advising their patients to use oral tobacco or codeine, (nor do we think that they should for those purposes) we find this to be yet another example of the "straw man" style of argument on which Tomar’s commentary was largely based.

  2. Selective reporting of findings
  Tomar et al accused us of selective reporting of findings. Any reviewer given a word-limit by a journal has to make selective judgements. This is problematic if methodologically strong studies, particularly those with results that conflict with the conclusions of the review, are omitted. We stand by the selection of both studies and results included in the review. For example, Tomar et al cite two reports by Lindstrom et al. [5,6] that we did not mention. These reports were from a single study based in a single city (Malmo) in the far south of Sweden. We did not include these reports because they were located in a small part of Sweden where snus use is markedly less prevalent than the country as a whole, as acknowledged by Lindstrom et al (e.g. the daily snus prevalence of 7% in men reported in the study is about one third of that for the country as a whole). Basing conclusions about snus use in Sweden on a study based exclusively in Malmo is like basing conclusions on smoking and smokeless use in the USA on studies in Utah. Given that Lindstrom et al. [5] concluded that:

  "Snuff consumption may explain a part of the increase in smoking cessation among men as opposed to women in Sweden,"

  we were also confident that this is not an example of omitting studies that don’t agree with the review’s conclusions.

  The other cohort study cited by Tomar et al [7] was flawed because it ignored the effect of the change in wording of the questions on snus in the Living Conditions surveys after 1980-81. The 1980-81 survey simply asked, "Do you use snus?" (thus including both daily and occasional users) whereas the subsequent surveys asked specifically about daily and occasional use.[8] This study therefore mistakenly compared all snus use in 1980-81 with only daily use in 1988-9.

  As a test of who has "selectively reported findings", lets compare our reporting on the prevalence of smoking in young people, and that of Tomar et al. We summarized the data as follows:

  "Looking only at daily smoking prevalence among 16 year olds in Sweden, this has remained remarkably stable at around 11% for boys and 16% for girls for the past 20 years." (p357)

  Tomar et al, on the other hand, state that:

  "Between 1981 and 2001 daily smoking declined more rapidly for 15-16 year old girls (23% to 16%) than boys (13% to 10%), snus use remained rare among girls, and the sex difference in smoking prevalence decreased."

  The full data for daily smoking prevalence by sex for 1981 to 2001 are shown in the figure below.[8] We’ll let the readers decide whether this shows a more rapid decline for girls or a stable pattern (other than normal fluctuations due to sampling differences and factors affecting both sexes equally such as price changes). We’d suggest that the 1981 figure for girls smoking prevalence was an outlier (possibly associated with changes in the wording of the survey questions and definitions of "daily smoking" that took place 1981-3), and that choosing to emphasize it is an example of Tomar et al’s own "selective reporting"

  Prevelance of daily smoking in Sweden by boys and girls ages 15 and 16 with linear regression lines.
  From annual surveys by CAN, Swedish Council for Information on Alcohol and other Drugs

  

  3. Tomar et al’s errors in critical appraisal of health effects of snus
  Tomar et al. accuse us of "misinterpreting the findings from the Lewin et al. study", claiming that we cited only the univariate analyses of results. On the contrary, in addition to their confusion over whether they were quoting relative risks or odds ratios, Tomar et al chose to cite the results of the univariate analyses based on only 9 cases and 10 referents (not controlling for factors such as alcohol use). We concurred with the authors of the original article [9] in choosing to emphasize the results based on a larger number of cases after adjusting for factors such as smoking and alcohol consumption.

  Tomar et al also chastized us for "ignoring" the Institute of Medicine Report.[10] We did in fact cite that report and its findings many times, but perhaps we should have gone further and quoted that report’s conclusion on snus and oral cancer:

  " In Sweden, there is a very high rate of Swedish snuff (snus) use. But, the use of snus in Sweden has generally not been associated with oral cavity cancer (Idris et al, 1998; Kresty et al, 1996; Lewin et al, 1998; Nilsson, 1998; Schildt et al, 1998). Snus is not fermented and so has a much lower level of N-nitrosamines (Nilsson, 1998) and has a lower genotoxic potential (Jansson et al, 1991), which might be related to the lack of increased risk." Institute of Medicine, [10] 2000, p428, para 2.

  Again, it is apparent that our choice not to draw from that report more heavily was not because it contradicted our overall conclusions. In fact the IOM report’s assessment of the snus-cancer relationship is at odds with that of Tomar et al. Perhaps Tomar et al feel that the IOM report was also, "uncritical, misinterpreted the findings", or is "illustrative of the type of simplistic conclusions that might be reached when the nuances of epidemiologic research are not fully appreciated, findings are not fully evaluated." Or perhaps it is Tomar et al. who are out on a limb in their interpretation of the evidence?

  Among the litany of inaccurate criticisms and repetitions of points that we and others have already made (e.g. the need for proper regulation of tobacco and medicinal nicotine products,[2,11] the possible cultural-specificity of Sweden’s experience,[2] etc), Tomar et al made two potentially substantive points: those relating to the pattern of cohort effects in Sweden, and their suggested alternative explanation for the sex-difference in Sweden’s smoking prevalence.

  4. Birth-cohort patterns relating smoking and snus use.
  Tomar et al suggested that the people who initiated snus use in Sweden are not the same people who have quit smoking, and present an analysis of birth cohort effects claiming to demonstrate this. They compared the snus use prevalence among males in different age groups (16-24, 25-34 etc) in 1988/89 with the prevalence of snus use with a different sample (but born in the same years) collected in a survey in 1996/97. They then compared this with the relatively small reduction in cigarette smoking prevalence between 1989 and 2000 among different samples from the same birth cohort (offering no explanation for the change in survey year, to 2000, for the smoking analysis). They imply that the relatively large increase in snus use and the relatively small reduction in prevalence of daily cigarette use within the same age cohort (if not the same sample) shows that the snus use increase and smoking cessation are independent phenomena.

  There are major problems with this analysis, some of which stem from the fact that the changes in tobacco use are not based on the same people over time. It is no big surprise that people tend to take up tobacco use when they are young and try to stop it when they are older (as they do for just about every other kind of substance use). It is perfectly plausible that despite this being the over-riding pattern of snus and cigarette use, a meaningful proportion of smokers in the older age cohorts take up snus (sometimes temporarily) as a way of stopping smoking and are more successful in their quit attempt as a consequence. So long as this number is smaller than the number ceasing snus use (without having smoked), one wouldn’t necessarily observe an increase in snus prevalence in these older age groups. This is particularly likely when a sizable proportion of those taking up snus to replace smoking do so only on a short-term basis, ending up tobacco free by the next survey. The cohort analysis presented by Tomar et al. is therefore irrelevant to the issue of whether men who quit smoking were helped to do so by snus.

   

  A recently published study [12] followed the same cohort of 3244 (75% of the original 4349) participants in the 1980-81 National Survey of Living Conditions through the 1988-9 and 1996-7 surveys. A strength of this data-set is that it follows the same participants over a long time period (16 years), and the weaknesses are that those participating tended to be have slightly lower smoking prevalence than non-participants, some participants were lost to follow-up (201 men and 129 women due to death) and snus use was not reported in the published paper. As shown in the table below, smoking prevalence fell uniformly across the birth cohorts for men (around –14% prevalence) but the reduction in smoking in women was greater in the younger age group (-16% in those aged 18-25 in 1980-1, compared with –3% in those aged 66-73 in 1980-81), and lower overall in women than men (-9 vs –14). This study did not report snus use data in this cohort and we do not believe it is appropriate to guess it based on prevalence in a different sample. However, it is noteworthy that male snus use (occasional plus regular) increased in Sweden from 16.6% to 25.4% from 1980-81 to 1996-7. Thus a higher proportion of male than female smokers have succeeded in quitting smoking in every age group except for 18-40 (most likely due to the extra boost to cessation surrounding pregnancy in women), and a much higher proportion of men than women use snus in every age group. The question of how many men quit smoking by using snus is best addressed by other surveys (discussed below).

  	Men (n=1834)				Women (n=1610)
  Agein 1980-1	% Smoking 1980-1	% Smoking 1988-9	% Smoking 1996-7		% Smoking 1980-1	% Smoking 1988-9	% Smoking 1996-7	Change in Smoking 1980-97 (Men)	Change in Smoking 1980-97 (Women)
  18-25	32	27	18		41	34	25	-14	-16
  26-33	36	29	21		41	38	31	-15	-10
  34-41	40	32	26		36	33	29	-14	-7
  42-49	28	22	15		33	27	23	-13	-10
  50-57	30	23	15		23	21	16	-15	-7
  58-65	27	17	11		17	13	10	-16	-7
  66-73	32	24	18		12	12	9	-14	-3
  Total Population	33	26	19		32	28	23	-14	-9

  Table 1. Prevalence of daily smoking in Sweden in a cohort recruited in 1980-1 and followed up in 1988-9 and 1996-7, by sex and age in 1980-1.[12]

  5. Is the sex difference in smoking prevalence due to fewer women in the smoke-free workplace?
  Tomar et al. proposed a speculative and entirely evidence-free explanation for the differences in smoking prevalence trends for men and women in Sweden: Smokefree workplace regulations have prompted more men than women to quit because a lower proportion of women than men are in full time employment (i.e.men are more likely to be impacted by smokefree workplace regulations).

  One thing that doesn’t seem to fit with that is the data on the older age groups presented in the table above. Those aged 50+ in 1980-1 in that data would mostly have retired from the workforce by 1996-7 and so might be less affected by workplace smoking bans. However, despite that, the sex-difference in cessation is actually stronger in that age group than any other.

  Similarly, at the opposite end of the age spectrum, the sex differences in smoking among school children (shown above) cannot be explained by policies on smoke-free environments as boys and girls in Sweden are subject to the same school environment. Examination of the 2002 Swedish Survey of Living Conditions smoking data [8] by profession also casts doubt on Tomar et al’s proposal. For examples, among adult students (presumably both sexes sharing the same campus environment), 11.6% of men smoke compared with 18.6% of women (no sign of a "gateway effect" here either). Among lower level office staff, smoking prevalence in men fell from 32.6% in 1989 to 18.0% in 2002, whereas the change was only from 29.0% to 26.4% for women. In short, while smoke-free workplace legislation almost certainly triggers smokers to try to quit, any (non snus-related) sex difference in the effects is extremely unlikely to be of sufficient magnitude to account for the relatively large sex differences in smoking patterns that occur even within occupational groups in Sweden. More persuasive is the data from surveys on the use of snus as a smoking cessation aid by Swedish men.

  6. Use of snus as a smoking cessation aid.
  Tomar et al failed to address the evidence [13-16] that a substantial minority (around 30%) of Swedish men who had quit smoking, state that they used snus to help them quit smoking. Rather strangely, Tomar et al tried to brush this important piece of evidence under the carpet by stating that "the majority" of men quit smoking without snus. If this statistic had referred to the proportion of ex-smokers who quit by using some other method (e.g. doctor’s/dentist’s advice, or use of nicotine replacement therapy) we suspect that Tomar et al would have more honestly acknowledged that anything that helps 30% of successful quitters to do so is having a meaningful and important role in smoking cessation. 

  7. This is about Sweden, not the USA.
  Tomar et al belatedly suggest that the rhetoric be toned down and that their differing focus relates to differences between national regulations, companies and products (presumably referring to differences between the USA and Sweden). However, these national differences are not directly relevant here because (for once) these papers were NOT about the U.S. These papers were published in an international journal and focused very specifically on the evidence to date in Sweden,[2] and the potential implications for European policy.[3]

  8. Both snus and Swedish tobacco control deserve some of the credit

  Finally, we’d like to address another point made by Tomar et al – namely that Sweden quite rightly deserves credit for its tobacco control efforts. In addition to inventing nicotine replacement therapy, the Swedes have implemented a number of positive tobacco control interventions and the reduction in smoking prevalence among women (which has very little to do with snus use) has been impressive. Sweden’s tobacco control movement has had a particularly strong component designed to reduce tobacco use among women, thanks in no small part to the efforts of Margaretha Haglund, who has also been the President of the International Network of Women Against Tobacco (INWAT) for many years. However, it is in that context of strong tobacco control measures, often targeting women, that the larger reduction in smoking prevalence in Swedish men is all the more remarkable. To deny that snus has played some part in that success (which is the issue we were asked to review) is to deny the weight of the evidence.

   Acknowledgements

  Jonathan Foulds and Michael Burke are primarily funded by New Jersey Department of Health and Senior Services. Jonathan Foulds, Karl Fagerstrom , and Lars Ramstrom have worked as consultants and received honoraria from pharmaceutical companies involved in production of tobacco dependence treatment medications. Lars Ramstrom has also received project support from the Swedish National Institute of Public Health and salary from short term employment with WHO. None of the authors has accepted any funding from the tobacco industry.

  Jonathan Foulds
  University of Medicine and Dentistry of New Jersey- School of Public health, Tobacco Dependence Program, New Brunswick, USA
  
  Lars Ramstrom
  <sup>Institute for Tobacco Studies, Stockholm, Sweden
  
  Michael Burke
  University of Medicine and Dentistry of New Jersey- School of Public health, Tobacco Dependence Program, New Brunswick, USA
  
  Karl Fagerstrom
  Fagerstrom Consulting and The Smokers Information Center, Helsingborg, Sweden</sup>

   

  References

  1. Tomar SL, Connolly GN, Wilkenfeld J, Henningfield JE. Declining smoking in Sweden: Is Swedish Match getting the credit for Swedish tobacco control’s efforts? Tob Control 2003; 12:368-59

  2. Foulds J, Ramstrom L, Burke M, Fagerstrom K. The effect of smokeless tobacco (snus) on smoking and public health in Sweden. Tob Control 2003; 12:349-59

  3. Bates C, Fagerstrom K, Jarvis MJ, Kunze M, McNeill A, Ramstrom L. European Union policy on smokeless tobacco: A statement in favor of evidence-based regulation for public health. Tob Control 2003; 12:360-7

  4. Critchley JA, Unal B. Health effects associated with smokeless tobacco: a systematic review. Thorax 2003; 58:435-443.

  5. Lindstrom M, Isacsson SO. Smoking cessation among daily smokers, aged 45-69 years: a longitudinal study in Malmo, Sweden. Addiction 2002;97 (2):205-15.

  6. Lindstrom M, Isacsson SO. Long term and transitional intermittent smokers: a longitudinal study. Tob Control 2002;11:61-7.

  7. Tillgren P, Haglund BJ, Lundberg M, Romelsjo A. The sociodemographic pattern of tobacco cessation in the 1980s: results from a panel study of living condition surveys in Sweden. J Epidemiol Cummunity Health 1996;50:625-630.

  8. Wicklin B. Nordic Tobacco Statistics. Statistikbyran VECA. Vallingby, Sweden, 2002. www.statveca.com.

  9. Lewin F, Norell SE, Johansson H, Gustavsson P, Wennerberg J. Smoking Tobacco, oral snuff, and alcohol in the etiology of squamous cell carcinoma of the head and neck. Cancer 1998; 82:1367-1374.

  10. Stratton K, Shetty P, Wallace R, Bondurant S, eds. Clearing the smoke: assessing the science base for tobacco harm reduction. Institute of Medicine, National Academy of Sciences. Washington DC: National Academy Press, 2001.

  11. McNeill A, Foulds J, Bates C. Regulation of nicotine replacement therapies (NRT): a critique of current practice. Addiction 2001; 96: 1757-1768.

  12. Rasmussen F, Tynelius P, Kark M. Importance of smoking habits for longitudinal and age-matched changes in body mass index: a cohort study of Swedish men and women. Preventive Medicine 2003; 37:1-9

  13. Ramstrom L. Patterns of use of Swedish smoke-free tobacco, snus: A gate leading to smoking, or a way to give it up? Abstract from 4^th SRNT European Conference, Santandar, October 2002. Nicotine & Tobacco Research 2003;5: 268.

  14. Rodu, B. Stegmayr, B. Nasic, S., and Asplund, K. Inmpact of smokeless tobacco use on smoking in northern Sweden. Journal of Internal Medicine 2002;252: 398-404.

  15. Rodu, B, Stegmayr B, Nasic S, Cole P, Asplund K. Evolving patterns of tobacco use in northern Sweden. Journal of Internal Medicine 2003;253:660-665.

  16. Gilljam H, Galanti MR. Role of snus (oral moist snuff) in smoking cessation and smoking reduction among Swedish men. Addiction 2003; 98:1183-9.

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top