Berry et al (1) report an analysis of two waves of the Population Assessment of Tobacco and Health (PATH) study focused on the association between the initiation of e-cigarette use by Wave 2 and cigarette abstinence/reduction assessed at Wave 2. They conclude that daily e-cigarette use is associated with both cigarette abstinence and reduced consumption among continuing smokers. While this addresses an important question, we argue that such analyses should be adjusted for the reason e-cigarettes are being used.
From Wave 1 of PATH (2), we know that ~75% of smokers agreed that e-cigarettes were useful to help people quit. However, ~80% agreed that e-cigarettes allowed someone to replace a cigarette where smoking was prohibited. From the first reason, we can hypothesize that e-cigarette use might be associated with cigarette abstinence/reduction. However, from the second reason, we can also hypothesize that e-cigarettes would be associated with neither cigarette abstinence nor reduction. The recent National Academies report (3) recommended that any assessment of the role of e-cigarettes in cigarette cessation/reduction should focus on smokers who used e-cigarettes to help them quit.
PATH Wave 2 data does include information on whether smokers tried to quit in the previous year, as well as whether they used e-cigarettes to aid the last quit attempt. Previous research (4) has shown that over half of the smoking population will not ha...
Berry et al (1) report an analysis of two waves of the Population Assessment of Tobacco and Health (PATH) study focused on the association between the initiation of e-cigarette use by Wave 2 and cigarette abstinence/reduction assessed at Wave 2. They conclude that daily e-cigarette use is associated with both cigarette abstinence and reduced consumption among continuing smokers. While this addresses an important question, we argue that such analyses should be adjusted for the reason e-cigarettes are being used.
From Wave 1 of PATH (2), we know that ~75% of smokers agreed that e-cigarettes were useful to help people quit. However, ~80% agreed that e-cigarettes allowed someone to replace a cigarette where smoking was prohibited. From the first reason, we can hypothesize that e-cigarette use might be associated with cigarette abstinence/reduction. However, from the second reason, we can also hypothesize that e-cigarettes would be associated with neither cigarette abstinence nor reduction. The recent National Academies report (3) recommended that any assessment of the role of e-cigarettes in cigarette cessation/reduction should focus on smokers who used e-cigarettes to help them quit.
PATH Wave 2 data does include information on whether smokers tried to quit in the previous year, as well as whether they used e-cigarettes to aid the last quit attempt. Previous research (4) has shown that over half of the smoking population will not have tried to quit in the previous year. By including these non-attempters, who by definition cannot have quit, Berry et al (1) may have introduced an important bias toward finding a higher daily e-cigarette effect on abstinence. We expect that daily e-cigarette use at Wave 2 will be much higher among those who made a recent quit attempt than in those who did not.
In their supplement tables (TableS3), they include an analysis of those who made a quit attempt prior to Wave 1(rather than between Waves 1 and 2). Using this analysis, there is a drastic reduction in the effect size amplitude and in the absolute number of involved smokers. We would expect similar, or even larger, reduction in effect estimates were they to have restricted their analysis to those who made a quit attempt in the year prior to Wave 2 and included reason for using e-cigarettes.
In order to know the effect of e-cigarettes on cessation, those who used an e-cigarette to help them to quit should be contrasted with comparable non-users: those who used other aids to quit as well as to those who quit unaided. There are numerous important potential confounders for these comparisons as it is well known that those who are least likely to be successful in the quit attempt are the most likely to use an aid. (5) For unbiased analyses, the exposure of interest needs to be isolated and covariate balance achieved between exposed and unexposed. There is a role for methodological approaches that help achieve covariate balance, such as propensity score matching, in deciding whether e –cigarettes improve population smoking cessation.
References:
1. Berry KM, Reynolds LM, Collins JM, Siegel MB, Fetterman JL, Hamburg NM, Bhatnagar A, Benjamin EJ, Stokes A. E-cigarette initiation and associated changes in smoking cessation and reduction: the Population Assessment of Tobacco and Health Study, 2013-2015.Tob Control. 2018 Mar 24. pii: tobaccocontrol-2017-054108. doi: 10.1136/tobaccocontrol-2017-054108.
2. Coleman BN, Rostron B, Johnson SE, Ambrose BK, Pearson J, Stanton CA, et al. Electronic cigarette use among US adults in the Population Assessment of Tobacco and Health (PATH) Study, 2013–2014. Tobacco Control. 2017. doi: 10.1136/tobaccocontrol-2016-053462.
3. National Academies of Sciences Engineering, and Medicine,. Public Health Consequences of E-Cigarettes. Washington, DC: Health and Medicine Division, National Academies of Sciences, Engineering, and Medicine, 2018
4. Zhu S-H, Lee M, Zhuang Y, Gamst A, Wolfson T. Interventions to increase smoking cessation at the population level: How much progress has been made in the last two decades? Tob Control. 2012;212:110–118
5. Leas EC, Pierce JP, Benmarhnia T, White MM, Noble ML, Trinidad DR, Strong DR. Effectiveness of Pharmaceutical Smoking Cessation Aids in a Nationally Representative Cohort of American Smokers. J Natl Cancer Inst. 2017 Dec 21. doi: 10.1093/jnci/djx240
Feliu et al’s conclusion “in the European Union countries with the higher scores in the Tobacco Control Scale, which indicates higher tobacco control efforts, have lower prevalence of smokers, higher quit ratios and higher relative decreases in their prevalence rates of smokers.” deserved comment.
First, it seems a tautology. Tobacco control policies are robustly evidence based. Accordingly, more efforts, less smokers.
Second, a PubMed search with “"tobacco control scale" only retrieved 27 articles since 2006 and no validation published yet. Obviously, the Scale poorly correlated with smoking rate: r2 being .58 in 2002/3, .15 in 2006/7 and .06 in 2010/11.(From table 3 in 2; n= 11 European countries).
Third, why make simple stuff complex? This surrogate is complex to calculate and its items are subjective because issuing a decree is useless if no implementation were enforced. In contrast, the smoking rate and its evolution are simple and reliable! How France can be ranked 4th among 28 countries with a 57/100 score (1) while smoking prevalence has been plateauing for so long at more than 30%? In France, from 2004 to 2017 no relevant increase in tobacco taxes, no implementation of the legal smoking ban in cafés or of the ban of sale to minors despite sting operations by NGO showing evidence of serious breaches.(3)
Fourth, claiming “the European Union should continue implementing comprehensive tobacco control pol...
Feliu et al’s conclusion “in the European Union countries with the higher scores in the Tobacco Control Scale, which indicates higher tobacco control efforts, have lower prevalence of smokers, higher quit ratios and higher relative decreases in their prevalence rates of smokers.” deserved comment.
First, it seems a tautology. Tobacco control policies are robustly evidence based. Accordingly, more efforts, less smokers.
Second, a PubMed search with “"tobacco control scale" only retrieved 27 articles since 2006 and no validation published yet. Obviously, the Scale poorly correlated with smoking rate: r2 being .58 in 2002/3, .15 in 2006/7 and .06 in 2010/11.(From table 3 in 2; n= 11 European countries).
Third, why make simple stuff complex? This surrogate is complex to calculate and its items are subjective because issuing a decree is useless if no implementation were enforced. In contrast, the smoking rate and its evolution are simple and reliable! How France can be ranked 4th among 28 countries with a 57/100 score (1) while smoking prevalence has been plateauing for so long at more than 30%? In France, from 2004 to 2017 no relevant increase in tobacco taxes, no implementation of the legal smoking ban in cafés or of the ban of sale to minors despite sting operations by NGO showing evidence of serious breaches.(3)
Fourth, claiming “the European Union should continue implementing comprehensive tobacco control policies in Europe.”(1) is optimistic, at best. The European Union is the chimney of rich countries: smoking prevalence in Italy, France and Germany is almost twice that in Australia and 1.5 fold that in the US. Almost no tobacco control in the European Union but Finland! The Scale is a smokescreen for tricky politicians cherry picking the weakest measures without even providing tools for implementation or monitoring.
Last, I am not aware that a critical assessment of the Eurobarometer method is available, and the limitations of such surveys cannot be overlooked. This deserves scrutiny as other data from the European Union on such a topic are a cause for concern: eg. the European School Project on Alcohol and other Drugs estimates smoking prevalence only on a declarative basis, roughly 10% of the data are missing despite only recruiting those attending school and only 80 % of the students said that they thought that their classmates had answered the questions honestly.(4)
1 Feliu A, Filippidis FT, Joossens L et al. Impact of tobacco control policies on smoking prevalence and quit ratios in 27 European Union countries from 2006 to 2014. Online Feb 22.
2 Kuipers MA, Monshouwer K, van Laar M, Kunst AE. Tobacco control and socioeconomic inequalities in adolescent smoking in Europe. Am J Prev Med 2015;49:e64-e72.
3 Braillon A, Mereau AS, Dubois G. [Tobacco control in France: effects of public policy on mortality]. Presse Med 2012;41:679-81.
4 Hibell B, Molinaro S, Siciliano V, Kraus L. The 2013 ESPAD validity study. European Monitoring Centre for Drugs and Drug Addiction. Publications Office of the European Union. Luxembourg. 2015.
NOT PEER REVIEWED
This paper’s core findings are quite helpful: (1) Tax/price increases for non-cigarette tobacco products can effectively reduce their use; and (2) Tax/price increases for non-cigarette tobacco products could prompt some users to increase their cigarette smoking if comparable tax/price increases for cigarettes are not done at the same time. But the paper’s related analysis is incomplete, producing misleading conclusions, largely because the paper focuses on cigarettes versus non-cigarette tobacco products without also considering the more important distinction for health-directed tobacco tax strategies between smoked tobacco products and non-combustible tobacco products.
In its abstract, the paper concludes that the “positive substitutability between cigarettes and non-cigarette tobacco products suggest that tax and price increases need to be simultaneous and comparable across all tobacco products.” But the paper does not appear to consider that the only substitutions that could significantly increase public health harms would be if the tax increases prompted some non-combusted tobacco product users to move to more-harmful smoking or prompted some smokers who would otherwise do so not to move to less-harmful non-combusted tobacco products. As a result, the paper fails to acknowledge that significant tax/price increases for only combusted tobacco products would not prompt any harm-increasing substitution and would directly secure desirable...
NOT PEER REVIEWED
This paper’s core findings are quite helpful: (1) Tax/price increases for non-cigarette tobacco products can effectively reduce their use; and (2) Tax/price increases for non-cigarette tobacco products could prompt some users to increase their cigarette smoking if comparable tax/price increases for cigarettes are not done at the same time. But the paper’s related analysis is incomplete, producing misleading conclusions, largely because the paper focuses on cigarettes versus non-cigarette tobacco products without also considering the more important distinction for health-directed tobacco tax strategies between smoked tobacco products and non-combustible tobacco products.
In its abstract, the paper concludes that the “positive substitutability between cigarettes and non-cigarette tobacco products suggest that tax and price increases need to be simultaneous and comparable across all tobacco products.” But the paper does not appear to consider that the only substitutions that could significantly increase public health harms would be if the tax increases prompted some non-combusted tobacco product users to move to more-harmful smoking or prompted some smokers who would otherwise do so not to move to less-harmful non-combusted tobacco products. As a result, the paper fails to acknowledge that significant tax/price increases for only combusted tobacco products would not prompt any harm-increasing substitution and would directly secure desirable public health gains by directly reducing smoking. It would be foolish for the public health community to reject or not support such a tax increase because it did not also increase taxes on non-combustible tobacco products. But the paper suggests otherwise.
A tobacco tax increase on cigarettes and all other combustibles would secure even larger public health gains if it also used different sized tax increases on different combustible products in order to raise the prices of all smoked tobacco products to the same level as cigarettes (typically the most highly taxed tobacco products), thereby making it more likely that cigarette smokers would respond to the tax increases by cutting back their smoking or quitting altogether, rather than by switching to less-taxed and less-expensive smoked tobacco products, such as little cigars or RYO. But the paper does not consider this option, and its support for comparable tax increases for all tobacco products argues against it.
A tax increase for combusted tobacco products could secure even larger public health gains if it also increased non-combustible tobacco product taxes and prices. But only if the increases to the combusted and non-combusted tobacco product taxes did not make the non-combustibles significantly more expensive relative to cigarettes and other smoked tobacco products, either relatively or in real terms, that they dampened moves from smoking to e-cigarettes or other non-combustibles or prompted some non-combustible users to increase their smoking. But that, also, is not discussed in the paper.
Instead, the main text of the paper talks about taxes to produce equal price increases for each type of tobacco product and concludes by offering the overly broad and potentially misleading recommendation that countries with tobacco-diverse markets “should raise taxes on non-cigarette tobacco products to prevent premature death.”
Perhaps all the additional policy analysis suggested here is too much to expect from a study with the rather simple stated objective of just systematically reviewing the price elasticity of demand of non-cigarette tobacco products. But the paper goes well beyond providing that information and also encourages tax increases for non-cigarette tobacco products, either on their own or along with comparable increases for cigarettes. Given that the paper has decided to make policy recommendations, it should do so more thoughtfully, with more explanation and more detailed guidance.
NOT PEER REVIEWED
The paper by Filippidis et al [1] provides data re-confirming the well-known fact that most ex-smokers attempt to quit without using any form of assistance, whether pharmaceutical, professional or via e-cigarettes. Moreover, the proportion of ex-smokers trying to quit unaided increased substantially in Europe between 2012-17 (ex-smokers using no assistance increased from 73.9% to 80.7%), a period where e-cigarette use accelerated in some nations.
Regrettably however, this study does not permit any comparison of success rates by method, as no data are reported on which method of cessation (assisted v unassisted) was used by ex-smokers on their last, final (and so successful) quit attempt.
The authors report that those “who successfully quit reported much lower use of cessation assistance compared with smokers who had tried to quit without success” and suggest that this might reflect indication bias, whereby those who find it harder to quit self-select to use assistance, leaving the low hanging fruit of non- or less addicted smokers to fall off the smoking tree using their own determination.
While this will be true for some, there are many former heavy smokers who quit without assistance. This argument also borrows assumptions from the discredited hardening hypothesis [2], which holds, in the face of evidence to the contrary, that as smoking prevalence falls the concentration of hardened, more deeply addicted smokers increase...
NOT PEER REVIEWED
The paper by Filippidis et al [1] provides data re-confirming the well-known fact that most ex-smokers attempt to quit without using any form of assistance, whether pharmaceutical, professional or via e-cigarettes. Moreover, the proportion of ex-smokers trying to quit unaided increased substantially in Europe between 2012-17 (ex-smokers using no assistance increased from 73.9% to 80.7%), a period where e-cigarette use accelerated in some nations.
Regrettably however, this study does not permit any comparison of success rates by method, as no data are reported on which method of cessation (assisted v unassisted) was used by ex-smokers on their last, final (and so successful) quit attempt.
The authors report that those “who successfully quit reported much lower use of cessation assistance compared with smokers who had tried to quit without success” and suggest that this might reflect indication bias, whereby those who find it harder to quit self-select to use assistance, leaving the low hanging fruit of non- or less addicted smokers to fall off the smoking tree using their own determination.
While this will be true for some, there are many former heavy smokers who quit without assistance. This argument also borrows assumptions from the discredited hardening hypothesis [2], which holds, in the face of evidence to the contrary, that as smoking prevalence falls the concentration of hardened, more deeply addicted smokers increases.
If our concern is (as it should be) to better understand the means of quitting that produce the largest net volume of ex-smokers across whole populations, studying the methods these former smokers used when they succeeded is critical. Yet the “inverse impact law of smoking cessation [3] shows that unassisted cessation, which undisputedly delivers more ex-smokers than any other method, is hugely neglected in smoking cessation research [4].
It is almost as if researchers want to turn away from learning more about the most successful route that has always delivered the largest number of successful quits. [5]
Rather than seeing the increase in unassisted quitting as something to be highlighted as a positive, motivating celebration of agency that could be megaphoned in campaigns to smokers imbued with pessimistic messages about how hard quitting is going to be, the authors conclude that their findings ”highlight the need for approaches to ensure that smokers get support”.
In 40 years of tobacco control, I cannot ever recall attending a meeting or conference on cessation where those whose living depended on them selling smoking cessation aids or providing professional cessation services did not reach similar conclusions. Yet 40 years on, the same cracked record is being played: we need to convince more smokers that they should not try foolishly to quit alone and that they need our help!
Analysis at the level of the success of quit “attempts” often shows that head-to-head, unassisted cessation attempts are less successful than those using assistance. But many so-called cessation attempts are empty gestures akin to those who attempt to get fit by buying an exercise bike, use it once or twice and then consign it to the corner. West and Sohal’s work on catastrophe theory noted that many who were not planning to quit at time 1, had succeeded at time 2. They suggested that “smokers have varying levels of motivational “tension” to stop and then “triggers” in the environment result in a switch in motivational state. If that switch involves immediate renunciation of cigarettes, this can signal a more complete transformation than if it involves a plan to quit at some future point.” [6]
The importance of continually stimulating the motivational tension to stop smoking and providing both informational and policy triggers for quitting cannot be over-emphasised.
It is long overdue that we gave far more attention to the net contribution of unassisted cessation at the population level. [7 ] Many smokers have little interest in being helped to quit. In this, they are very aware of many friends and acquaintances who quit alone when they were sufficiently motivated to do so. Over 40 years of professional hand-wringing, research and campaigning about how to undermine unaided quitting and sell more drugs and clinic appointments have thankfully done little to erode this.
References
1. Filippidis FT, Mons U, Jiminez-Ruiz C, Vardavas CI. Changes in smoking cessation assistance in the European Union between 2012 and 2017: pharmacotherapy versus counselling versus e-cigarettes. Tobacco Control http://dx.doi.org/10.1136/tobaccocontrol-2017-054117
2. Cohen JE, McDonald PW, Selby P. Softening up on the hardening hypothesis. Tobacco Control ttp://dx.doi.org/10.1136/tobaccocontrol-2011-050381
3. Chapman S. The Inverse Impact Law of Smoking Cessation. Lancet 2009; 373(9665):701-3.
4. Chapman S, Mackenzie R. The global research neglect of unassisted smoking cessation: causes and consequences. PLoS Medicine 2010; 7(2): e1000216. doi:10.1371/journal.pmed.1000216.
5. Smith A, Chapman S. Quitting unassisted: the 50 year neglect of a major health phenomenon. JAMA 2014;311(2):137-138. doi:10.1001/jama.2013.282618.
3. West R, Sohal T. “Catastrophic” pathways to smoking cessation: findings from national survey. BMJ. 2006 Feb 25; 332(7539): 458–460.
doi: 10.1136/bmj.38723.573866.AE
4. Smith A, Carter SM, Chapman S, Dunlop S, Freeman B. Why do smokers try to quit without medication or counseling? A qualitative study with ex-smokers. BMJ Open 5:e007301 doi:10.1136/bmjopen-2014-007301
There is also very clear evidence that tobacco industry interference is either delaying or dumbing down implementation of each of the MPOWER policies particularly in LMICs. The TC vaccine is a good concept but the framework needs to include monitoring, exposing and countering industry tactics.
I am grateful to Bashash et al. for raising some important methodological and policy-related issues. Responding to their specific points:
(1) Very high formaldehyde concentrations may arise in aerosols when atomisers generate excessive heat[1]. Under these circumstances recommended safety limits for formaldehyde may indeed be exceeded and this compound contributes most to the cancer potency summation.
(2) Goodson et al. [2] provide a framework for assessing whether low dose compounds that are not necessarily individual carcinogens may become involved in carcinogenesis when acting in concert. Although discussed under "Strengths and limitations" synergystic phenomena were not accommodated in the cancer potency model as it is not yet possible to predict the mechanism and magnitude of such interactions in tobacco or e-cigarette aerosols. Under the Goodson et al. model adverse effects reflect adventitious synergystic combinations. These may be statistically more likely in tobacco smoke where the number of different compounds greatly exceeds those of simpler aerosols, however this effect is expected to be minor compared with the exceptionally high carcinogenic potencies of some well-established carcinogens in tobacco smoke.
(3) Lifetime cancer risk is linearly dependent on the daily volume of vapour inhaled (equation 7) and the effect on risk of increased consumption after switching to heat not burn (HnB) products is directly related to the chang...
I am grateful to Bashash et al. for raising some important methodological and policy-related issues. Responding to their specific points:
(1) Very high formaldehyde concentrations may arise in aerosols when atomisers generate excessive heat[1]. Under these circumstances recommended safety limits for formaldehyde may indeed be exceeded and this compound contributes most to the cancer potency summation.
(2) Goodson et al. [2] provide a framework for assessing whether low dose compounds that are not necessarily individual carcinogens may become involved in carcinogenesis when acting in concert. Although discussed under "Strengths and limitations" synergystic phenomena were not accommodated in the cancer potency model as it is not yet possible to predict the mechanism and magnitude of such interactions in tobacco or e-cigarette aerosols. Under the Goodson et al. model adverse effects reflect adventitious synergystic combinations. These may be statistically more likely in tobacco smoke where the number of different compounds greatly exceeds those of simpler aerosols, however this effect is expected to be minor compared with the exceptionally high carcinogenic potencies of some well-established carcinogens in tobacco smoke.
(3) Lifetime cancer risk is linearly dependent on the daily volume of vapour inhaled (equation 7) and the effect on risk of increased consumption after switching to heat not burn (HnB) products is directly related to the change in number of sticks. Any implication that HnB products pose acceptable risks was certainly not intended: the adjective 'safe' should never be used to encourage the use of HnB products, nor indeed any VNP with the possible exception of medicinal devices.
(4) Neither emissions nor biomarkers are suitable for accurate determination of absolute risks; long-term clinical evidence is needed for reliable estimates. In the meantime the emissions data used in the paper suggest that the relative lifetime cancer risk from HnB devices may be up to 50 times less than that of combustible cigarettes. The risk is still large - about 10,000 times greater than inhaling an equivalent volume of ambient air. In contrast the modelling suggests that many e-cigarettes pose cancer risks within a factor of 10 of a nicotine inhaler that has been approved for use by the public. E-cigarettes, if used as the manufacturer intended, appear to offer significantly lower risk alternatives to HnB for the smoker intent on using these novel products as aids to quitting to avoid cancer. The relative risks of other diseases are yet to be quantified.
References
[1] Farsalinos KE, Voudris V, Spyrou A, Poulas K. E-cigarettes emit very high formaldehyde levels only in conditions that are aversive to users: A replication study under verified realistic use conditions. Food and Chemical Toxicology. 2017;109:90-4.
[2] Goodson WH, Lowe L, Carpenter DO, Gilbertson M, A. MA, Lopez de Cerain Salsamendi A, et al. Assessing the carcinogenic potential of low-dose exposures to chemical mixtures in the environment: the challenge ahead. Carcinogenesis. 2015;36 (Suppl 1):S254-S96.
NOT PEER REVIEWED The Jawad et al systematic review and meta-analysis examining price effects for non-cigarette tobacco and nicotine products appears methodologically sound and was a registered analysis. It provides information that could be used productively by advocates and policymakers seeking to reduce harm. The cross-elasticities reported in this paper can be used to the advantage of public health by increasing the impact of policies that seek to drive down smoking.
However, this work does not take into account the fact that not all tobacco and nicotine products cause the same level of health harms as combustible cigarettes. The paper examines own- and cross-price elasticity across a wide array of products – from combustible tobacco products such as kreteks and little cigars to nicotine-only products such as e-cigarettes and nicotine patches – and then discusses consumption patterns in terms of an undifferentiated aggregate of nicotine use. Jawad and colleagues do not consider the health implications of policies to move nicotine users from more-harmful to less-harmful means of administration (see, for example, Chaloupka, Warner and Sweanor, 2015, recommending differential taxation according to differential risk).
From a public health perspective, any analysis of nicotine-use patterns should consider differential harm levels. A focus on nicotine use as the sole outcome variable can be seriously misleading and detrimental to the goal of reducing smoking....
NOT PEER REVIEWED The Jawad et al systematic review and meta-analysis examining price effects for non-cigarette tobacco and nicotine products appears methodologically sound and was a registered analysis. It provides information that could be used productively by advocates and policymakers seeking to reduce harm. The cross-elasticities reported in this paper can be used to the advantage of public health by increasing the impact of policies that seek to drive down smoking.
However, this work does not take into account the fact that not all tobacco and nicotine products cause the same level of health harms as combustible cigarettes. The paper examines own- and cross-price elasticity across a wide array of products – from combustible tobacco products such as kreteks and little cigars to nicotine-only products such as e-cigarettes and nicotine patches – and then discusses consumption patterns in terms of an undifferentiated aggregate of nicotine use. Jawad and colleagues do not consider the health implications of policies to move nicotine users from more-harmful to less-harmful means of administration (see, for example, Chaloupka, Warner and Sweanor, 2015, recommending differential taxation according to differential risk).
From a public health perspective, any analysis of nicotine-use patterns should consider differential harm levels. A focus on nicotine use as the sole outcome variable can be seriously misleading and detrimental to the goal of reducing smoking.
Cited Reference
Chaloupka, FJ, Sweanor D, Warner KE. Differential taxes for differential risks—toward reduced harm from nicotine-yielding products. N Engl J Med 2015 Aug 13;373(7): 594-7.
NOT PEER REVIEWED
We thank Dr. Jarvis for his appreciation of our historical scholarship but disagree that our conclusion, “the promotion of tobacco harm reduction may serve the interests of tobacco companies more effectively than the public,” is an attack.
Our paper is about how policy affects ideas and vice versa. The ideas guiding the product modification program led to bad outcomes. That these ideas have been reanimated merits critical assessment. Voluntary agreements led to industry influence over the ISCSH’s recommendations, which in turn undermined public health. We point out that some of the same premises that led the ISCSH astray are popular again. Jarvis claims that current UK harm reduction policy has nothing to do with the product modification program, and everything to do with the influence of the late Michael Russell. Russell’s impressive scholarship – and oft-quoted statement, “people smoke for the nicotine, but die from the tar” – is indeed hugely influential among proponents of tobacco harm reduction. Jarvis posits that Russell’s work serves as a “paradigm shift” on which the UK’s current embrace of long-term nicotine maintenance and tobacco harm reduction actually rests, and which severs any link between the failures of product modification and widespread fears of a redux today.
Yet Russell’s work represents more a variation in theme than it does revolution in content. Russell’s policy recommendations operate from the same premises a...
NOT PEER REVIEWED
We thank Dr. Jarvis for his appreciation of our historical scholarship but disagree that our conclusion, “the promotion of tobacco harm reduction may serve the interests of tobacco companies more effectively than the public,” is an attack.
Our paper is about how policy affects ideas and vice versa. The ideas guiding the product modification program led to bad outcomes. That these ideas have been reanimated merits critical assessment. Voluntary agreements led to industry influence over the ISCSH’s recommendations, which in turn undermined public health. We point out that some of the same premises that led the ISCSH astray are popular again. Jarvis claims that current UK harm reduction policy has nothing to do with the product modification program, and everything to do with the influence of the late Michael Russell. Russell’s impressive scholarship – and oft-quoted statement, “people smoke for the nicotine, but die from the tar” – is indeed hugely influential among proponents of tobacco harm reduction. Jarvis posits that Russell’s work serves as a “paradigm shift” on which the UK’s current embrace of long-term nicotine maintenance and tobacco harm reduction actually rests, and which severs any link between the failures of product modification and widespread fears of a redux today.
Yet Russell’s work represents more a variation in theme than it does revolution in content. Russell’s policy recommendations operate from the same premises as the ISCSH. While Russell criticized the ISCSH for its discount of compensation, (which he believed medium- high- nicotine cigarettes could help remedy) the same foundations appear in his recommendations: reduced toxicity is reduced risk, (1) collaboration with the tobacco industry doesn’t influence outcomes, (2) nicotine addiction is inevitable, (3) and products must be popular to curtail use. (4)
We caution that enthusiasm for the tantalizing fantasy of reduced risk products and tobacco industry resources may lead us to readopt the premises promoted by the tobacco industry for decades.
In their response, Drs. Britton and O’Connor implore public health to promote evidence-based tobacco harm reduction. We could not agree more.
1. The Lancet. Nicotine Use After the Year 2000. Lancet. 1991; 337: 1191-2.
2. Russell M. Conversation with Michael AH Russell. Addiction (Abingdon, England). 2004; 99: 9-19.
3. Jarvis M and Russell M. Comment on the Hunter Committee's second report. British medical journal. 1980; 280: 994.
4. Russell M. The future of nicotine replacement. British Journal of Addiction. 1991; 86: 653-8.
NOT PEER REVIEWED Martin Jarvis is right to describe the Hunter Committee era as “a sorry tale” that by and large is well told by Elias and Ling, but his assertion that “taken as a whole their paper reads more as an attack on current UK policy than as a scholarly contribution to the history of tobacco control” is way over the top, as is his criticism of “the editorial processes and decision-making of Tobacco Control”.
In a paper that runs to a little over five pages of text, there are very brief references to current policies on the first page, then further brief references towards the end, suggesting that there are lessons to be drawn from the earlier episodes.
The paper might indeed have expanded further on the industry-friendly record of the Hunter Committee, noting that after his term as Chairman of the Committee ended, Lord Hunter became a consultant for Imperial Tobacco, while a civil servant who worked on smoking and serviced the Hunter Committee went on to work for Gallahers. It might also have included more emphasis on the way tobacco substitutes dominated public discourse on tobacco policy issues during the 1970s (1), although in fairness to the authors they appear to have been misled by the re-writing of history evident in some of the material they cite, particularly from industry actors.
But this would simply have added more weight to the conclusion that during the 1970s discussion, debate and massive promotion of tobacco substitutes by t...
NOT PEER REVIEWED Martin Jarvis is right to describe the Hunter Committee era as “a sorry tale” that by and large is well told by Elias and Ling, but his assertion that “taken as a whole their paper reads more as an attack on current UK policy than as a scholarly contribution to the history of tobacco control” is way over the top, as is his criticism of “the editorial processes and decision-making of Tobacco Control”.
In a paper that runs to a little over five pages of text, there are very brief references to current policies on the first page, then further brief references towards the end, suggesting that there are lessons to be drawn from the earlier episodes.
The paper might indeed have expanded further on the industry-friendly record of the Hunter Committee, noting that after his term as Chairman of the Committee ended, Lord Hunter became a consultant for Imperial Tobacco, while a civil servant who worked on smoking and serviced the Hunter Committee went on to work for Gallahers. It might also have included more emphasis on the way tobacco substitutes dominated public discourse on tobacco policy issues during the 1970s (1), although in fairness to the authors they appear to have been misled by the re-writing of history evident in some of the material they cite, particularly from industry actors.
But this would simply have added more weight to the conclusion that during the 1970s discussion, debate and massive promotion of tobacco substitutes by tobacco companies succeeded in hijacking the tobacco public policy and media agenda, and diverted the government of the day from evidence-based approaches recommended by health authorities to reduce smoking. Elias and Ling have good reason to suggest that there are similarities between then and now.
In passing, Jarvis’s summary that the novel products of the day failed “because consumers rejected them” omits much. One important reason for the failure of tobacco substitutes was that industry rhetoric and hype to the contrary, there was ultimately justifiable concern, with concomitant media coverage, that “all the generally available brands contained 75 per cent tobacco, and were no more than additions to the low and low-to-middle tar range” (1). Indeed, it would have been quite possible for smokers switching to part-substitute brands to find themselves moving to higher tar products.
But before this message hit, there had been a mountain of industry-generated publicity promoting the notion that “safer” and (as inevitably reported) “safe” new brands were on the way. This approach was supported with public relations ploys such as an Imperial Tobacco press conference about part-substitute cigarettes the day after the launch of the 1977 Royal College of Physicians’ “Smoking Or Health” report. Not surprisingly, the tobacco trade press was quick to congratulate the company concerned on its success in diluting the impact of the College report.
1. Daube M. The politics of smoking: thoughts on the Labour record. Community Medicine. 1979; 1, 306-314
NOT PEER REVIEWED
Elias & Ling throw useful light on the slow-motion disaster that was the series of voluntary agreements begun in the 1970s between government and the tobacco industry in the UK, overseen by the Independent Scientific Committee on Smoking and Health (ISCSH). These had as their aim to address the issue of tobacco product modification to reduce the health risks of smoking. Industry produced new smoking materials with the aim of reducing the biological activity of the tar fraction of smoke from cigarettes, and agreed to a programme of gradual tar yield reduction across the years. The novel products failed because consumers rejected them (there were too few users even to recruit for trials to examine their potential benefits), and the reductions in machine-smoked tar yields were achieved largely through increasing filter ventilation. The material cited shows that the low tar programme fiasco was characterized by undue influence from tobacco industry and a complete lack of understanding of the dynamics of smoking behaviour on the part of the scientific experts charged by government with supervision of the programme.
This is a sorry tale from the early days of tobacco control, and Elias & Ling tell it well. So far so good. But in framing and interpreting their material they go well beyond the data they cite, and draw quite unwarranted conclusions about what they see as the deficiencies of the current UK harm reduction policy. Indeed, tak...
NOT PEER REVIEWED
Elias & Ling throw useful light on the slow-motion disaster that was the series of voluntary agreements begun in the 1970s between government and the tobacco industry in the UK, overseen by the Independent Scientific Committee on Smoking and Health (ISCSH). These had as their aim to address the issue of tobacco product modification to reduce the health risks of smoking. Industry produced new smoking materials with the aim of reducing the biological activity of the tar fraction of smoke from cigarettes, and agreed to a programme of gradual tar yield reduction across the years. The novel products failed because consumers rejected them (there were too few users even to recruit for trials to examine their potential benefits), and the reductions in machine-smoked tar yields were achieved largely through increasing filter ventilation. The material cited shows that the low tar programme fiasco was characterized by undue influence from tobacco industry and a complete lack of understanding of the dynamics of smoking behaviour on the part of the scientific experts charged by government with supervision of the programme.
This is a sorry tale from the early days of tobacco control, and Elias & Ling tell it well. So far so good. But in framing and interpreting their material they go well beyond the data they cite, and draw quite unwarranted conclusions about what they see as the deficiencies of the current UK harm reduction policy. Indeed, taken as a whole, their paper reads more as an attack on current UK policy than as a scholarly contribution to the history of tobacco control. Current policy, which had its beginnings some 20 years or more later, is painted as being an outgrowth of the low tar programme and as being similarly tainted with industry influence. It reflects no credit on the editorial processes and decision making of Tobacco Control that authors were permitted to get away with such unsupported conclusions in their published paper.
The idea that reduction in tar yield of cigarettes could be an effective way of reducing the health risks of smoking goes back to the 1960s at least. It was supported by many distinguished epidemiologists, including Doll, Peto, and Wynder. See for example this 1979 statement by Ernst Wynder: ““One of the things that has always appealed to me about science is that if I find something that makes biological sense, then I feel reassured. Thirty years ago, when we had a 40mg tar cigarette, if you smoked 30 cigarettes a day you were exposed to about 1200 mg tar a day. Today’s cigarettes have 20 mg tar, so you are exposed to 600 mg daily. If there’s one thing everybody can agree on, it is that all tobacco-related cancers are dose related.” (1)
What was lacking in this formulation was an understanding that the risk from products is not simply determined by their composition, but resides in the interaction between the user and the product – as Michael Russell pointed out, if tar yield was everything, the most dangerous product would be the large Cuban cigar (1).
Russell was an outspoken critic of the ISCSH’s record. In a 1980 comment on the committee’s second report (2) he noted “Most of all it [the report] is remarkable for the stunning naivety of its implicit model of smoking behaviour”. He pointed out that: “Lower risk cigarettes are equated with cigarettes with lower tar and nicotine yields. If people smoked cigarettes in the same way that smoking machines do, this would indeed be the case. But there is much evidence that they do not….The tendency for smokers to regulate their smoke intake has been ignored by the Hunter Committee”.
Russell was not a member of the ISCSH (indeed, the membership of the committee was heavily tilted towards epidemiology and toxicology, and included no behavioural science expertise). It is in Russell’s insights and policy prescriptions that we can see the real intellectual underpinnings of UK policy on harm reduction in recent years.
Russell had a number of key insights, the most fundamental of which is that people smoke for nicotine (3). This coupled with the observation that the harms of smoking come not from the drug people are seeking, but from contaminants of the delivery system used to obtain it, led to his dictum “People smoke for the nicotine, but die from the tar”. As early as the mid-1970s he began to advocate an approach to tobacco harm reduction that was based on an appreciation of nicotine’s role (4, 5). He was the first to expose the fallacy of low tar cigarettes (6) and the first to explore the potential of non-combustible nicotine products for harm reduction (7-9). His 1991 paper in the British Journal of Addiction (10) and editorial in The Lancet (11) set out a vision for a transformed nicotine market: “If a strategy were adopted to sanction and encourage the use of purified nicotine products as substitutes for smoking, and at the same time impose stringent regulations on permissible constituents of cigarette smoke …….. the virtual elimination of smoking could become a more realistic health promotion target” (11). This represents a paradigm shift from the failed approach adopted in the 1970s. How it, or something springing from it, became a part of UK tobacco control policy in recent years, is a story for others to tell. But it did not happen through tobacco industry influence.
References
1. Gori GB, Bock FG. Banbury Report 3: A Safe Cigarette? Cold Spring Harbor Laboratory; 1980.
2. Jarvis MJ, Russell MAH. Comment on the Hunter Committee's second report. British Medical Journal. 1980;5; 280(6219):994-5.
3. Russell MA. Cigarette smoking: natural history of a dependence disorder. Br-J-Med-Psychol. 1971;44(1):1-16.
4. Russell MA. Realistic goals for smoking and health. A case for safer smoking. Lancet. 1974;16; 1(851):254-8.
5. Russell MA. Low-tar medium-nicotine cigarettes: a new approach to safer smoking. Br-Med-J. 1976;12; 1(6023):1430-3.
6. Russell MAH, Jarvis M, Iyer R, Feyerabend C. Relation of nicotine yield of cigarettes to blood nicotine concentrations in smokers. British Medical Journal. 1980;280(6219):972-6.
7. Russell MAH, Jarvis MJ, Feyerabend C. A new age for snuff? Lancet. 1980;1; 1(8166):474-5.
8. Russell MAH, Jarvis MJ, Devitt G, Feyerabend C. Nicotine intake by snuff users. Br Med J. 1981;26; 283(6295):814-7.
9. Russell MAH, Jarvis MJ, West RJ, Feyerabend C. Buccal absorption of nicotine from smokeless tobacco sachets. Lancet. 1985;2(8468):1370.
10. Russell MAH. The future of nicotine replacement. Br-J-Addict. 1991;86(5):653-8.
11. Russell MAH. Nicotine use after the year 2000. The Lancet. 1991;337:1191-2.
NOT PEER REVIEWED
Berry et al (1) report an analysis of two waves of the Population Assessment of Tobacco and Health (PATH) study focused on the association between the initiation of e-cigarette use by Wave 2 and cigarette abstinence/reduction assessed at Wave 2. They conclude that daily e-cigarette use is associated with both cigarette abstinence and reduced consumption among continuing smokers. While this addresses an important question, we argue that such analyses should be adjusted for the reason e-cigarettes are being used.
From Wave 1 of PATH (2), we know that ~75% of smokers agreed that e-cigarettes were useful to help people quit. However, ~80% agreed that e-cigarettes allowed someone to replace a cigarette where smoking was prohibited. From the first reason, we can hypothesize that e-cigarette use might be associated with cigarette abstinence/reduction. However, from the second reason, we can also hypothesize that e-cigarettes would be associated with neither cigarette abstinence nor reduction. The recent National Academies report (3) recommended that any assessment of the role of e-cigarettes in cigarette cessation/reduction should focus on smokers who used e-cigarettes to help them quit.
Show MorePATH Wave 2 data does include information on whether smokers tried to quit in the previous year, as well as whether they used e-cigarettes to aid the last quit attempt. Previous research (4) has shown that over half of the smoking population will not ha...
NOT PEER REVIEWED
Feliu et al’s conclusion “in the European Union countries with the higher scores in the Tobacco Control Scale, which indicates higher tobacco control efforts, have lower prevalence of smokers, higher quit ratios and higher relative decreases in their prevalence rates of smokers.” deserved comment.
First, it seems a tautology. Tobacco control policies are robustly evidence based. Accordingly, more efforts, less smokers.
Second, a PubMed search with “"tobacco control scale" only retrieved 27 articles since 2006 and no validation published yet. Obviously, the Scale poorly correlated with smoking rate: r2 being .58 in 2002/3, .15 in 2006/7 and .06 in 2010/11.(From table 3 in 2; n= 11 European countries).
Third, why make simple stuff complex? This surrogate is complex to calculate and its items are subjective because issuing a decree is useless if no implementation were enforced. In contrast, the smoking rate and its evolution are simple and reliable! How France can be ranked 4th among 28 countries with a 57/100 score (1) while smoking prevalence has been plateauing for so long at more than 30%? In France, from 2004 to 2017 no relevant increase in tobacco taxes, no implementation of the legal smoking ban in cafés or of the ban of sale to minors despite sting operations by NGO showing evidence of serious breaches.(3)
Fourth, claiming “the European Union should continue implementing comprehensive tobacco control pol...
Show MoreNOT PEER REVIEWED
This paper’s core findings are quite helpful: (1) Tax/price increases for non-cigarette tobacco products can effectively reduce their use; and (2) Tax/price increases for non-cigarette tobacco products could prompt some users to increase their cigarette smoking if comparable tax/price increases for cigarettes are not done at the same time. But the paper’s related analysis is incomplete, producing misleading conclusions, largely because the paper focuses on cigarettes versus non-cigarette tobacco products without also considering the more important distinction for health-directed tobacco tax strategies between smoked tobacco products and non-combustible tobacco products.
In its abstract, the paper concludes that the “positive substitutability between cigarettes and non-cigarette tobacco products suggest that tax and price increases need to be simultaneous and comparable across all tobacco products.” But the paper does not appear to consider that the only substitutions that could significantly increase public health harms would be if the tax increases prompted some non-combusted tobacco product users to move to more-harmful smoking or prompted some smokers who would otherwise do so not to move to less-harmful non-combusted tobacco products. As a result, the paper fails to acknowledge that significant tax/price increases for only combusted tobacco products would not prompt any harm-increasing substitution and would directly secure desirable...
Show MoreNOT PEER REVIEWED
The paper by Filippidis et al [1] provides data re-confirming the well-known fact that most ex-smokers attempt to quit without using any form of assistance, whether pharmaceutical, professional or via e-cigarettes. Moreover, the proportion of ex-smokers trying to quit unaided increased substantially in Europe between 2012-17 (ex-smokers using no assistance increased from 73.9% to 80.7%), a period where e-cigarette use accelerated in some nations.
Regrettably however, this study does not permit any comparison of success rates by method, as no data are reported on which method of cessation (assisted v unassisted) was used by ex-smokers on their last, final (and so successful) quit attempt.
The authors report that those “who successfully quit reported much lower use of cessation assistance compared with smokers who had tried to quit without success” and suggest that this might reflect indication bias, whereby those who find it harder to quit self-select to use assistance, leaving the low hanging fruit of non- or less addicted smokers to fall off the smoking tree using their own determination.
While this will be true for some, there are many former heavy smokers who quit without assistance. This argument also borrows assumptions from the discredited hardening hypothesis [2], which holds, in the face of evidence to the contrary, that as smoking prevalence falls the concentration of hardened, more deeply addicted smokers increase...
Show MoreThere is also very clear evidence that tobacco industry interference is either delaying or dumbing down implementation of each of the MPOWER policies particularly in LMICs. The TC vaccine is a good concept but the framework needs to include monitoring, exposing and countering industry tactics.
I am grateful to Bashash et al. for raising some important methodological and policy-related issues. Responding to their specific points:
(1) Very high formaldehyde concentrations may arise in aerosols when atomisers generate excessive heat[1]. Under these circumstances recommended safety limits for formaldehyde may indeed be exceeded and this compound contributes most to the cancer potency summation.
(2) Goodson et al. [2] provide a framework for assessing whether low dose compounds that are not necessarily individual carcinogens may become involved in carcinogenesis when acting in concert. Although discussed under "Strengths and limitations" synergystic phenomena were not accommodated in the cancer potency model as it is not yet possible to predict the mechanism and magnitude of such interactions in tobacco or e-cigarette aerosols. Under the Goodson et al. model adverse effects reflect adventitious synergystic combinations. These may be statistically more likely in tobacco smoke where the number of different compounds greatly exceeds those of simpler aerosols, however this effect is expected to be minor compared with the exceptionally high carcinogenic potencies of some well-established carcinogens in tobacco smoke.
(3) Lifetime cancer risk is linearly dependent on the daily volume of vapour inhaled (equation 7) and the effect on risk of increased consumption after switching to heat not burn (HnB) products is directly related to the chang...
Show MoreNOT PEER REVIEWED The Jawad et al systematic review and meta-analysis examining price effects for non-cigarette tobacco and nicotine products appears methodologically sound and was a registered analysis. It provides information that could be used productively by advocates and policymakers seeking to reduce harm. The cross-elasticities reported in this paper can be used to the advantage of public health by increasing the impact of policies that seek to drive down smoking.
However, this work does not take into account the fact that not all tobacco and nicotine products cause the same level of health harms as combustible cigarettes. The paper examines own- and cross-price elasticity across a wide array of products – from combustible tobacco products such as kreteks and little cigars to nicotine-only products such as e-cigarettes and nicotine patches – and then discusses consumption patterns in terms of an undifferentiated aggregate of nicotine use. Jawad and colleagues do not consider the health implications of policies to move nicotine users from more-harmful to less-harmful means of administration (see, for example, Chaloupka, Warner and Sweanor, 2015, recommending differential taxation according to differential risk).
From a public health perspective, any analysis of nicotine-use patterns should consider differential harm levels. A focus on nicotine use as the sole outcome variable can be seriously misleading and detrimental to the goal of reducing smoking....
Show MoreNOT PEER REVIEWED
We thank Dr. Jarvis for his appreciation of our historical scholarship but disagree that our conclusion, “the promotion of tobacco harm reduction may serve the interests of tobacco companies more effectively than the public,” is an attack.
Our paper is about how policy affects ideas and vice versa. The ideas guiding the product modification program led to bad outcomes. That these ideas have been reanimated merits critical assessment. Voluntary agreements led to industry influence over the ISCSH’s recommendations, which in turn undermined public health. We point out that some of the same premises that led the ISCSH astray are popular again. Jarvis claims that current UK harm reduction policy has nothing to do with the product modification program, and everything to do with the influence of the late Michael Russell. Russell’s impressive scholarship – and oft-quoted statement, “people smoke for the nicotine, but die from the tar” – is indeed hugely influential among proponents of tobacco harm reduction. Jarvis posits that Russell’s work serves as a “paradigm shift” on which the UK’s current embrace of long-term nicotine maintenance and tobacco harm reduction actually rests, and which severs any link between the failures of product modification and widespread fears of a redux today.
Yet Russell’s work represents more a variation in theme than it does revolution in content. Russell’s policy recommendations operate from the same premises a...
Show MoreNOT PEER REVIEWED Martin Jarvis is right to describe the Hunter Committee era as “a sorry tale” that by and large is well told by Elias and Ling, but his assertion that “taken as a whole their paper reads more as an attack on current UK policy than as a scholarly contribution to the history of tobacco control” is way over the top, as is his criticism of “the editorial processes and decision-making of Tobacco Control”.
In a paper that runs to a little over five pages of text, there are very brief references to current policies on the first page, then further brief references towards the end, suggesting that there are lessons to be drawn from the earlier episodes.
The paper might indeed have expanded further on the industry-friendly record of the Hunter Committee, noting that after his term as Chairman of the Committee ended, Lord Hunter became a consultant for Imperial Tobacco, while a civil servant who worked on smoking and serviced the Hunter Committee went on to work for Gallahers. It might also have included more emphasis on the way tobacco substitutes dominated public discourse on tobacco policy issues during the 1970s (1), although in fairness to the authors they appear to have been misled by the re-writing of history evident in some of the material they cite, particularly from industry actors.
But this would simply have added more weight to the conclusion that during the 1970s discussion, debate and massive promotion of tobacco substitutes by t...
Show MoreNOT PEER REVIEWED
Show MoreElias & Ling throw useful light on the slow-motion disaster that was the series of voluntary agreements begun in the 1970s between government and the tobacco industry in the UK, overseen by the Independent Scientific Committee on Smoking and Health (ISCSH). These had as their aim to address the issue of tobacco product modification to reduce the health risks of smoking. Industry produced new smoking materials with the aim of reducing the biological activity of the tar fraction of smoke from cigarettes, and agreed to a programme of gradual tar yield reduction across the years. The novel products failed because consumers rejected them (there were too few users even to recruit for trials to examine their potential benefits), and the reductions in machine-smoked tar yields were achieved largely through increasing filter ventilation. The material cited shows that the low tar programme fiasco was characterized by undue influence from tobacco industry and a complete lack of understanding of the dynamics of smoking behaviour on the part of the scientific experts charged by government with supervision of the programme.
This is a sorry tale from the early days of tobacco control, and Elias & Ling tell it well. So far so good. But in framing and interpreting their material they go well beyond the data they cite, and draw quite unwarranted conclusions about what they see as the deficiencies of the current UK harm reduction policy. Indeed, tak...
Pages