Editor,—If regulators attempt to reduce nicotine in tobacco products, there is a danger that smokers will smoke more intensely to achieve a given nicotine dose from products with less available nicotine. This phenomenon, known as smoker compensation, is why the International Standards Organisation/Federal Trade Commission tar and nicotine yield measurements are highly misleading and why so-called low-tar cigarettes are far more harmful than their nominal yields suggest.

We feel that the AMA Council on Scientific Affairs is much too sanguine on the issue of smoker compensation, and that this effect could undermine the AMA’s suggested programme.1-1 The authors assert that “even though compensatory oversmoking is a concern, the extent may not prove to be great, and the accompanying risk of increased temporary exposure to tobacco toxins may be acceptable if the ultimate benefit will be cessation in a few years.” Oversmoking may in fact be a serious problem, and it is very risky to suggest any policy where health impacts get worse before they get better. In our view, the currently available research evidence and the disclosures of internal tobacco industry research from the Minnesota litigation,1-2 clearly establish the pervasiveness of the compensation phenomenon, and cannot justify rejection of the hypothesis that compensation is complete.

Evidence on nicotine compensation comes from two main sources: comparison of nicotine intakes among smokers who have self-selected to smoke brands of differing nominal yields; and forced-switching studies in which smokers, usually for a brief period, are experimentally shifted to lower yielding brands. These studies have established that smokers characteristically oversmoke low-yielding brands and achieve extensive compensation for reduced nicotine delivery. The extent of compensation observed varies, but is usually somewhat less than complete.1-3 But neither of these study designs gives definitive information. Those who self-select to lower yielding brands may do so in part because their preferred nicotine dose is lower and can be satisfied from these brands without smoking more intensely; no causal effect of brand yield in lowering exposure can be inferred. With forced-switching studies, smokers may not achieve full compensation, but equally they may find the cigarettes unpalatable and the increased intensity of puffing required1-4 unacceptable, so that in the real world they would not elect to smoke them. It is probably not irrelevant that brands with less than 0.5 mg nicotine yield captured only 3% of the market in the United Kingdom in 1994.1-5 If, through regulation, smokers had no choice but to switch to brands whose nicotine delivery was so low that full compensation was impossible, it would be necessary to acknowledge that the proposal embodies a degree of coercion and compulsory withdrawal. Faced with “unsatisfactory” cigarettes, smokers may try to compensate by seeking nicotine from other sources, such as smuggled cigarettes.

The AMA’s review overlooked the two largest, long-duration studies of switching, both of which randomised smokers and followed them over a six-month period.1-6 1-7 These studies found it difficult to recruit and retain smokers for the study duration, illustrating smokers’ resistance to shifting.1-8 One study observed complete nicotine compensation1-6 and in the other it was close to complete.1-7

The AMA report makes much of a short-term study of switching in an in-patient setting in which little compensation for reduced nicotine was observed.1-9 But several considerations challenge the relevance of this finding for the general population of smokers. The study lasted three days, and involved only 12 smokers, almost all of whom smoked filter brands, who were switched to a non-filter Kentucky reference cigarette acknowledged to be unpalatable. They were also required to smoke according to an imposed schedule.

There is an overwhelming case for the United States Food and Drug Administration’s assertion of regulatory jurisdiction over tobacco products, and there are many ways in which harm could be reduced. But in our view it is not yet possible to justify on health grounds a programme to remove nicotine from tobacco products. We believe that any measures, particularly radical approaches, should be rooted in a well-founded understanding of smokers’ behaviour.

Editor,—As proposed by Henningfieldet al for the American Medical Association,2-1 serious nicotine reduction will result in increased quitting, and importantly, fewer relapses. Smokers will become ex-smokers, as cigarette nicotine content falls gradually below the addictive threshold.

Most smokers in New Zealand are periodically attempting to quit: 10% said they were likely to quit within a month; 38% said they had quit in the past year for a week or more; and another 43% quit for a day but less than a week.2-2

Though lowering nicotine could make smoking seem less risky to adolescents, and encourage experimentation, it would help adolescents to quit sooner. As it is, four out of five New Zealand smokers at ages 14 and 15 say it is difficult to quit,2-3 61% of 18-year-old daily smokers say they are addicted,2-4 and 89% of those experimenting at age 15 were still smoking at age 18 years.2-5

As with advertising bans, some countries will legislate ahead of others. Health agencies will need to research the policy into a winnable piece of legislation, commit resources for a long fight, build a coalition, feed the media, educate the public, and allay smokers’ fears. Tobacco manufacturers will oppose any cigarette nicotine reduction law with every resource at their command.

Is this prohibition? No. Smokers will not be fined for possession. Smoking and cigarettes will remain, and so will nicotine, though mainly in other forms. Cigarette brand names will survive. The tobacco trade will have 10 years to overturn the new law. Smokers will stop buying cigarettes, and will live 7 years longer on average. Smoking will cause fewer than 2% of all deaths.

Smuggling would not be a major problem in an island state such as New Zealand. Mail order might be permitted in the meantime for registered nicotine-dependent smokers, but shops would no longer sell addictive cigarettes. The key to success may be pharmaceutical companies developing nicotine products truly satisfying to those who quit smoking but who still want a regular nicotine buzz.

In New Zealand tobacco control agencies will carefully consider this proposal. It is a potential winner. At the very least, if only one group in each country supports it, it will scale up the debate several notches, and make it easier for other groups to get support for current policies.

In reply,—We appreciate the interest that our proposal to gradually remove the nicotine from cigarettes has generated as evidenced by these three letters and many other communications we have received and seen. These letters give us an opportunity to clarify what seem to be the major issues expressed by colleagues in the scientific and public health community.

One of the many debates is whether such a proposal ought to be implemented all at once with sudden elimination of nicotine from cigarettes (for instance, on some future New Year’s Day) or gradually, as we proposed. The main argument for the sudden nicotine elimination approach is that it would reduce the period during which people seeking to sustain nicotine intake would inadvertently increase exposure to tar and other cigarette toxins (the “compensatory smoking” concern raised by Jarvis and Bates).

Shatenstein argues that many smokers are not yet interested in quitting (they are “contemplators” or “pre-contemplators” in the vernacular of the stages-of-change model referred to in his letter) and they would strongly resist such an approach. On the contrary, 70–80% of smokers in the United States say they want to quit.3-1Even though some may not be “ready” or “prepared” to quit at a given point in time, we believe a nicotine reduction policy accompanied by the other programmatic elements recommended in our report3-2 would motivate many or most of them to quit.

In any event, to address this concern, we suggested phasing nicotine out of cigarettes gradually, over up to 10 years from the time the protocol was initiated. As discussed in the AMA report,3-2enabling the process would require supportive education to consumers and health professionals alike, expansion of treatment options including alternate forms of nicotine delivery (nicotine replacement medicines), and other supportive measures preceding and accompanying the phasing out of nicotine from cigarettes. We are encouraged by the overall experience in California and Massachusetts in which there has been remarkably strong support from smokers as well as non-smokers for tobacco control measures and restrictions on smoking that a few years ago met with objections. A major lesson from these experiences is that the viability of a tobacco control measure cannot be determined simply by examining the measure in isolation. Rather, the context in which it is presented—for example, educational support and increased availability of alternative products for treatment—is also important.

A gradual nicotine reduction approach is also supported by clinical research on nicotine withdrawal and treatment, which has demonstrated that minimising withdrawal symptoms can be achieved by gradual, as opposed to precipitous, nicotine deprivation. Furthermore, the AMA report recommended more rapid and extensive surveillance of all forms and patterns of tobacco and nicotine use to enable rapid detection of unintended consequences (including strong resistance) to guide course corrections. Thus, we support gradual nicotine withdrawal, accompanied by supportive policies such as increases in tobacco taxes, expanded treatment access, and increased availability of safe forms of nicotine.

We share the concerns of Jarvis and Bates, that gradual nicotine reduction could lead to increased toxin exposure because of compensatory increased smoke intake as smokers smoked more cigarettes or extracted more toxins per cigarette. This issue was discussed on pages 285–286 in the report.3-2 There is an enormous literature on this topic and we welcome the mini-review provided in the letter from Jarvis and Bates. Our intent was not to imply that compensation would not be a problem, but rather that, with consideration given to research findings (including those referred to by Jarvis and Bates), potential adverse consequences could be minimised. This conclusion follows from several observations.

First, compensation is not an all-or-none phenomenon; there are known factors that could influence the degree of compensation as well as the consequences of compensation. Second, prior marketplace experience differs from our proposal in that our proposal would provide extensive guidance to consumers (including meaningful product labelling) to encourage and enable them to minimise compensatory smoking. Third, the degree to which compensation occurs and its consequences can be affected by the design of cigarettes themselves. It appears that most commercially marketed cigarettes have been designed so as to maximise the likelihood of compensatory smoking by employment of physical and chemical engineering techniques to ensure delivery of high doses of nicotine deep into the lungs.3-3 3-4 Such designs could be prohibited if they undermined the goals of nicotine reduction and if they contributed to increased occurrence and consequences of compensation. Finally, in parallel to reductions in nicotine, tobacco manufacturers could be required to provide reductions in specific toxins—for example, carbon monoxide and tobacco-specific nitrosamines—for which there is already evidence of significant potential for reduction.3-5-3-7

This proposal is like many programmes in public health, in which no alternative is devoid of risk, so the course with presumed lesser risk is chosen. An example is immunisation—vaccinations are not risk-free, but the benefits of immunisation clearly outweigh the risks. Thus, as we discussed in our proposal, every effort should be taken with future cigarettes (whether or not a nicotine reduction strategy is implemented) to reduce the occurrence and adverse consequences of compensation—for example, by banning hidden ventilation holes and providing meaningful labelling of the dosing characteristics of cigarettes.

Shatenstein notes that tobacco manufacturers would not cooperate in an enterprise that would threaten their survival. The AMA report mentioned possible industry cooperation only in reference to “ways in which contraband could be suppressed by intensifying anti-smuggling law enforcement activities and by enlisting improved cooperation of manufacturers.” Manufacturers might be forced to cooperate in anti-smuggling efforts because smuggling is illegal. But even if the manufacturers don’t cooperate, anti-smuggling activities can proceed quite well without them, if governments muster the political will and devote the resources necessary to address the problem. The fact that an RJ Reynolds (RJR) marketing partner recently pleaded guilty to federal criminal charges of smuggling,3-8 and that a former RJR marketing executive has been charged with aiding tobacco smugglers,3-9 provides evidence that this is already happening.

We would add two more points about smuggling. First, it is not plausible that a massive contraband market for commercially produced cigarettes could occur without the complicity of the cigarette makers or marketers; this is true whether or not cigarettes use high-nicotine yielding strains of tobacco. Therefore, as discussed in the AMA report, more effective monitoring of the transactions of major tobacco companies would be essential. Second, even if there was a contraband market, the size and health impact of that market would be much less than the current cigarette market, which is supplied by manufacturing operations that produce and ship billions of cigarettes every day in the United States alone. The potential for smuggling is another reason for the recommendation in the report to include rigorous surveillance to enable rapid detection of unintended consequences and corrections in course.

Shatenstein also mentioned the cigarette marketed by Philip Morris under the brand name “Next”, as an example of a denicotinised cigarette that was widely rejected by cigarette smokers. However, the circumstances surrounding Next were dramatically different from those envisioned under our proposal. For example, Next cigarettes were marketed using terms previously applied to high-nicotine yielding cigarettes, such as “rich taste”. They were not developed or marketed as a brand that might be used as an aid to cessation. Consumers were not told that switching to these cigarettes might lead to nicotine withdrawal symptoms and that they might need to prepare for this, perhaps by gradual reduction of smoking from their regular cigarettes. They were not marketed as part of a national programme that would have included the kinds of educational and medical support that our report discussed. These strategies might have required some supportive research by the company and approval by regulatory agencies. Lastly, Next might have failed because high-nicotine yielding cigarettes, backed by aggressive marketing campaigns, were still available as alternatives to Next; those choices and countervailing influences would not be available in a nicotine reduction policy imposed on a national scale. The only surprise at all concerning Next was that Philip Morris spent so much money on a cigarette marketing approach that its own history and research might have suggested was a doomed approach.

The comments by Laugesen offer the observation with which we concur—that nicotine reduction is not the same as cigarette prohibition. Interestingly, the so-called official position of the tobacco industry is mixed on this topic. As noted in the letter from Shatenstein, and elsewhere,3-3 3-4 in their previously secret documents, tobacco industry officials did equate cigarettes with nicotine, calling nicotine the sine qua nonof cigarettes. Paradoxically, in the sworn testimony of senior tobacco industry officials to the United States Congress in 1994, in recent court cases by experts on behalf of the tobacco industry, and in their official comments to the United States Food and Drug Administration (FDA),3-10 the role of nicotine has been down-played and framed primarily as a natural cigarette constituent with minor pharmacological and sensory effects. Setting aside, however, any hypocrisy (and possibly perjury) on the part of the tobacco industry, there is a difference between a proposal to ban the product and one that would strip it of a chemical that causes addiction. For those who would seek some form of image or status by consumption of cigarettes, the product could remain. For those who require nicotine to sustain desirable functions, presumably in most cases as a consequence of their history of chronic nicotine use, other forms of nicotine delivery would be available which would have been approved by the FDA for safety and efficacy. This would include existing FDA-approved nicotine delivery systems and new forms of “clean” nicotine delivery to either fully or partially replace the nicotine delivered by conventional tobacco products, as has been discussed elsewhere.3-5-3-7 3-11

Therefore, a major goal to improve health is not to sustain the current system whereby the most toxic and addictive forms of nicotine delivery are the most readily available, nor a system in which cigarettes and nicotine are both unavailable. Rather, we believe we should be moving toward a day when commercially marketed cigarettes are not addictive and no more toxic than necessary; and that conversely, acceptable forms of treatment to reduce or eliminate tobacco toxin exposure are more readily available. This is not prohibition of a product or elimination of the opportunity to obtain nicotine; it is the opportunity to make choices about tobacco and nicotine-product use that are not compromised by the seductive effects of nicotine in cigarettes.

In reply,—While I found the letters from Shatenstein and Jarvis and Bates to be useful contributions to the debate, and generally very well informed, several of their points warrant comment.

Shatenstein argues that the bull’s-eye on the tobacco control target should be “a denormalised product: in essence, rendering the palatable unpalatable”—in other words, to make tobacco use so socially unacceptable that people will stop partaking of it, despite its addictiveness. After more than a decade as a lawyer-advocate, in which many of my efforts have been devoted to doing this very thing, I cannot agree more that the denormalisation of tobacco in the public mind is a critical feature of any comprehensive tobacco control campaign. However, even the most effective efforts at denormalisation are destined to take decades longer than necessary in the absence of loosening the vice grip of the product itself on the brains and central nervous systems of hundreds of millions of users. Millions more will become addicted as adolescents and die painfully and prematurely as adults, a direct and certain consequence of leaving control of the product in the hands of the cigarette makers.

In my editorial,4-1 I observed that the Ford Motor Company had to change its defective Pinto car when it was found to be killing people in low-speed crashes, an analogy I used to help explain why cigarette companies should likewise be forced to change their products. Shatenstein responds by noting that “there are no more Pintos on the roads these days” and that “[c]onsumers moved on to more trustworthy models”. He notes in the same paragraph that American consumers chose not to switch to the Next (denicotinised) cigarettes when they were marketed briefly in the early 1990s. First, Shatenstein overlooks the fact that the Pinto is merely one example of a much larger phenomenon, in which government safety authorities around the world have required improvements in the design and manufacture of all motor vehicles. In the United States, every mass-marketed automobile has long been required to meet strict safety standards, which have been progressively strengthened over time. Thus, once the Pinto was deemed hazardous, it could no longer be sold without first being fixed.

As for the example of Next cigarettes, he fails to recognise that consumers failed to buy the denicotinised product in part because its manufacturer, Philip Morris, deliberately neglected to inform the public that the product was not only non-addicting, but was also, through the extraction of nicotine, nearly bereft of tobacco-specific nitrosamines, which are among the most lethal of cigarette smoke carcinogens. In short, Next probably was a less hazardous cigarette, but the public did not know it.

In stark contrast to the automobile example, no country has yet to subject tobacco products to meaningful health and safety standards. The historic failures of political will in the face of the industry’s money and power are no excuse for failing now to regulate the manufacture of tobacco products in light of the recent disclosures regarding cigarette makers’ design of those products to addict unsuspecting consumers, millions of children among them.

Jarvis and Bates make the misplaced argument that the AMA proposal4-2 “embodies a degree of coercion and compulsory withdrawal”. Shatenstein similarly likens the AMA proposal to “foist[ing] nicotine-reduced cigarettes on an unwilling public”. First, the only coercion and withdrawal that could conceivably be at issue involves nicotine itself, since tobacco products would continue to be readily available to adult consumers. Under the AMA proposal, nicotine, too, would continue to be made readily available through increasingly effective alternative delivery systems. Any withdrawal experienced by the nicotine-dependent consumer would come as a result of a choice, freely made, not to use any nicotine delivery product. That is precisely the choice that should remain available to adults. Moreover, as Laugesen notes, to take but one example, more than eight of 10 smokers in New Zealand have tried to quit, but with extreme difficulty. Helping them—and millions of others like them around the world—conquer their addiction to nicotine-containing cigarettes could hardly be characterised as coercive, especially since the drug itself would still be readily available to them in other forms.

Finally, Shatenstein asserts that proposing the gradual reduction of nicotine in tobacco products will spawn accusations of nannyism and the hurling of the “health Nazi” insult, and that this will place us, not tobacco products, in the bull’s-eye. On the contrary, while the tobacco industry and its sympathisers will always reserve their choicest epithets for their adversaries, the proposal to phase out nicotine should help to shift the focus of public anger squarely toward the tobacco industry. The AMA report should be used as a platform for focusing on the industry’s malfeasance, not on the assertive good intentions of the public health and medical communities. The tobacco companies’ misconduct caused the nicotine-fuelled calamity from which we now seek to extricate society, and they must be challenged as to their “right” to continue to make and sell products that they deliberately render addictive at the same time that they fail to render them less hazardous.

The letters assailing the AMA report essentially wave the white flag of surrender, ceding to the tobacco industry the extraordinary discretion to use, manipulate, and freely adulterate their products to exploit the effects of a highly addictive drug on a planet of human guinea pigs. This position inevitably brings to mind an observation I once heard made by my friend Garfield Mahood. In his inimitable style, the Canadian tobacco-control leader intoned to an audience of health advocates, “Those who say it cannot be done should not get in the way of those who are doing it”.

The dramatic first step toward eradication of the tobacco epidemic must be taken thoughtfully and cautiously, not fearfully or timidly.