Summary of Regulatory Principles

A science-based regulatory framework for implementing tobacco harm reduction should conform to the following 11 principles:

1.

Manufacturers of tobacco products, whether conventional or modified, should be required to obtain quantitative analytical data on the ingredients of each of their products and to disclose such information to the regulatory agency.

2.

All tobacco products should be assessed for yields of nicotine and other tobacco toxicants according to a method that reflects actual circumstances of human consumption; when necessary to support claims, human exposure to various constituents of tobacco smoke should be assessed using appropriate biomarkers. Accurate information regarding yield range and human exposure should be communicated to consumers in terms that are understandable and not misleading.

3.

Manufacturers of all PREPs should be required to conduct appropriate toxicological testing in preclinical laboratory and animal models and appropriate clinical testing in humans to support the health-related claims associated with each product and to disclose the results of such testing to the regulatory agency.

4.

Manufacturers should be permitted to market tobacco-related products with exposure reduction or risk reduction claims only after agency approval based on scientific evidence (a) that the product substantially reduces exposure to one or more tobacco toxicants and (b) if a risk reduction claim is made, that the product can reasonably be expected to reduce the risk of one or more specific diseases or other adverse health effects, compared with whatever benchmark product the agency requires to be stated in the labeling. The “substantial reduction” in exposure should be sufficiently large that independent scientific experts would anticipate finding a measurable reduction in morbidity and/or mortality in subsequent clinical or epidemiological studies.

5.

The labeling, advertising, and promotion of all tobacco-related products with exposure reduction or risk reduction claims must be carefully regulated under a “not false or misleading” standard, with the burden of proof for the claim resting on the manufacturer not the government. The responsible agency should have the authority and resources to conduct surveys of consumer perceptions relating to these claims.

6.

The regulatory agency should be empowered to require manufacturers of all products marketed with claims of reduced risk of tobacco-related disease to conduct postmarketing surveillance and epidemiological studies as necessary to determine the short-term behavioral and long-term health consequences of using their products and to permit continuing review of the accuracy of their claims.

7.

In the absence of any claim of reduced exposure or reduced risk, manufacturers of tobacco products should be permitted to market new products or modify existing products without prior approval of the regulatory agency after informing the agency of the composition of the product and certifying that the product could not reasonably be expected to increase the risk of cancer, heart disease, pulmonary disease, adverse reproductive effects, or other adverse health effects, compared to similar conventional tobacco products, as judged on the basis of the most current toxicological and epidemiological information.

8.

All added ingredients in tobacco products, including those already on the market, should be reported to the agency and be subject to a comprehensive toxicological review.

9.

The regulatory agency should be empowered to set performance standards (e.g., maximum levels of toxicants; definitions of terms such as “low tar”) for all tobacco products, whether conventional or modified, or for classes of products.

10.

The regulatory agency should have enforcement powers commensurate with its public health mission, including the power to issue subpoenas.

11.

Exposure reduction and risk reduction claims for drugs and devices that are supported by appropriate scientific and clinical evidence should be allowed by the FDA.

The following sections elaborate on each of these principles.

Principle 1: Disclosure of Product Ingredients

Manufacturers of tobacco products, whether conventional or modified, should be required to obtain quantitative analytical data on the ingredients of each of their products and to disclose such information to the regulatory agency.

The manufacturers of tobacco products have detailed and extensive knowledge of the composition, curing, and blending of tobacco; the ingredients of their products; and the composition of tobacco smoke. They also have detailed quantitative information on the tobacco in each of their brands of cigarettes and smokeless products, since each is blended to achieve the desired taste and levels of nicotine and other ingredients. Manufacturers also conduct major research and development programs aimed at developing new products. Currently, there is no requirement for disclosure of this information to an appropriate regulatory body. Such disclosure is prerequisite to any meaningful scientific appraisal of the comparative risks of different tobacco-containing products or the potential for risk reduction offered by modified products. For smoked products, analytical information on the smoke and the concentrations of smoke components that are absorbed under actual smoking conditions may be even more important than knowledge of the product ingredients themselves. For smokeless tobacco products, whether conventional or modified, similar information on the concentrations of major ingredients in saliva and blood are equally important. The disclosure of quantitative information on ingredients (with appropriate safeguards to protect trade secrets) is a standard requirement in regulatory laws relating to drugs, biologics, devices, food additives, toxic chemicals, and environmental contaminants and should be required of all tobacco products, whether conventional or modified.

Although FTC regulation requires public reporting of some constituents in cigarette smoke, manufacturers are not required to report brand-specific information about the nicotine content or other properties (e.g., nitrosamine level) of the material that forms the tobacco rod. Under legislation enacted in 1986, manufacturers of smokeless products are required to report total nicotine content to the secretary of the Department of Health and Human Services (HHS), but the secretary may not release the data. Under the same legislation, tobacco manufacturers are required to submit lists of additives to the tobacco (but not to filters or papers) to the secretary of HHS. Information about the quantity of additives and their presence in specific brands is not required, and the secretary is bound to safeguard the information from public disclosure. In 1993, attorneys for six cigarette manufacturers released a combined list of 599 additives. The following year, ten manufacturers of smokeless products released a list of additives in their products. Three states have enacted legislation requiring disclosure of additives in tobacco products (U.S. DHHS, 2000).

Principle 2: Yield Assessment

All tobacco products should be assessed for yields of nicotine and other tobacco toxicants according to a method that reflects actual circumstances of human consumption; when necessary to support claims, human exposure to various constituents of tobacco smoke should be assessed using appropriate biomarkers. Accurate information regarding yield range and human exposure should be communicated to consumers in terms that are understandable and not misleading.

The actual yield—that is, the amount of toxicants inhaled by an individual—from particular cigarettes varies considerably among smokers. As discussed in Chapter 11, this is because the standard yield is measured by a machine that smokes cigarettes in a mechanical and standardized way, whereas smokers can and do smoke their cigarettes with different numbers of puffs and different depths of inhalation. As a result, for any product, the temperature of combustion and the composition of the smoke varies among smokers depending on the pattern of smoking. The information required to address the relative harmfulness of tobacco products will be available only if an improved methodology for ascertaining the range of actual toxicant yields in human consumers is developed and applied; only then will there be a scientific basis for developing some of the features of a harm reduction program, including, for example, accurate labeling, meaningful definition of terms such as “low tar”, and reasonable standards for yields of tar, carbon monoxide (CO), or nicotine.

Although the limitations of the FTC's current methodology for assessing cigarette yield is generally acknowledged, no definitive steps have been taken to replace it. In 1994, the President's Cancer Committee and the National Cancer Institute (NCI) issued a report criticizing the current approach and recommending adoption of a new one. In 1997, the FTC issued a proposed revision of the methodology and a proposed format for disclosing test results in advertising and labeling. However, many proponents of a new approach urged the FTC to defer action pending judicial clarification of the FDA's authority and possible congressional enactment of comprehensive regulation for tobacco products. It is time for the FTC— or, ideally, an agency with regulatory authority over tobacco products— to adopt a new standardized methodology to assess the range of toxic exposures to smokers under actual conditions of human smoking.

Although the focus of yield assessment has been smoked tobacco, an analogous methodology should also be developed for smokeless products. Without quantitative information on the absorption of various constituents under normal product use, it is impossible to assess comparative exposures and comparative risks.

The committee embraces the conclusions and recommendation on this issue set forth in the IOM (1994) report Growing Up Tobacco Free:

[T]he regulatory agency, as its first step, should develop a sound methodology for ascertaining the actual yields of nicotine, tar, or any other constituents of tobacco products, based on human consumption. Human exposure to some constituents of tobacco smoke can be assessed by use of biochemical markers of exposure to those constituents, although at present this methodology is technically difficult and still imprecise; however, it is likely that better exposure measures will be developed in the future. In any case, even the currently available measures of human exposure are likely to provide a better indicator of relative risk than do the standard cigarette smoking machine yields. At a minimum, the smoking machine tests can be modified to reflect the range of ways in which people actually smoke, including numbers of puffs and blocking of ventilation holes, to determine likely ranges of delivery. In addition, the manufacturers of tobacco products could be directed to submit information to the regulatory agency regarding the actual yields of their products in humans for particular brands of tobacco products, based on use of prescribed protocols.

A variety of regulatory initiatives relating to tar and nicotine yields can be envisioned. At a minimum, the regulatory agency should take steps to assure that consumers are informed about the meaning of statements regarding tar and nicotine yields, about the behavioral influences on exposure, and about the relative importance of the characteristics of the cigarette and the way it is smoked. The agency might also require that tar and nicotine yields be presented in a standard format, such as absolute content together with a statement of the range of expected systemic yield determined according to number of puffs or other behavioral factors. Currently, marketed cigarettes typically contain 8–9 mg nicotine in the tobacco rod, and have an expected actual yield to the smoker of 0.5–3.0 mg nicotine. Stating the nicotine content of the tobacco contained in the cigarettes is important because the content reflects the maximum possible yield, and reduction of content would be expected to result in a reduction in actual yield. Manufacturers might be required to convey this information on the package or through package inserts. In order to avoid misunderstanding, the agency might require consumers to be told that small differences in nominal yield do not reflect significant differences in health risks. Regulations of this nature will improve risk perception among consumers, and will correct any misleading impression about the relative hazards of cigarettes containing different levels of tar and nicotine. From the same perspective, the agency should be authorized to ban or regulate use of misleading terms (such as “light”) in advertising or on packaging, and should be authorized to require the use of standard terms.

Principle 3: Toxicity Testing

Manufacturers of all PREPs should be required to conduct appropriate toxicological testing in preclinical laboratory and animal models and appropriate clinical testing in humans to support the health-related claims associated with each product and to disclose the methods and results of such testing to the regulatory agency.

Under the regulatory arrangements recommended by the committee, tobacco manufacturers would have to obtain appropriate toxicological data from the scientific literature or conduct toxicological testing in connection with any new product or modification of an existing product (1) to support a claim of reduced exposure or reduced risk (see principle 5), (2) to demonstrate that a new ingredient does not increase the product risk (see principle 7), or (3) to satisfy the requirements of the added-ingredients review described in principle 8.

Because clinical trials on the long-term health risks of individual ingredients in tobacco-related PREPs cannot reasonably be conducted in humans, any scientific assessment of the risks, except perhaps low birthweight and sudden cardiac death, of added ingredients must necessarily be based on animal studies, short-term human studies, and epidemiological studies. For ingredients that have not previously been used in conventional or modified tobacco products, animal studies to support their safety should be conducted by the manufacturer prior to human exposure to those ingredients. The committee believes that the practice of introducing new ingredients into smoked products without full and adequate testing of such ingredients, as judged by a competent regulatory agency, poses a substantial risk to the consumer and should cease. A requirement for adequate preclinical testing of potentially toxic ingredients is a basic feature of all regulatory laws relating to drugs, biologics, devices, food additives, toxic chemicals, and environmental contaminants and should apply as well to conventional and modified tobacco products.

Principle 4: Premarket Approval of Claims

Manufacturers should be permitted to market tobacco-related products with exposure reduction or risk reduction claims only after agency approval based on scientific evidence (a) that the product substantially reduces exposure to one or more tobacco toxicants and (b) if a risk reduction claim is made, that the product can reasonably be expected to reduce the risk of one or more specific diseases or other adverse health effects, compared with whatever benchmark product the agency requires to be stated in the labeling. The “substantial reduction” in exposure should be sufficiently large that independent scientific experts would anticipate finding a measurable reduction in morbidity and/or mortality in subsequent clinical or epidemiological studies.

Products accompanied by explicit or implicit claims of reduced exposure to tobacco toxicants or reduced risk of adverse health effects should be subject to premarket approval of the specific claim. Under this approach, manufacturers would be permitted to market products with claims of reduced exposure or reduced risk (compared with whatever benchmark product is stated in the labeling) as long as the claim is approved on the basis of appropriate scientific data by the regulatory agency before marketing and, where judged necessary for risk reduction claims, appropriate plans for postmarketing surveillance are also approved.

When the committee refers to a “reduced risk” claim in this chapter, it is referring to any statement, however qualified, indicating that use of the product presents or may present a reduced risk of disease or other adverse health effect, compared with use of some other tobacco product.

Premarket approval of tobacco-related PREPs with claims of reduced exposure or reduced risk is essential to ensure that full information is presented to potential consumers. Under the current regulatory system, manufacturers may, in the absence of drug claims, introduce new or modified tobacco products into the marketplace without any prior regulatory review, subject only to the FTC's authority to enjoin “deceptive acts and practices” after the deception has already occurred. The result has been the long history, cited elsewhere in this report, of promotional practices by tobacco manufacturers that have, at the least, obscured the adverse health consequences of tobacco use. The committee considers it highly unlikely that tobacco-related PREPs marketed under this system would be labeled and promoted with appropriate caution, especially if the market becomes competitive.

The premarket approval model used for the review of new drugs, while not now applied to new brands and modifications of conventional tobacco products marketed without disease claims, in fact provides the only fail-safe procedure for the review of exposure reduction or risk reduction claims relating to tobacco products, whether made on behalf of novel products like Eclipse or in connection with products already on the market (e.g., smokeless products). The committee thus recommends that any new legislation must provide for premarket approval of claims of exposure reduction or risk reduction. The medical device provisions of the FDCA provide a model for this policy in that high-risk products are subject to premarket approval, while products of lesser risk are subject only to premarket notification. The marketing of a new tobacco-related product with an unsubstantiated claim of reduced disease risk (including a claim of reduced exposure that implies a reduced disease risk) is probably a greater threat to public and personal health than the marketing of a product already known to be dangerous, unless the claim is proven to be accurate. The scientific evidence behind such a claim thus deserves full and independent scrutiny.

The committee judges that exposure reduction and risk reduction claims for tobacco-related PREPs should be treated as analogous to “effectiveness” claims for drugs and devices. Just like drug claims, they are intended to assure consumers of a potential health benefit if the product is used as directed, and just like drug claims, they should be presented accurately and supported by appropriate scientific data. As discussed in Section II of this report, the committee is not aware of a sound scientific basis at the present time for an unqualified risk reduction claim for any tobacco-related PREP. Although the potential for reduced-risk products exists, data supporting such claims have never been subject to comprehensive, independent scientific scrutiny or regulatory review. In no case has the degree of potential risk reduction been estimated from animal or human studies, nor have surveillance programs been implemented to monitor the long-term outcome of the use of any of these products. Uncertainty about the health effects of all PREPs will therefore continue for years even if an effective regulatory framework is created in the near term. If no such framework is created, consumers will continue to act on deficient information.

With respect to the tobacco-related PREPs known to the committee, the most that can reasonably be said today on the basis of the publicly available data supplied by manufacturers is that these products reduce exposure to some but not all of the toxic ingredients in cigarettes, assuming no increase in the total number of cigarettes smoked or in other forms of compensatory smoking. Direct evidence will not be available for many years, if ever, to prove that the degree of exposure reduction achieved, if maintained over years, is sufficient to result in a significant reduction in morbidity and mortality. The use of reduced-nitrosamine tobacco, for example, seems likely to reduce the exposure of smokers to nitrosamines, but whether this will reduce lung cancer rates, given the other carcinogens in tobacco smoke, is unknown (see Chapter 12). Similarly, the use of the cigarette-like product Eclipse may reduce exposure to carcinogens but increase exposure to CO. Whether these and similar products, if smoked chronically like cigarettes, will provide a net benefit in terms of all-cause mortality is unknown and can only be determined by long-term epidemiological studies.

Products that increase exposure to one or more tobacco toxicants while reducing exposure to others will present a complex challenge to the regulatory agency. At a minimum, approved claims of reduced exposure should be accompanied by warning statements regarding any increased exposure. Moreover, in the committee's view, reduced-risk claims should be permitted for such a product only if the agency finds, based on scientific evidence, that the use of the product can reasonably be expected to reduce the overall risks of tobacco-related disease or death. While recognizing that the proposed premarket approval requirement raises the possibility of regulatory delays and may inhibit the marketing of new tobacco-related PREPs, the committee believes that undue delay can be minimized by an efficiently administered agency for the following reasons:

• The body of scientific information submitted for review of each new product is likely to be substantially less than that in a typical new drug application to the FDA. For most such products the scientific evidence supporting exposure reduction claims will come from in vitro studies, animal studies, and pharmacokinetic studies in humans. Depending upon the claim, evidence may also come from short-term clinical studies of pulmonary function and/or cardiovascular function. The huge clinical trial programs required to show the effectiveness and safety of drugs in humans would not be necessary to support exposure reduction claims for tobacco-related PREPs.
• Only a handful of tobacco-related PREPs are currently known to be under development, suggesting a relatively low volume of new applications compared with the hundreds of applications for drugs and devices that are handled annually by FDA.
• The success at FDA of the user fee mechanism, in combination with a budgetary appropriation, in providing resources to support an adequate staff suggests a model of public-private funding for the review of claims for tobacco-related PREPs.

Principle 5: Labeling, Advertising, and Promotion

The labeling, advertising, and promotion of all tobacco-related products with exposure reduction or risk reduction claims must be carefully regulated under a “not false or misleading” standard, with the burden for proof of the claim resting on the manufacturer, not the government. The responsible agency should have the authority and resources to conduct surveys of consumer perceptions relating to these claims.

Labeling requirements and restrictions on advertising and other promotional activity are core elements of product safety regulation. In a previous report focusing on preventing initiation of tobacco use (IOM, 1994), another IOM committee recommended that warnings concerning the risks of using tobacco products be strengthened and made more salient, that advertising be restricted to a text-only format, and that many types of promotional activity be severely curtailed. These measures are still needed to reduce youth uptake, to promote cessation, and—most pertinent to this committee's charge—to ensure that public understanding of the dangers of tobacco use is not obscured or compromised by the marketing and promotion of a new generation of modified tobacco products purporting to be “safer” than conventional products. From this perspective, adequate regulatory authority to promote accurate public understanding of the health effects of tobacco-related PREPs is perhaps the single most important feature of effective regulation.

The burden of proving any claim of reduced exposure or reduced risk should rest on the manufacturer. The evidentiary base for such claims must meet contemporary scientific standards and be fully available for independent review by a responsible regulatory body. Independent review and replication of key findings—the “gold standard” for validating scientific assertions—is a compelling necessity for assessing health-related claims associated with tobacco products. Unfortunately, in today's regulatory environment for tobacco products, this proper burden of proof is reversed. Manufacturers are free to market new tobacco products with explicit or implied health claims until challenged by the FTC, which must then show in court, through expert testimony and consumer surveys, that a particular advertising claim is deceptive and not adequately substantiated. Regulation of claims through this process may take years and offers no chance for prompt correction of misleading advertising. The committee anticipates that under the current regulatory conditions, tobacco manufacturers are no more likely to present—accurately and with appropriate caveats, warnings, and uncertainties—what is known, and unknown, about the potential health effects of PREPs than they have been to describe accurately the risks of conventional cigarettes.

The committee therefore strongly recommends that new legislation be enacted to ensure that the labeling, advertising, and promotion of all tobacco-related products are carefully regulated so that exposure reduction and risk reduction claims are supported by adequate scientific evidence and are not false or misleading. Claims made in the product labeling should be subject to premarket approval by the regulatory agency, as explained in connection with principle 4. Although a routine requirement for prior approval of advertising might not be constitutionally permissible, the agency should have the authority to take prompt action against specific advertising or promotional campaigns or practices that are false or misleading. This authority should include the power to require corrective advertising.

The committee considered the alternative of recommending that the advertising and promotion of tobacco-related PREPs be included under the present system of regulation for tobacco products in general. For the reasons noted previously, however, the committee considers this system to be insufficient to the task. In the judgment of the committee, a harm reduction strategy using tobacco-related PREPs cannot be implemented successfully unless consumers are fully informed through accurate labeling and advertising about the health consequences of using all types of tobacco products and about the substantial gaps in scientific knowledge. The committee finds no justification for continuing the current situation in which pharmaceutical products intended to aid smoking cessation— the desired public health goal—must be labeled and promoted truthfully under a “not false or misleading” standard, while tobacco products claiming to reduce exposure or risk—at best a partial and uncertain step toward disease reduction—can be promoted under generic consumer deception laws, with the burden of proof on the FTC to police violations after the fact. Unless this imbalance in the regulation of labeling and advertising is corrected, the committee finds it likely that aggressive promotion of tobacco-related PREPs, accompanied by incomplete and possibly inaccurate or misleading health-related claims, may well undermine other public health efforts to promote cessation of smoking and other forms of tobacco use. In short, unless the regulation of labeling and advertising of all tobacco-related products with health claims can be brought to parity, the impending availability of tobacco-related PREPs could cause more harm than good.

As a corollary, the committee concludes that the evidentiary base for regulating labeling and advertising must include not only scientific studies related to toxicology and human biology but also scientific analyses of the impact of particular claims, as well as messages and images in advertising and promotional activity, on risk perception and risk communication. Such analyses should take into account the overall public health goal of reducing tobacco-related morbidity and mortality both in individuals and in the population as a whole as well as the goal of enhancing the capacity of each adult to make an informed decision concerning the use of tobacco. The regulatory agency should have the resources and funding to conduct relevant consumer surveys and analyses as well as the authority to require such studies from manufacturers.

The committee has aimed to balance two possible types of errors: (1) allowing reduced-risk claims that turn out to be erroneous, and (2) deterring the development of modified products that would reduce exposure to known toxicants (and that would eventually be shown to reduce disease risk) if they were marketed. The committee acknowledges that manufacturers might forego the necessary investment in product development because of doubts that the accuracy of their claims can be defended successfully to unduly skeptical regulators.

However, the most important considerations are unproven or premature reduced-risk claims and reduced-exposure claims that are mistakenly interpreted as reduced-risk claims. The regulatory process should not discourage or impede scientifically grounded claims of reduced exposure, as along as steps are taken to ensure that consumers are not misled into believing (in the absence of sound evidence) that smoking the modified product is (or is likely to be) less hazardous than smoking the conventional product. How the complex of claims and caveats associated with PREPs can best be articulated in labeling is one of the major challenges facing the regulatory agency. On the one hand, the public health is not well served by the continued use of poorly defined terms such as “light,” “low tar,” or other phrases that imply a benefit when none has been proven to exist. On the other hand, neither is the public health served if smokers are discouraged by unduly cautionary language from using a new product with the potential for real risk reduction. The problem of conveying balance in communicating health benefits and risks is not unique to tobacco-related PREPs, and the large body of experience in other areas of health and safety regulation may be applicable to these products as well.

The agency will have to direct its attention to the language used as well as the labeling format. Some illustrations, based on existing formats, follow:

• Current cigarette labeling contains warnings that smoking causes lung cancer, heart disease, and emphysema and may cause birth defects. If warranted by scientific evidence, such warnings could be accompanied by a statement that the modified product might carry a reduced risk of one or more of these conditions.
• Current food labeling has on each package a table displaying a quantitative analysis of nutritional content. This approach could be applied to selected ingredients and constituents of tobacco products or tobacco smoke, such as known carcinogens and CO.
• Food-labeling tables express nutritional content not only as grams per serving but also as a percentage of average daily intake. One could envision tobacco product labeling with a similar table that shows exposure or yield ranges for particular toxicants or perhaps ranks the levels of exposure or yield as high, average, or low. These terms could be specifically defined and would perhaps be less misleading than such terms as “light.”
• One could also envision the use of words such as “high,” “average,” or “low,” again carefully defined, in a “risk” column in such a table. Pictograms, such as those that appear in poison control warnings, or icons might be used instead of words. It is essential that such labeling in the end be perceived as denoting degrees of risk, not as signifying or implying safety. The message that cessation is the only safe choice must not be obscured or lost.
• The agency should also consider requiring that labels for PREPs that make exposure reduction claims disclose that the reduction in exposure depends upon the user not compensating for the reduction by increasing use or by inhaling more deeply. Consideration should also be given to a disclosure that the health benefits of the product's exposure reduction have not yet been established in scientifically recognized tests or ongoing studies. Such a disclosure would guard against consumer confusion that risk reduction benefits have been proven. Furthermore, such a disclosure would provide an incentive to manufacturers to do more research on the health effects of exposure reduction. The FTC should consider requiring similar disclosures under its existing authority if new legislation is not adopted.

The committee does not have recommendations on the specific form of labeling that is optimal for tobacco-related PREPs. (The formats cited in the foregoing examples are meant only to be illustrative.) This topic deserves the sustained attention of experts in communication and marketing, with the intent of identifying the optimal approach for conveying in a simple format specific information on exposure reduction, the potential for risk reduction, and the associated uncertainty. This information must be balanced, accurate, informative, and not misleading.

Growing Up Tobacco Free (IOM, 1994) reviewed the history of mandated tobacco health warnings; called attention to the inadequacy of the existing warnings (both in content and format); reviewed recent initiatives in other countries, especially Canada; and recommended a number of legislative measures to increase the salience and effectiveness of health warnings on packaging and in advertising. Specifically, the IOM (1994) report made the following recommendations:

Based on current knowledge, Congress should enact specific warning and format requirements now, and should delegate regulatory responsibility for future modifications to the secretary of health and human services. The secretary should ensure that ongoing research is conducted on the effectiveness of prescribed warnings. Congress lacks the institutional expertise and flexibility to monitor the efficacy of regulatory innovations and to respond to new information. Authority to amend the warnings should be delegated to an appropriate regulatory agency. The secretary should be empowered to modify the format or content of existing warnings and to prescribe additional warnings, whenever such action is reasonably necessary to achieve effective communication of significant information regarding the health consequences of tobacco use or to discourage consumption of tobacco products.

The committee endorses this recommendation. The need to implement this approach is all the more evident now in light of the emerging market for products that purport to reduce the risk of tobacco use.

One of the recurrent questions in the regulation of health claims is when to permit a claim that a product may reduce the risk of disease. Such a statement may be misleading even though it is not false. Assuming substantial proof of exposure reduction, what type of evidence must be produced in support of a claim that the product may, as a result, reduce the risk of disease? Is inference based solely on the existence of a dose-response relationship sufficient? How strong must the evidence be? What is the significance of counterevidence? How much weight should be given to favorable changes in biomarkers that are biologically rational but not yet proven to be true surrogate end points? Although these are questions that will have to be addressed by the regulatory agency in due course, the committee itself is disinclined to permit claims that the product “may” or “might” reduce the risk of disease (or that “growing evidence suggests that the risk is reduced”) unless, as judged by the regulatory agency after review of data by an independent scientific advisory committee, the degree of exposure reduction is likely to result in a meaningful decrease in tobacco-related morbidity or mortality that can be measured in subsequent epidemiological studies.

Another question must then be faced. Assume that the agency finds insufficient evidence under such a demanding standard to support a claim that the exposure-reducing product “may reduce the risk” of a particular disease. Should the manufacturer be permitted to make the statement (which is not false) as along as it is accompanied by a disclaimer to the effect that the regulatory agency has not approved the statement? This approach is receiving increasing attention in other regulatory contexts in light of a few judicial decisions holding that regulatory agencies must allow manufacturers to make the unproven statement, together with a disclaimer, under the First Amendment (Gilhooley, 2000). The committee assumes that the First Amendment does not require such a permissive approach and is inclined to be much more cautious in light of the history of public health mistakes in tobacco product innovation. In the committee's judgment, in the absence of sufficient scientific evidence to support a claim that a product may reduce the risk of disease, any claim of reduced exposure to tobacco toxicants should be accompanied by a statement that the health consequences of the change are unproven (or unknown).

Given that considerable time will be needed to enact appropriate legislation, to staff a new regulatory agency, and to bring whole classes of products into a new regulatory system, a number of modified tobacco products with risk-reduction claims may enter the marketplace in coming years without regulatory review, assuming they are positioned not to fall under the drug or device laws. The committee views this situation with alarm but cannot offer a ready solution. Perhaps tobacco manufacturers would be willing to demonstrate their good faith by agreeing to voluntarily submit claims of reduced exposure or reduced harm to FDA, FTC, CDC, or some other appropriate agency for its review and comment, and to conform to agency suggestions pending legislative action. The committee emphasizes, however, that any purported harm-reducing products that have been put on the market before enactment of regulatory legislation should not be grandfathered or exempted from rigorous scientific and regulatory scrutiny.

Principle 6: Postmarketing Surveillance

The regulatory agency should be empowered to require manufacturers of all products marketed with claims of reduced risk of tobacco-related disease to conduct postmarketing surveillance and epidemiological studies as necessary to determine the short-term behavioral and long-term health consequences of using their products and to permit continuing review of the accuracy of their claims.

The committee has repeatedly emphasized that since definitive information on the ultimate value of a PREP cannot be determined at the time of its initial marketing, the true extent of its long-term benefit (or possible harm) will remain uncertain unless specific surveillance and epidemiological studies are conducted after the product is marketed. The FDA already has the authority to require pharmaceutical manufacturers to conduct such studies for the purpose of monitoring drug-related health risks. The designated regulatory agency for tobacco-related PREPs should have equivalent authority to require manufacturers who seek to make reduced-risk claims to sponsor surveillance and epidemiological studies in appropriate cases. These studies should be designed to assess not only the benefit of the product in reducing the risk of smoking but also the degree to which the product has served to maintain as smokers those who might otherwise have quit and/or to recruit new smokers. The quality of these studies should be high so that the results can permit modification of claims as knowledge evolves.

The committee envisions that a plan or protocol for such a study (or studies) would commonly be required at the time of approval of the claim. Whether a postmarketing study is necessary should be judged by the regulatory agency on a case-by-case basis. The agency can also be expected to take steps to ensure the scientific integrity of such studies by requiring appropriate monitoring and, where necessary, an independent safety monitoring board. In addition to the authority to require manufacturers to sponsor postmarketing studies, the regulatory agency should also have in-house expertise in epidemiology. Such expertise is needed to deal competently with scientists in industry and peers in the field, to operate an adverse event reporting system, and to sponsor selected studies when necessary. As discussed in Chapter 6, the regulatory agency, in collaboration with other public health agencies, should conduct an active program of surveillance for all tobacco products, including conventional ones, modified products that are not accompanied by health claims, and PREPs. It is particularly important for public health authorities to monitor patterns of initiation and progression of use, including the interactions of product use (substitution and supplementation), for all PREPs.

Principle 7: “No Increased Risk” Threshold for All Tobacco Products

In the absence of any claim of reduced exposure or reduced risk, manufacturers of tobacco products should be permitted to market new products or modify existing products without prior approval of the regulatory agency after informing the agency of the composition of the product and upon certifying that the product could not reasonably be expected to increase the risk of cancer, heart disease, pulmonary disease, adverse reproductive effects, or other adverse health effects, compared to similar conventional tobacco products, as judged on the basis of the most current toxicological and epidemiological information.

Conventional tobacco products are well established in the marketplace, and the committee recognizes that, regardless of the adverse health consequences of using these products, this situation is likely to continue for the foreseeable future. These products are already subject to a limited regulatory system that includes certain health warnings and other labeling requirements and some restrictions on advertising and promotion. The committee considered whether modified versions of established tobacco products, or new brands, that are marketed without health claims should be required, on public health grounds, to satisfy any new regulatory requirements and, if so, whether manufacturers should continue to be permitted to put such conventional products on the market without prior agency approval.

The committee concludes that any new regulatory system should ensure that all newly marketed tobacco products pose no increase in health risks, and ideally a new system should move products toward reduced potential for harm. To this end, the committee recommends that manufacturers be permitted to introduce a new product or make significant changes to an already marketed conventional tobacco product only if, on the basis of currently known toxicological and epidemiological information, the product could not reasonably be expected to increase the risk of disease. Implementation of this regulatory principle requires the definition of a standard for comparison. Although the responsibility for developing such a standard will rest with the regulatory agency, the committee can envision several possibilities: the most popular brand, the most (or least) harmful brand on specified dimensions, a toxicity profile that is representative of a sales-weighted average brand, or a standardized product of known composition. Because toxicant yields do not necessarily reflect health effects and because different standards may be appropriate for different products, the establishment of standards or reference products may require the type of information that would evolve from the regulatory agency's experience with the disclosure of ingredients in individual products (principle 1) and the added-ingredients review (principle 8). Such standards might be promulgated through guidelines and performance standards (principle 9).

To implement this approach, manufacturers should be required, before marketing a product, to submit ingredient information to the agency and to certify that the criterion of no increased risk is met (premarket notification). Whatever comparative standard is adopted for the product, the committee believes that a judgment of no increased risk should be based on the most current toxicological and epidemiological data. Manufacturers should not use the most dangerous products of the past as an appropriate standard for any new product. Consumers should rightfully expect that all new tobacco products, even those marketed without health claims, are at least no more hazardous than similar contemporaneously marketed products.

The committee considered whether premarket approval by the regulatory agency should be required for modifications and new brands of conventional tobacco products without claims of reduced exposure or reduced risk, and rejected this approach in favor of premarket notification and certification by the manufacturer. Although some product modifications may be undertaken to improve taste or other features linked to consumer satisfaction, future innovation in the tobacco market in this country is likely to focus on changes that purport to reduce risk. Because the main regulatory purpose of a requirement for no increased risk is to help establish the baseline for future comparisons and to stimulate development and use of protocols for tobacco product risk assessment, premarket notification is sufficient. The committee also did not want to see the regulatory agency burdened with a high volume of premarket approval decisions for products that have no claims of and no potential for improving the public health. Nevertheless, the regulatory agency should have the authority to seek the removal of a product from the market in the event that the “no increased risk” standard is not met.

Principle 8: Added-Ingredient Review

All added ingredients in tobacco products, including those already on the market, should be reported to the agency and be subject to a comprehensive toxicological review.

The ingredients added to tobacco products should be subject to a review that is similar to the FDA review of food additives conducted more than two decades ago to determine if those additives were generally recognized as safe and not subject to additional regulation. Although many of the ingredients added to tobacco products are generally recognized as safe when used as food additives, such an understanding may not apply when these substances are combusted and/or inhaled in smoke. A review of added ingredients would establish a baseline of knowledge about the ingredients, including chemicals, papers, and filters, as well as the toxicological testing that has been conducted on them; it would thereby facilitate an informed appraisal of the potential effects of product modifications. This review should be conducted by independent panels of experts reporting to the regulatory agency, with the objective of identifying those ingredients that add no significant toxicity to tobacco products and therefore can be considered safe in the context of this use. Sufficient knowledge of the toxicity of many of the constituents and ingredients of tobacco products may already be available (see Chapter 10) to begin this review. If adequate information is not yet available, the regulatory agency should have the authority to require the in vitro and animal testing that is judged necessary by the expert panels conducting the review. Such a review would begin to rationalize the risk assessment of ingredients in tobacco products. It would also open this field to the same scientific discourse that is now applied to food additives, environmental contaminants, and toxic chemicals, substances that are no longer introduced into use without independent scientific scrutiny or regulatory review. Tobacco products are the last remaining legally marketed toxic products in our marketplace that are tolerated without such review. A review of added ingredients would provide a scientific basis for guidelines or performance standards to limit possible toxicants like pesticides or filter fibers, if such limits are judged necessary for health reasons. It would also facilitate the review of new products by establishing conditions under which added ingredients, such as flavoring agents and papers, would be considered safe for use without additional data or review.

An added-ingredients review would be a large administrative undertaking requiring several years of effort. It would also require the cooperation of the tobacco industry in submitting data. The committee believes that such a review would be in the interests of both manufacturers and consumers. It would permit credible scientific judgment regarding the extent to which nontobacco ingredients in tobacco products contribute to overall toxicity, would build scientific bridges between the tobacco industry and the toxicology community at large, and as noted above, would facilitate the regulatory review of new products. The committee notes that the review of food additives has been conducted without public disclosure of trade secrets, such as the composition of spices and flavoring agents, and believes that this principle should be respected in any review of the ingredients added to tobacco products.

Principle 9: Performance Standards

The regulatory agency should be empowered to set performance standards (e.g., maximum levels of toxicants; definitions of terms such as “low tar”) for all tobacco products, whether conventional or modified, or for classes of products.

Performance standards, including maximum permissible levels, are promulgated under many regulatory laws, including those related to drugs, medical devices, and environmental pollutants. As noted above, successful implementation of a harm reduction strategy as an element of the nation's tobacco policy will require proactive government efforts in research, education, and surveillance as well as regulation of specific products. Accordingly, as scientific knowledge evolves regarding product risks and consumer preferences, the regulatory authority should be empowered to require product modifications to eliminate unreasonable risks. (A similar recommendation appears in the 1994 IOM report.) The committee has not attempted to draft precise statutory criteria to guide the agency's discretion in adopting performance standards. However, the committee does assume that the agency would not be empowered to ban nicotine from tobacco products.

Performance standards might take the form of limits on the concentrations of toxic ingredients in the product or the smoke; sets of limits that taken together would qualify a smoked product to be labeled as high, average, or low on a risk scale; or a list of reviewed ingredients that could be used without challenging the no increased-risk standard. A performance standard cannot be adopted without good scientific data, deliberate planning, and careful monitoring to ensure that it is achieving the desired goal. As the FTC test for tar and nicotine illustrates, even well intended performance standards can sometimes be subverted, with perverse and unintended health consequences. Performance standards aimed at setting definitions for terms must therefore be thought through with great care and be subject to change as experience is gained. Such standards will undoubtedly require a public rule-making process, meaning considerable time for their adoption.

Principle 10: Enforcement Powers

The regulatory agency should have enforcement powers commensurate with its public health mission, including the power to issue subpoenas.

The committee envisions that any regulatory agency for tobacco-related products would need the usual enforcement authorities conferred on public health regulatory agencies, such as the FDA, FTC, CPSC, and Environmental Protection Agency (EPA). The committee anticipates that this agency would also have an appropriate technical and legal staff concerned with issues of enforcement.

The committee specifically recommends that the agency have among its powers the authority to require the monitoring of scientific studies sponsored by manufacturers and to inspect manufacturers, investigators, and contractors to verify the data submitted in these studies. Such powers are necessary to ensure the quality and integrity of these studies. The agency should also have the authority to remove from the market ingredients or products that do not meet the test of no increased risk; to prohibit claims that are not supported by adequate scientific data; to seize products that are improperly labeled; to act promptly against advertising campaigns and promotional materials that are false or misleading; and to require corrective action.

Principle 11: Regulation of Drugs and Devices

Exposure reduction and risk reduction claims for drugs that are supported by appropriate scientific and clinical evidence should be allowed by the FDA.

The committee recommends no new legislation regarding the regulation of pharmaceutical products and medical devices intended to assist people to stop smoking. What will be needed is effective policy coordination between FDA regulators responsible for drugs and devices and whatever agency is charged with administering any new legislation related to tobacco products.

The committee also recommends that the FDA be prepared attitudinally and technically to approve both exposure reduction and risk reduction claims for drugs, when such claims are supported by appropriate scientific data. It is clear that the current FDA standard for all approved smoking cessation products is “quit rates” and that long-term use of pharmaceutical PREPs with cigarettes, which might be employed in a harm reduction strategy, is discouraged by the labeling approved for these products (Hughes et al., 1999). While the committee again emphasizes that cessation of smoking is the desired goal for all smokers, it also concludes that drugs with exposure reduction and risk reduction claims, if properly regulated and proved efficacious, have a place among the treatment modalities that should be available to current smokers. The committee judges, for example, that statements such as “helps recruitment into treatment clinics” or “reduces the use of cigarettes without compensation” are appropriate indications if supported by adequate data from clinical trials. The committee also concludes that for persons addicted to nicotine, a nicotine-containing drug product is preferable to a cigarette or other tobacco-containing product as a chronic source of nicotine. The FDA should therefore be prepared to consider the chronic administration of nicotine products as a reasonable exposure reduction strategy, again if supported by valid clinical data.

Transitional Regulation of Tobacco-Related PREPs as Drugs

Although the committee believes that adequate regulation of tobacco-related PREPs requires new legislation conforming to the preceding principles, the potential utility of the existing authority of the FDA and FTC should not be overlooked, and some of the committee's recommendations could be implemented by these agencies in the absence of new authority. The FDA presently has the authority to regulate any product, including new or existing tobacco products, as drug-delivery devices or as drugs, on the basis of an expressed or implied claim by the manufacturer that the product prevents disease by reducing the health risks of tobacco (Page, 1998). FDA can find that a manufacturer intends to make such a claim by looking at any relevant source, including advertising. See United States v. “Sudden Change” (United States v. “Sudden Change,” 1969) and National Nutritional Foods Assn. v. FDA (National Nutritional Foods Assn. v. FDA) (objective evidence). Claims that PREPs respond to concerns about smoking-related illnesses or cancer can be regarded as disease prevention claims.

In its recommendations for regulating the risk reduction claims for PREPs, this report identifies the type of testing and post-approval studies that should be conducted. The committee recommends that FDA use these testing standards under its existing authority in regulating PREPs that make claims for reduced risk of disease. FDA can regulate products that are both drugs and devices under its authority over medical devices. In the case of medical devices that require premarket approval, the agency can determine what testing needs to be done to provide a “reasonable assurance” of safety and effectiveness and can set performance standards for class II devices where appropriate (21 U.S.C.; Page, 1998). The products can be considered safe on a relative basis in light of their capacity for risk reduction, even though tobacco itself was not considered safe in FDA v. Brown & Williamson (Food and Drug Administration, et al. v. Brown & Williamson Tobacco Corporation, et al., 2000). Post-approval testing has been allowed and required for new drugs (Page, 1998).

Exposure reduction claims, such as low tar, have not been regulated by FDA as disease claims, and its authority to do so is uncertain (Page, 1998). A new law designed specifically to regulate tobacco-related PREPs can give a regulatory agency clear authority with respect to exposure-related and smokeless claims. Even for disease prevention claims, new legislation can confirm the agency's authority and the type of tests and manner in which the agency should regulate claims. Moreover, the legislation recommended by the committee would provide the agency new authority to obtain data on ingredients of all tobacco products, permit additional disclosures to consumers about toxicants in tobacco, give enhanced authority (and review when needed) over advertising and labeling claims for PREPs, guard against increased risks from new or modified tobacco products, and authorize a toxicology review and performance standards for tobacco products.

FDA's authority with respect to disease claims would continue if the new authority recommended under this report were given to FDA, with whatever modifications were made by statute. If the authority were given to a new agency, FDA's authority over disease claims for PREPs should be revoked only if the new agency has authority that is at least as extensive as FDA's present authority.