The principle of proportionality

    Photo: Milkos | Dreamstime

    How to regulate reduced-risk products

    By Clive Bates

    How should society regulate reduced-risk products like e-cigarettes, heated tobacco products, novel nicotine products or smokeless tobacco? These products challenge existing systems of tobacco regulation, which generally assume the products are an unqualified threat to be contained. In reality, the products represent a public health opportunity with relatively minor risks to be mitigated.

    If there is too little regulation, people may be harmed by dangerous products or fooled by misleading claims. But if there is too much regulation, the excessive compliance burdens will make perfectly good products ineffective or unviable, and small firms will exit the market. The U.K. Royal College of Physicians provides the best summary of this dilemma in its 2016 report, Nicotine Without Smoke: Tobacco Harm Reduction:

    “A risk-averse, precautionary approach to e-cigarette regulation can be proposed as a means of minimizing the risk of avoidable harm, e.g., exposure to toxins in e-cigarette vapor, renormalization, gateway progression to smoking, or other real or potential risks.
    However, if this approach also makes e-cigarettes less easily accessible, less palatable or [less] acceptable, more expensive, less consumer-friendly or pharmacologically less effective, or inhibits innovation and development of new and improved products, then it causes harm by perpetuating smoking. Getting this balance right is difficult.” (Section 12.10 page 187)

    The question then is: How do we get this balance right? How do we find a “sweet spot” between inadequate and excessive regulation?

    In its August 2018 report on e-cigarettes, the U.K. House of Commons Select Committee on Science and Technology provides some direction. This respected parliamentary committee calls for:

    “… a wider shift to a more risk-proportionate regulatory environment, where regulations, advertising rules and tax/duties reflect the evidence on the relative harms of the various e-cigarette and tobacco products available.”

    This is an elaboration of the “principle of proportionality,” an idea that underpins good policymaking in many jurisdictions. In the European Union, it is coded into the Treaty on European Union at Article 5.4. In the United States, the underlying concept can be found in longstanding Executive Orders (https://bit.ly/2zfCUhu and https://bit.ly/2KeFM2g) that govern regulatory planning and review. This is not really a revolutionary idea; it should be approached as part of normal good practice in policymaking.

    What would it mean for regulation? First, we would start with an overall objective.

    Many possible overarching objectives could be chosen, and many different goals are evident in the statements of public health advocates. These could include reducing the health consequences of nicotine use, reducing smoking or tobacco use, reducing nicotine use, protecting bystanders, preventing youth uptake, or even destroying the tobacco industry. For many years, it was possible for campaigners and governments just to say “all of the above” and duck any hard choices. This is because with cigarettes and other combustibles dominating the market, these goals seemed aligned and unified. But that is not the case now—many of these goals are now in conflict. If a regulator focuses on preventing nicotine use, they risk ending up with more cancer and heart disease because they have closed down reduced-risk nicotine products and made tobacco harm reduction more difficult. We may be concerned about youth uptake, but this is primarily because of later harm to health. In any situation where we give higher priority to a goal other than health outcomes, we implicitly accept that there are situations where we will accept more disease and premature death. It follows that the focus should be reducing disease—and this forms a basis for judging trade-offs between different objectives. A variation on this could be to pursue the greatest possible welfare—taking in enjoyment, stigma and distributional impacts of policies like taxation.

    Second, we should adopt a broad finding: The key issue for public health is not whether nicotine products are tobacco or nontobacco or whether they are novel or long established. The key distinction is combustion versus noncombustion. It is the inhalation of smoke that dominates the harms arising from tobacco use. The difference in risk between combustible and noncombustible products is inherent in the chemical and physical processes involved, and it allows for a much more liberal approach to regulation and risk assessment. For reduced-risk products, there is no case to follow the playbook for regulating products that are, by universal consensus, very dangerous.

    Third, we should consider some approaches to regulation that would be risk-proportionate for noncombustible products.

    Notification not authorization. The route to market for noncombustible products should require notification of a regulator not authorization by a regulator. This would permit access to the market if a product complies with appropriate standards but gives powers to regulators to intervene if there is a material concern about health or safety. In an authorization system, regulators have to approve thousands of products but find it notoriously difficult to say “yes” to anything to do with tobacco. An authorization regime can create a very narrow and costly bottleneck preventing useful reduced-risk products reaching the market.

    Focus on individual risk. The most relevant characteristic of a nicotine product for regulators is its impact on health and safety relative to other products, such as cigarettes, or by comparison with other common risks. However, some jurisdictions, including the United States, require assessments of “population effects” or how the pattern of use of consumers in the market changes as a result of introducing a new product. Some honesty is needed here: It ought to be acknowledged that this is completely unknowable in advance. Population effects are not even a characteristic of the product, but an emergent property of a complex system of thousands of interactions. Many of these have little to do with the product itself, for example, the innovation and pricing strategies of rival products. The effect of demanding answers to impossible to answer questions is predictable and damaging. So, for noncombustible products, the focus should be strictly on individual risk, with any concern about population effects addressed through postmarket surveillance and retrospective corrective action if needed.

    Product and production standards. It is important that manufacturers and importers have a rulebook to work from and that this is comprised of standards that are transparent and proportionate—allowing producers to gear up to meet commonly applied standards. These should cover electrical, thermal, mechanical and chemical safety; standardized testing regimes; labeling requirements; and supply chain quality assurance. It should be possible to set purity standards for ingredients and to blacklist or place limits on any problematic ingredients.

    Consumer risk information. Consumers should be empowered to make good product choices in their own interests and at their own expense. That means they need reliable information about risks. Health Canada has shown real leadership in proposing seven government-approved risk communication statements designed to inform consumers. These include, for example, “If you are a smoker, switching completely to vaping is a much less harmful option,” and, “Completely replacing your cigarette with a vaping product will significantly reduce your exposure to numerous toxic and cancer-causing substances.” The Canadian example shows how a public authority can take responsibility for consumer risk information and support informed consumer choice.

    Allow advertising of an adult product. If regulators prevent producers of reduced-risk products from advertising their products and building brands, then they are protecting the harmful incumbent products from entrants that can benefit health. At the same time, there are concerns about advertising reaching teenagers and recruiting new users. There is no perfect way to segregate audiences, but the pragmatic solution, widely adopted for alcohol advertising, is to place restrictions on content and placement of advertising rather than ban advertising outright.

    Let owners and managers decide policies on indoor use. The case for the state intervening to ban smoking rests on material risk to bystanders, especially workers. No such rationale has been established for vapor and heated-tobacco products, and it is unlikely to be. From a public health point of view, there is a strong rationale for treating vapers differently to smokers to encourage switching and to support switchers to remain smoke-free. This does not mean unconstrained use of these products would be permitted everywhere, but that the decision should rest with property owners to balance the needs of their customers and users. This would allow diverse and nuanced policies that are not possible under the terms of a mandatory ban. Public Health England has approached this issue by publishing guidance on setting vaping policies for property owners.

    Excise tax should be zero or limited and proportionate to risk. One of the greatest drivers of switching from combustible products to reduced-risk products has been the relative pricing. Taking on something new is certainly a lot more appealing if it saves money too. Tax policy is usually underpinned by three objectives: raising revenue, avoiding harmful distortions or achieving positive goals, and keeping tax transaction costs to a small fraction of the revenue raised. These objectives point toward keeping excise at zero on vapor products and ensuring heated-tobacco products and smokeless tobacco products are taxed at zero or a small fraction of the lowest rates for combustible tobacco products.