Bringing the flamethrower of reform to FDA’s Center for Tobacco Products

News has reached England of the appointment by President-elect Trump of Elon Musk and Vivek Ramaswamy to the “Department of Government Efficiency” (DOGE). We understand this will be a non-governmental entity that will advise the Office of Management and Budget (OMB -the efficiency engine of the White House) on reforms and efficiencies to be made in the federal government. Is there potential for some good to come from this? Regarding the FDA’s regulation of tobacco products by its Center for Tobacco Products, the answer is a resounding “yes”.

Bureaucracies are notoriously tenacious and will attempt to exhaust reformers with a thicket of detail, obtuse impediments, and a parade of “bleeding stumps”, the supposed victims of deregulation and downsizing. So, for orientation purposes, this note outlines a non-exhaustive overview of issues with the FDA’s regulation of tobacco and nicotine products and suggests efficiency reforms under each section.  I will update this over the coming days and weeks; feedback is welcome.

Summary: major efficiencies are essential in tobacco and nicotine regulation

The emergence of nicotine products with far lower risks than cigarettes presents a major opportunity to reduce the burden of death and disease caused by smoking. In the United States, the Food and Drug Administration is squandering that opportunity by creating insurmountable bureaucratic barriers to entry to the market for nearly all responsible manufacturers. The effect has been to consolidate the oligopoly of the tobacco industry, throttle innovative American businesses, and allow illicit Chinese-manufactured products to dominate the market, undercutting law-abiding manufacturers. Over 90% of the vaping products sold in the United States are NOT authorized by the FDA, and the market has become a chaotic mess.

With the vast majority of the market functioning without authorisation, the FDA plays no meaningful role as the guardian of public health or the public interest despite receiving $8 billion in tobacco user fees since 2009. Thankfully, Americans are not waiting for the FDA’s permission to protect their own health at their own expense. Smoking is at a record low and has almost disappeared among young people. Despite a moral panic in 2018-20, youth vaping has fallen by 72% since its peak in 2019.  Tobacco harm reduction is proceeding despite the FDA.

Yet, a complete liberalization of the market would not be optimal. FDA’s Center for Tobacco Products has the flexibility within the foundational Tobacco Control Act to radically lower barriers to entry and reduce the costs of regulation to itself and to applicants. FDA does not need additional funds; it needs to use the funds granted by Congress far more efficiently, to focus on its core regulatory purpose, and to withdraw from activities outside its legal mandate. 

Lack of overall coherent regulatory strategy

The Reagan Udall Foundation evaluation of the FDA’s Center for Tobacco Products highlighted the absence of a coherent regulatory strategy (cross-cutting recommendation 1) and the FDA’s arbitrary approach to the public health standard “appropriate for the protection of public health” (Recommendation 5c and 5d). Though the FDA subsequently published a regulatory strategy, “The CTP Strategic Plan”, this primarily focuses on process generalities and falls far short of the requirement for a coherent framework for evaluating and authorizing tobacco and nicotine products in the public interest.  The Tobacco Control Act grandfathered thousands of cigarette products, and they remain on the market with a relatively light touch regime for product updates (substantial equivalence). Today, 3,816 distinct cigarette varieties and 14,777 combustible products are on the market lawfully [FDA database, accessed November 2024].

In contrast, enormous barriers to entry face market entrants with reduced-risk products. Only eight vaping systems (34 products) have been authorized, mainly in one flavour (tobacco) with two menthol products [see current status], and all from tobacco companies with mass-market products. No pouch products have been authorized so far. Authorized vaping products account for less than 10% of the market, with the rest either wholly illicit or caught in an ambiguous legal twilight as they are illegal but subject to continuing review or with marketing denial orders stayed by the courts. In contrast, the European Union registry contains 300,000 authorized vaping products. Twenty-four mainly US-based academics and experts have proposed an updated strategy for the FDA (see the University of Michigan Tobacco Research Network – and full letter and briefing). 

Reform proposal: formulate a clear, transparent strategy for discharging the responsibilities vested in FDA/CTP by Congress and for focusing only on its legal responsibilities under the Tobacco Control Act.

Hostility to innovation

Congress designed the Tobacco Control Act (TCA) both to encourage innovation beneficial to health (as signaled in the 2001 Institute of Medicine report, Clearing the Smoke: assessing the scientific basis for tobacco harm reduction) but also to block deceptive innovations like “Light” and “Mild” cigarettes that appear to be safer but deliver just as much tar as regular cigarettes. The 2001 IoM report had excellent proposals for regulating nicotine in the US, with 11 principles (here – summarized by me here). It is reasonable for Congress to expect the FDA to sort the genuine pro-health harm-reduction innovation from misleading scam innovations and to promote the former and block the latter.

But the FDA is doing the opposite: for example, by approving the 22^nd Century low-nicotine cigarette while denying the Juul system on purported toxicology grounds and thousands of other low-risk nicotine products that millions of Americans use as alternatives to cigarettes (there are now 19 million vapers and about 30 million smokers). FDA is even boastful about doing this [for example, see here, here, here] even though it puts smokers and vapers at risk and destroys perfectly viable, pro-health businesses.

The application process is so cumbersome and lengthy that Americans only have access to authorized products approaching ten years old. American companies’ creative scientific, engineering, marketing, and legal capabilities are consumed in justifying products to the FDA that are already obsolete for consumers, providing a major advantage to consumer-orientated innovators in Shenzhen, China, and so supporting black market demand.

Reform proposal: FDA must demonstrate a capability to review and authorize new products within 180 days (already a requirement under TCA Section 910(c)(1)(a)) and ensure the process requirements and assessment regime make this feasible, which it currently is not.

FDA’s regime is only viable for big players

The primary factors predicting the FDA’s vape marketing authorizations to date are the company’s size and homogeneity of its products, not the “appropriate for the protection of public health” (APPH) standard. Many smaller companies have products that are used by adults as an alternative to smoking and would be APPH. Yet, these companies do not have the financial resources or sales volumes to prove APPH to the satisfaction of the FDA. Companies do not routinely disclose how much they spend on their Pre-market Tobacco Applications (PMTAs). Still, we know from legal filings (see Juul FTC filing from para 86-104) that Juul spent $100 million on its initial PMTA. PMTAs generally require years of effort, tens of millions of dollars, and considerable expertise (also here).

Very few companies and products have revenue streams that can absorb such large compliance costs. Through this size and compliance cost filter, the FDA exerts an aggressive anti-competitive chokehold on the market that will radically reduce the number of businesses and diversity of products lawfully available. The FDA’s approach also presents a significant advantage to tobacco companies – they can cross-subsidize their mass-market vaping product compliance costs from strong balance sheets built on cigarette profits. But if the vaping market falters, they sell more cigarettes. The high cost is not the only barrier to entry: the uncertainty of success and opaque success criteria also add significant risk to the costs. Finally, the barriers to entry stimulate illicit trade by suppliers that have not incurred these giant regulatory costs, undercutting legitimate businesses aiming for compliance. 

It’s worth recalling the original Regulatory Planning and Review Executive Order 12866 of 1993 and the requirement for cost-effectiveness in regulation.

(5) When an agency determines that a regulation is the best available method of achieving the regulatory objective, it shall design its regulations in the most cost-effective manner to achieve the regulatory objective. In doing so, each agency shall consider incentives for innovation, consistency, predictability, the costs of enforcement and compliance (to the government, regulated entities, and the public), flexibility, distributive impacts, and equity.

Reform proposal: FDA must cut the costs of the PMTA authorisation regime to applicants by at least 80% and provide an affordable fast-track for small and medium enterprises. It should aim for similar savings in its internal assessment costs. Costs of application should be benchmarked against the European Union.

Wildly inaccurate regulatory impact analysis

Usually, such gigantic and distorting barriers to entry would be picked up in the rulemaking process, including through scrutiny by OMB/OIRA. It is doubtful that the 2016 Deeming Rule that brought vaping under the TCA and FDA jurisdiction would have survived scrutiny if the supporting analysis had even approximately reflected the reality post-implementation. FDA’s 2016 Regulatory Impact Analysis (RIA) for its deeming rule put the cost of a PMTA for e-liquids at $131,643 (first application) and $117,486 (subsequent applications) – see Table 11b. FDA estimated average PMTA costs for devices at $466,563 (first application) and $192,654 (subsequent applications) – see Table 12b. These are likely to be underestimated by two orders of magnitude. FDA also grossly underestimated the number of applications it would receive. It has had 26 million PMTA applications for vaping products, but its RIA estimate was for up to 4,050 e-liquids and 747 ENDS devices – see Table 7.

Reform proposal: FDA or a reliable third party should repeat the Regulatory Impact Analysis but with numbers that reflect the post-implementation reality. With a new RIA, OMB/OIRA should determine if the 2016 rule should have gone ahead, and, with the benefit of hindsight, if it should be rescinded and replaced.

Opaque decision-making criteria

FDA has not been clear about how it will authorize or deny these products – the uncertainty adds a further burden and reason for caution to vaping companies and their investors. The PMTA guidance was unclear and late, the standards or evidence required remain unknown and have changed over time without notification, and the method of assessing APPH has never been set out. This approach to APPH is now the subject of extensive litigation under the Administrative Procedures Act. Many filings by capable mid-size vaping companies provide a compelling account of arbitrary and capricious regulatory conduct.

Somehow, the FDA authorized a cigarette (the 22^nd Century VLN product) but denied the much safer Juul product on toxicological grounds. If it can do something this extreme, it shows it can approve or deny anything it chooses. The APPH test balances a range of impacts on behaviour arising from launching a new product. Yet, the FDA has never described how this balancing calculation will be performed or what weight it would give to different behavioural changes. For example, how much weight should be given to preventing adolescents from vaping compared to causing adults to switch from smoking to vaping? FDA has even ruled out counting some benefits: for example, young people diverted from smoking to vaping. There is no legal or ethical basis for ignoring such benefits. 

Reform proposal: FDA should provide unambiguous guidance on how it will assess products and meet the “appropriate for the protection of public health” standard. This should clarify how it will make trade-offs, value disparate benefits and detriments, and factor in the long-term impacts of behavioural transitions. It does not do this at present.

An authorization system leads to excessive caution and politicization

A system that requires a regulator to authorize a product will tend to be excessively but asymmetrically cautious. It will over-emphasize the possible risks of approval but undervalue the lost opportunities arising from a denial, as these are never visible. FDA’s monitoring metrics show that by 28 February 2023, it had refused to accept 19.6m PMTA applications, refused to file 5.1m applications, and denied 1.3m applications. 

However, the FDA has no idea about the aggregate consequences of doing this, as they were all rejected for procedural application reasons, not substantive public health reasons. None of these products was shown to be inappropriate for the protection of public health. However, all were rejected because they couldn’t meet the FDA’s exceedingly high but fuzzy and moving standard to prove they were appropriate for the protection of public health against opaque criteria. The difference matters.

In addition, authorization regimes are wide open to political interference as activists can apply a range of political pressures to secure their preferred outcomes. The risk of politicization is especially acute when the decision-making is opaque (see above). One example is Congressman Raja Krishnamoorthi claiming credit for influencing the FDA Commissioner to secure a particular regulatory outcome or, as he puts it, “victory” (the denial of Juul’s PMTA). 

Reform proposal: the process must become as “mechanistic” as possible, based on transparent guidance and standards and leave as little as possible to opaque and politically-motivated determinations. FDA should subject its own regulatory practices to an “appropriate to the protection of public health” standard so that it can assure Congress that its interpretation of the statute is not doing more harm than good and confounding the purpose of the Act.

FDA adopts a simplistic view of youth risk behaviors

Nearly all adults agree that young people should not vape, smoke, take drugs, drive under the influence, join gangs, have sex prematurely, etc. But these are simply adult preferences. It is not a basis for policymaking, given we know a subset of young people engage in risky behaviors whether we like it or not. Therefore, regulators must consider the harmful unintended consequences of their actions and try to mitigate real-world risks to young people. A range of studies now suggests smoking and vaping are substitutes, that vaping displaces smoking for some young people, and that the most intensive teenage vapers are those who would otherwise have been smoking. Measures to prevent vaping may increase smoking, illicit trade, or promote workarounds.

It is not good enough to deny these effects matter if doing so triggers unintended adverse consequences. Further, it is impossible to simply divide adolescents’ interests from the adults in their lives: young people experience harm from smoking through the role-modelling influence of parents, loss of household income, illness and death of parents, and secondhand smoke exposure. FDA does not recognize the effects of this on public health. It is not possible for PMTA applicants to provide product-specific evidence about such effects, even if they have material public health implications.

See my briefing on youth vaping for a more complete analysis with sources.

Reform proposal: FDA should use a measure of health impact, such as discounted Quality Adjusted Life-Years, to shape its approach to trade-offs between the health imperative of adult smoking cessation and concern about youth vaping uptake. FDA should also include a counterfactual analysis within its approach to account for vaping among young people who would otherwise smoke. If it took these steps, it would focus entirely on reducing smoking among adults and young people who would otherwise smoke.

FDA avoids accountability for the aggregate effect of millions of decisions

FDA has so far only authorized a small number of vaping products; all but one system is tobacco-flavored, in total accounting for less than 10% of the tracked market. In 2018, the FDA issued an advanced notice of proposed rulemaking signalling its intent to ban all or most flavoured e-liquids besides tobacco and possibly some menthol flavours. Under Section 907 of the TCA, such rulemaking (a tobacco product standard) places the burden of demonstrating APPH on the FDA. The agency would find it extremely difficult to meet this burden as it would need to assess the impact of collapsing the vape market to a few products not used by most adult vapers and closing all the small vape shops that provide what amounts to community-based stop-smoking support. It would need to consider illicit trade, DIY flavour products, and numerous possible workarounds. It would also have to account for the likelihood that many adolescents who currently vape would migrate back to smoking.

It would not take much of an uptick in smoking to completely overwhelm any hoped-for benefits of reduced youth vaping. So, the FDA has not attempted to do this. Instead, it has developed such restrictive criteria for authorizing flavours at the individual product application (PMTA) level that the net aggregate effect is almost the same.  Though the FDA recently authorized one menthol-flavoured variety, this was justified using a standard few other companies could meet. It may have been a tactical concession to further the FDA’s defences in contentious litigation rather than the opening of a viable path to market for flavoured products.

The example above illustrates a generic problem. No single applicant can show that the denial of their particular product will have material market-wide effects (e.g., shutting down all vape stores or triggering mass relapse to smoking among vapers) because one product will not have that much effect when there are others to switch to. However, when all products of a particular type are denied, there may be significant market-wide negative effects for which no one in the FDA appears to be accountable or even interested in understanding.

Reform proposal: The FDA should take a full system view of its role, as it is this, not individual products, that affects the American public. That means considering aggregated impacts arising from the denial of thousands of similar applications, in which none of these issues can be considered at the level of the individual product. If FDA took this approach, it would be far more reluctant to grant such regulatory protections to the cigarette trade and such lavish advantages to the China-based illicit market. It would value more highly the pro-health innovative US industry providing far safer alternative to cigarettes.

Stealth rule-making

Without issuing a rule and going through the rule-making process, FDA can impose de facto standards through using a common practice across thousands of applications. Using the PMTA process as a form of stealth rulemaking, the FDA does not have to assess or justify the aggregated impact of millions of individual decisions, each justified without reference to any other. However, because the FDA applies the same restrictive criteria to every product, there will be adverse market-wide effects that no one is responsible for assessing. However, by using the PMTA process rather than rule-making, the FDA escapes the need to evaluate and justify these aggregate impacts.  This potentially unlawful de facto rulemaking has now entered litigation with the Fifth Circuit Court of Appeal taking a sceptical view (see Reynolds vs FDA – see A.3 on page 11).  The fact that it has authorized one menthol flavoured vape system does not invalidate the concern that it is operating a de facto ban in practice.

Reform proposal: FDA should justify its own practices as a “appropriate for the protection of public health” regardless of whether it is doing this for rule-making purposes.

Fatally wrong risk perceptions

The general public, consumers, the media, healthcare practitioners, regulators, and policymakers will base their judgments and decisions to some extent on their understanding of risks – for example, the relative risks of vaping versus smoking, smokeless tobacco versus cigarettes, and whether nicotine is harmful or addictive. The HINTS survey by the National Cancer Institute shows these risk perceptions to be wildly out of kilter with sound scientific understanding. For example, in HINTS 2022, only 1.6% correctly thought vaping was much less risky than smoking. However, 50% thought it was as harmful or more harmful, and 37.5% did not know what to think. HINTS 2017 found that 71% did not know that smokeless tobacco was less risky than cigarettes. HINTS 2019 found that most people (56%) incorrectly think nicotine is the agent causing most cancer associated with smoking, and a further 19% did not know. FDA is not, of course, uniquely responsible for these misperceptions. CDC and multiple sources and drivers of news flow have a major effect. However, the FDA/CTP spends $125 million annually on public education campaigns and $17 million on communications (2023, Report to Congress). Very little of the totality of this communications expenditure does anything to address risk misperceptions, and most of it exaggerates risks or reinforces risk misperceptions. FDA’s own efforts at risk communication are wholly inadequate and unlikely to improve public risk perceptions. See Lethal miscommunication: FDA fails to explain the relative risks of tobacco products. Though it has been responsible since 2016, FDA commissioned research in 2024 into risk communication that will take years to complete and is unlikely to lead to the candour required: for example, matching this public communication from the UK National Health Service:

Reform agenda: FDA and CDC should agree on a package of informative and actionable risk communications about non-combustible nicotine products and nicotine itself. The agencies should work hard to combat misinformation that, in some cases, they have promoted, such as false concerns about EVALI, gateway effects, popcorn lung etc. CDC should be the lead agency on risk communication. FDA is a regulator that should be implementing risk proportionate regulation.

A science program excessively orientated to problem-seeking and restrictive regulation

FDA-funded science is a more extensive subject than it is possible to address here. Much of the science supported by user fees paid to the FDA and channelled through NIH does not contribute helpfully to proportionate regulation or realistic risk perceptions. It often exaggerates risks, asserts spurious causal relationships such as gateway effects, and downplays or ignores public health opportunities. FDA’s budget for science (including the Office of Science in FDA) is $305 million (2023, Report to Congress). Research incentive structures and institutional culture likely lie behind the distortions in science that arise from having a regulator as the dominant funder of science.  The funding of science should be wholly under the control of research funding agencies, and they should prioritise the funding based on scientific and public value. The channelling of research funds through the FDA is distorting and wasteful, leading to hundreds of studies with questionable scientific merit and unspoken conflicts of interest. FDA has an internal need for scientific capability, and its science funding should be confined to meeting that need from internal or external sources.

Reform proposal: FDA’s tobacco science budget should be confined to that necessary to discharge its legal duties. Other tobacco and nicotine science should be justified and funded from the budget of NIH or other research-funding entities. An independent review of the quality of FDA/NIH-funded tobacco and nicotine science should be conducted, and centres producing poor quality or irrelevant science should be defunded.

Counterproductive enforcement of a dysfunctional regulatory regime

 There have been many calls for robust and aggressive enforcement. However, caution is in order because the authorization regime works so badly, and most of the products on the market are currently technically unlawful. The consequences of the dislocation caused by widescale enforcement have not been assessed and would likely be highly damaging to public health. In the near term, there should be careful prioritization of enforcement focused on criminal networks and products for which no PMTA application has ever been made. As the authorization regime becomes more rational and proportionate, it makes sense to eventually step up enforcement to cover all non-compliant products.

The critical strategy for enforcement is to radically reduce the volume of illegal activity by authorizing several thousand vaping and pouch products – something that will only be possible if the barriers to entry are radically lowered. Recall that there are 3,816 cigarette variants lawfully on the market today, so a call for several thousand vapes is not outlandish.

Reform proposal: FDA should prioritise a major effort to simplify its PMTA process and authorise several hundred or thousand of reduced-risk products. Enforcment activity in the meantime should focus on tackling emerging criminal networks that FDA’s regulatory burdens have unintentionally nurtured.

There is no case for an increase in user fees

 Like all bureaucracies, the FDA likes money – the more, the better, and the FDA is pitching for additional funding of $114 million per year from 2025. Under Section 919 of the TCA, the FDA has received just over US$8 billion in tobacco user fees since it was given jurisdiction under the TCA in 2009 and currently receives $712 million per year. This should be enough. Much of its budget is not spent wisely or on its core regulatory functions – for example, it spends heavily on public education and funds science beyond that needed to conduct its regulatory duties.  Much of the core regulatory budget is consumed on addressing a problem of the FDA’s own making: the astonishingly expensive process of bringing new nicotine products to market – this is burdensome for applicants and for the FDA’s own assessors. A simpler process would reduce costs and improve efficiency for both.

There is a case for a much harder look at how the FDA spends its money and why so little has been achieved for the US$8 billion spent since 2009. There is no case for adding to it.  There may be a case for redistributing the burden of paying users fees to ensure that all products authorized by the FDA pay and carry an appropriate share. Against that is the hugely asymmetric compliance costs faced by vape or heated tobacco products vendors that must take the PMTA pathway to market rather than the grandfathering and substantial equivalence applications that make it cheap and easy for cigarettes to be on the market. 

Reform proposal: FDA’s budget should decrease significantly by eliminating spending on activity outside its legal remit and mandate from Congress (notably on public education, which is CDC’s job, and science, which should be prioritised and funded through research funding bodies). Further savings can be made by greatly simplifying its PMTA process, as this will reduce costs to both applicants and the FDA’s cost of assessment. Costs of realigning CTP staff salaries to the (higher) agency norm should be met by cost savings as described above and headcount reductions to reflect a reduced need.

Process simplifications are possible and essential

The main barrier to progress in tobacco harm reduction is the requirement for reduced-risk tobacco and nicotine products to undergo Premarket Tobacco Applications (PMTAs) under the Tobacco Control Act section 910. More precisely, the barrier to entry is created by the FDA’s expensive interpretation of the requirements of this section of the Act, which goes far beyond what Congress has written on the face of the legislation. A radical simplification consistent with the principles outlined in the Act would: 

1. Product. Premarket assessment would be limited to the characteristics of the product itself. This would include meeting guidance for chemical, electrical, and thermal safety and ensuring packaging and marketing were responsible and not targeted at youth.  This is the only way to meet a 180-day target that Congress set for the completion of the assessment process.
2. Product improvements. A fast and efficient process is necessary to authorise product improvements, especially if these improve health and safety, or will help with smoking cessation.
3. Manufacturing. It could include GMP standards for manufacturers and importers.
4. Surveillance. Instead of including surveillance obligations in individual product authorizations, which is simultaneously incomplete and wastefully duplicative, the FDA should conduct surveillance of the whole market, including illicit trade and unauthorized products. This should form the basis for post-market corrective action and enforcement activity.
5. Population effects. Any adverse population effects should be picked up in post-market surveillance and be subject to corrective action, including recall and rescinding authorization. 
6. Risk communication. FDA and CDC should provide a range of “findings” about the absolute and relative risks of nicotine and tobacco products. These should be simple, actionable, truthful statements aimed at consumers, health professionals, media, politicians, and industry. 
7. Enforcement. The problem of illicit trade should be managed primarily by crowding out illegal commerce with lawful products and suppliers (i.e., hundreds or thousands more authorized products). Enforcement has not proved successful so far at the scale of the whole market.  FDA’s intense regulatory control does not apply to >90% of the products used by 19 million Americans.  

A longer-term regulatory framework

In the longer term, FDA needs new legislation that is not built around the “appropriate for the protection of public health” concept. Nicotine products are recreational psychoactive substances and not intended to protect public health. No one would regulate, say, beer in this way. Yet beer is regulated in many of the ways that would be appropriate for regulating nicotine: ingredients, chemical quality, microbiology, manufacturing process, age restrictions, packaging and labelling, marketing controls, taxation, etc. The goal is to make products that are widely used available with tolerable risks and appropriate controls.

Reform proposal: OMB or another suitable entity should commission research on future efficient and rational regulation of consumer nicotine products with a view to the Executive Branch presenting a Bill to Congress for consideration.