e-cigarette sellers must substantiate greater population-level safety to make "safer" claims

Nicopure Labs, LLC v. Food & Drug Admin., No. 17-5196 (D.C. Cir. Dec. 10, 2019)

In the Tobacco Control Act, Congress gave the FDA additional authority to regulate tobacco because previous measures “failed adequately to curb tobacco use by adolescents.” Congress made a lot of findings about the addictiveness of tobacco and its dangers to children. Instead of banning tobacco and suffering the resulting black market, Congress took the then-current market as a baseline for improvement. The TCA grandfathered products that were on the market as of February 15, 2007. New products require premarket authorization, requiring the FDA to assess their health effects on the population as a whole in view of both the “likelihood that existing users of tobacco products will stop using such products,” and the “likelihood that those who do not use tobacco products will start.” Authorization shall be denied if, among other things, there’s no showing that permitting the new product would be appropriate for public health; the proposed labeling is false or misleading; or the new product deviates from an existing tobacco product standard without adequate information to justify the deviation.

Anything marketed as safer than existing tobacco products (“modified risk” tobacco products) has to meet more stringent public-health standards. A modified risk tobacco product is a product whose “label, labeling, or advertising … represents explicitly or implicitly that … the tobacco product presents a lower risk of tobacco-related disease or is less harmful than one or more other commercially marketed tobacco products; … contains a reduced level of a substance or presents a reduced exposure to a substance; or … does not contain or is free of a substance”; or a product whose label, labeling, or advertising uses “light,” “mild,” or “low” or similar descriptors; or whose manufacturer “has taken any action directed to consumers through the media or otherwise, other than by means of the tobacco product’s label, labeling, or advertising . . . respecting the product that would be reasonably expected to result in consumers believing that the tobacco product or its smoke may present a lower risk of disease or is less harmful than one or more commercially marketed tobacco products, or presents a reduced exposure to, or does not contain or is free of, a substance or substances.” There’s a statutory exemption allowing the use of “smokeless tobacco,” “smoke-free,” and similar defined terms for chewing tobacco.

The marketing of a modified risk product must “enable the public to comprehend the information concerning modified risk and to understand the relative significance of such information in the context of total health and in relation to all of the diseases and health-related conditions associated with the use of tobacco products.” A product may be marketed as presenting a lower risk only if “the applicant has demonstrated that such product, as it is actually used by consumers,” will both significantly reduce harm/risk to individual tobacco users and benefit the health of the population as a whole, taking into account current users and current non-users. 

There’s a special rule with “a less demanding and more targeted standard” for the subset of modified risk products that purport to contain a reduced level or none of an identified substance (e.g., “no diacetyl”). Such products aren’t required to “significantly” reduce harm or risk to the individual user and must be only “expected” to benefit the health of the population as a whole. Also, the substance identified as reduced or absent must actually be harmful; the reduction must be substantial and accurate as labeled; the product must not expose the consumer to increased levels of other harmful substances; and consumer perception testing must show that consumers will not misinterpret a specific claim as an assurance of relative overall safety. A user of this rule must also “conduct postmarket surveillance and studies” and submit the results to the FDA annually to allow it to “determine the impact of the order on consumer perception, behavior, and health and to enable the Secretary to review the accuracy of the determinations on which the order was based.”

Smoking cessation products have to meet even more exacting standards for a new drug or device. No e-cigarette has yet sought and received clearance from the FDA under any of the three pathways. The industry didn’t challenge either the new tobacco product or new smoking cessation product approval pathways. It wanted to make health claims with fewer restraints than those afforded by the modified risk pathway.

Nicopure, an e-cigarette manufacturer and distributor, and an e-cigarette industry group, argued that the FDA violated the APA by not providing an easier premarket authorization pathway for e-cigarettes. It didn’t and I will say no more.  They also challenged two provisions of the Tobacco Control Act as violating the First Amendment: (1) the premarket review standards applicable to modified risk tobacco products allegedly impermissibly burdened truthful, nonmisleading statements about e-cigarettes and (2) the ban on distribution of free samples of tobacco products, including e-cigarettes allegedly suppressed expressive conduct. Both challenges failed.

Although the court mostly discusses the evidence in the context of the APA challenge, it’s clearly relevant that both the numbers of young users and adverse reactions to e-cigarettes are rising sharply. While the evidence is insufficient about whether e-cigarettes reduce conventional smoking—and some evidence suggests they’re a gateway for some users—the industry wasn’t seeking approval of e-cigarettes as smoking cessation products, nor was it instructing users on how to stop using nicotine.  “But e-cigarette manufacturers nonetheless have actively marketed their products as if they were a safer, healthier substitute for conventional cigarettes.” We just don’t yet know the long-term impact on the general population.

Note: the following discussion largely assumes some of the analytical moves, which makes it just like every other commercial speech case.  Key among them: Can the government decide that the meaning of “safer” is “safer for the population as a whole,” rather than “safer for you than smoking tobacco would be, holding constant your likelihood of doing that instead”?  That’s a far more significant move than the general substantiation requirement—which is in theory applicable to all commercial advertising subject to the FTC’s jurisdiction, and which is bolstered here by a well-known and appalling history of deadly industry lies. The court gets to finesse the “safer” definition question by focusing on the substantiation requirement.  And I have my doubts that the industry could substantiate that its “safer for you” claims, and most crucially their limitations, are understood by most reasonable consumers.  But in the theoretical situation that consumers did understand “this is safer for me than tobacco, but I could easily end up worse off by using it because I probably wouldn’t have smoked tobacco in the first place,” can the government ban that marketing? I think the answer is yes, especially given all the uncertainties, but that is the strongest case in conventional First Amendment terms.  (It thus matters a lot that this is not an as-applied challenge, since no one in the industry appears ready to substantiate that consumers receive that risk message.)

The court held: The FDA can constitutionally bar advertising of e-cigarettes as safer than existing products until that safety benefit has been shown. “That conclusion is amply supported by nicotine’s addictiveness, the complex health risks tobacco products pose, and a history of the public being misled by claims that certain tobacco products are safer, despite disclaimers and disclosures.” Congress found that “modified risk tobacco products may encourage new users to take up tobacco products, rather than simply reduce risk to those who already use them.” Congress cited an FTC study and found that advertisements that claim one tobacco product is less harmful than another mislead consumers, even when the putatively less risky products contain “disclosures and advisories intended to provide clarification.” It specifically found that disclaimers and other “[l]ess restrictive and less comprehensive approaches have not and will not be effective” in communicating risks associated with tobacco products sold as safer. It concluded that “the only way to effectively protect the public health from the dangers of unsubstantiated modified risk tobacco products is to empower the Food and Drug Administration to require that products that tobacco manufacturers s[ell] or distribute[] for risk reduction be reviewed in advance of marketing, and to require that the evidence relied on to support claims be fully verified.”

As an initial matter, the industry argued that the use of advertising to identify what counts as a modified risk product burdened speech in violation of the First Amendment.  Not so.  “First, our precedent explicitly approves the use of a product’s marketing and labeling to discern to which regulatory regime a product is subject, and to treat it as unlawful insofar as it is marketed under a different guise.”  As with drug claims versus structure/function claims, or even whether the FDA has any jurisdiction at all, how a product is marketed can tell you what it is and therefore what regime applies to it. “Just as the government may consider speech that markets a copper bracelet as an arthritis cure or a beach ball as a lifesaving flotation device in order to subject the item to appropriate regulation, so, too, the FDA may rely on e-cigarette labeling and other marketing claims in order to subject e-cigarettes to appropriate regulation.”

The industry wanted to pitch e-cigarettes as safer, arguing that this would help current smokers who “routinely seek information that would be helpful when attempting to move away from cigarettes and learn more about the features of particular vapor products.”  But it wanted to do so without scrutiny based on public health and without addressing the risk of greater uptake by current nonsmokers. It argued that the FDA’s modified risk pathway regulated the message itself, not the product.  But that was wordplay. “Deliberately selling an e-cigarette as less risky without going through the requisite regulatory review for reduced-risk tobacco products renders the sale-as-labeled unlawful, just as selling saw palmetto extract as a drug without FDA premarket approval was unlawful. It is well established that ‘commercial speech related to illegal activity’ is not subject to constitutional protection.”

Even viewing the modified risk pathway as burdening speech, it was a legitimate restriction on commercial speech. Manufacturers could make accurate “less risky” claims, but only if substantiated with evidence of overall public health effects and with evidence that consumers wouldn’t be misled. “If a manufacturer shows its product is in fact safer, and shows that consumer perception accurately grasps the nature and limits of any safety claim, the product will be marketable. Because the Act withholds from market only those tobacco product claims that, upon review, are found to be misleading, it bars only commercial speech that by definition is unprotected by the First Amendment.”

Under Central Hudson, the government had a substantial interest in “ensuring that any modified risk statements are accurate and non-misleading in order to protect consumers from buying a highly addictive product with a false sense of the risks it presents,” before any marketing began. This interest was especially powerful given the combination of health risks and vulnerable young consumers.

And the modified risk product pathway directly advanced the government’s substantial interest. Requiring a “significant[]” reduction of harms and risks to individual users and a “benefit” to the population as a whole directly advanced the government’s interests in accuracy and public health. “Given that no tobacco product has ever been shown to be safe, Congress ensured that the FDA will not lightly authorize the sale of tobacco products as carrying reduced health risk.” The special rules for claims about specific substances also directly advanced the government’s interest. “Each element of the inquiry is targeted towards ensuring that any specific-substance claim that consumers may understand as a relative safety claim is accurate and not misleading.”

Finally, the regulation was “not more extensive than necessary” to serve the government’s interest. In making a “fit” determination, “the least restrictive means is not the standard; instead, the case law requires a reasonable fit between the legislature’s ends and the means chosen to accomplish those ends[.]” That standard was satisfied. Congress found that “the only way to effectively protect the public health from the dangers of unsubstantiated modified risk tobacco products is to empower the Food and Drug Administration to require that products that tobacco manufacturers s[ell] or distribute[] for risk reduction be reviewed in advance of marketing, and to require that the evidence relied on to support claims be fully verified.”

And the rule for specific substances “reasonably tailors the requisite substantiation to the type of product.” For products marketed as generally less harmful, scientific studies must show that a “substantial reduction in morbidity or mortality among individual tobacco users occurs” with their use, whereas for those marketed only as less harmful because they contain a reduced level of a substance, the manufacturer must show only that reduced morbidity and mortality is “reasonably likely.”

The industry objected to premarket review because it believed that its claims for healthfulness were accurate. “But modified risk claims that might be technically accurate if viewed in isolation are in fact often misunderstood by consumers. In particular, Congress specifically found that consumers have been misled about the health consequences of claims that a tobacco product did not contain or contained reduced level of a harmful substance.” Just as with low tar and light cigarettes, “product labeling or advertising that touts an e-cigarette as free of a specified ingredient may mislead consumers to view the product as generally safer, even if other chemicals it contains, such as formaldehyde, are equally or more harmful than the disclaimed ingredient. The Industry’s claims of accuracy are unsubstantiated, and it has yet to submit an application with appropriate consumer-perception evidence.”

The court emphasized that misleadingness is based on the understanding of a significant number of reasonable consumers, and not only on what is explicitly said. Because the rationale supporting First Amendment protection of commercial speech is “the informational function of advertising,” “[t]he government may ban forms of communication more likely to deceive the public than to inform it.” And, when the speech addresses matters on which the “public lacks sophistication,” then “misstatements that might be overlooked or deemed unimportant in other advertising may be found quite inappropriate.” That was the case here: “Tobacco products are by definition harmful and addictive, and choosing among them based on comparative safety is inherently risky and complex, making the public especially susceptible to being misled and harmed.”

The court pointed to Congress’s knowledge of the sordid, deadly history of tobacco marketing as strong support for premarket approval of modified risk products. The FDA has already found similar problems with e-cigarette marketing, especially to young people. “Consumers have frequently and erroneously read narrow safety statements about an identified substance as materially complete claims that the product is safe overall.” Thus, the modified risk pathway could require the “testing of actual consumer perception” to show that “consumers will not be misled into believing that the product . . . is or has been demonstrated to be less harmful” more broadly.

The industry suggested a bunch of supposedly less restrictive alternatives. The court found none convincing.  First, required disclaimers: Congress considered and rejected them because they’d been ineffective to prevent deceptive tobacco marketing in the past. “The risk of misinterpretation regarding a highly addictive product supports the FDA’s choice of preclearance over a disclaimer requirement.” Second, post-market enforcement, putting the onus on the government. But that would require the FDA

to investigate the harms of an open-ended litany of substances that might appear in e-cigarettes, and to continually test products for their presence. Restricting the government’s regulatory options in that way is inappropriate for products containing harmful and addictive substances about which the public is known to be easily misled and about which the manufacturer has superior information. The FDA has already noted inaccuracies in claims made by various e-cigarettes about their nicotine content, and significant variability between labeled and actual content of various chemicals. Once inaccurate or misleading information influences people to start using a powerfully addictive substance, damage has been done.

The court also rejected the industry’s appeal to Sorrell v. IMS Health Inc., 564 U.S. 552 (2011), which unconstitutionally restricted “sophisticated and experienced consumers,” namely prescribing physicians, from accessing “truthful, nonmisleading advertisements.” The targets here were ordinary laypeople, including adolescents. And, unlike Sorrell, the modified risk pathway didn’t ban information going to one speaker while allowing its dissemination to others (like researchers). Not only was there no absolute ban here, as in Sorrell—only a substantiation requirement—but there was no non-e-cigarette group authorized to make the same claims in connection with a commercial transaction. [This mishmash of speaker/recipient isn’t the DC Circuit’s fault; it’s an effect of Sorrell’s own incoherence about what’s protected and why.]

Nor was the special treatment of chewing tobacco and “smokeless” or “smoke free” claims arbitrary, even though e-cigarettes couldn’t use the same terms without preclearance as a modified risk product. Congress relied on decades of use of the term “smokeless” to distinguish chewing tobaccco from loose smoking tobacco, a rationale inapplicable to e-cigarettes. Moreover, chewing tobacco isn’t inhaled. “To the extent that consumers may view ‘smokeless’ as a claim about relative pulmonary risk, decades of experience supports the FDA’s allowance of that claim for chewing tobacco whereas the FDA lacks any similar track record regarding e-cigarettes.”

Separately, the ban on free samples didn’t violate the First Amendment.  The industry argued that, since free samples are a marketing technique, they constituted expressive conduct, and that, since the reason Congress banned free samples was to decrease uptake, it was regulating based on the expressive effect of the conduct. 

The reason that Congress banned free samples was “to eliminate an easily accessible source for youth that are especially vulnerable to the risks of tobacco use and addiction.” The ban targeted conduct, not speech, and it wasn’t obviously expressive conduct either.  The industry failed to identify its “entirely unstated” message. The industry argued that free samples were “expressive” because they “convey[] important information to smokers who want to switch to vapor products, including key consumer information about different e-liquid flavors and device performance characteristics.”

“This extraordinary argument, if accepted, would extend First Amendment protection to every commercial transaction on the ground that it ‘communicates’ to the customer ‘information’ about a product or service.” The Supreme Court long ago rejected the idea that conduct carried out with the intent of expressing an idea is therefore speech. “[T]he seller’s intention that those experiences leave consumers with helpful information that encourages future purchases does not convert all regulation that affects access to products or services into speech restrictions subject to First Amendment scrutiny.”

Even if there were an incidental burden on speech, the restriction on conduct was imposed “for reasons unrelated to the communication of ideas,” and thus unproblematic. The free sample ban wasn’t about communication of information, it was about the products themselves and the well-documented danger that children would obtain and use them via free samples, given the greater price sensitivity of young consumers. Expressions Hair Design v. Schneiderman, 137 S. Ct. 1144 (2017), “recently reaffirmed that ordinary price regulation does not implicate constitutionally protected speech,” and the ban here was an ordinary price regulation: e-cigarette sellers can’t charge zero dollars.  

Since this is rational basis review, it doesn’t matter that the TCA allows distribution of free samples of chewing tobacco at “qualified, adult-only” facilities:

Anyone with even basic awareness of e-cigarettes and chewing tobacco, and their differential health consequences for and uptake by youth, will readily discern rational reasons to treat free samples of chewing tobacco differently from free samples of e-cigarettes. E-cigarettes are discreet and trendy in a way that chewing tobacco is not. Additionally, Congress’ limited exemption for free samples of chewing tobacco in specified, controlled circumstances reflects Congress’ knowledge of youth access and usage derived from years of experience. As the Industry concedes, no comparable information exists for e-cigarettes. Additionally, users of e-cigarettes inhale into their lungs myriad potentially hazardous substances not limited to those derived from tobacco. Congress’ decision to exempt chewing tobacco but not e-cigarettes from the free sample ban readily survives rational basis review.

The Sixth Circuit previously characterized a free sample ban as “an attempt to regulate the ‘communicative impact’ of the activity, not the activity itself.” But that case “addressed a regulation covering a range of clearly communicative promotional activities—including the distribution of tobacco- branded merchandise (t-shirts, baseball caps, bobblehead dolls) and event sponsorships—together with a prohibition on free product samples, and its First Amendment analysis grouped them together as ‘marketing bans.’” That was wrong, but even so, the Sixth Circuit concluded that any burden on the expressive element of free samples was easily justified by the FDA’s “overwhelming evidence” of the danger that free samples could fall into the hands of young people.  The industry argued that e-cigarettes were different because “consumers are searching for truthful information regarding a novel and potentially life-saving product category.” “Given the relatively unknown and potentially grave risks of e-cigarettes to all users, and their extraordinary allure to middle and high school students, we cannot agree.” [Also, trying an e-cigarette provides exactly zero information to the user about whether they are better, healthwise, than cigarettes.]