Alliance for Natural Health US v. Sebelius

Citations: 786 F. Supp. 2d 1; 2011 U.S. Dist. LEXIS 39908; 2011 WL 1898920Docket: Civil Action 09-1546(BAH)

Court: District Court, District of Columbia; April 13, 2011; Federal District Court

SummaryOriginalCytator Case TreatmentBackward CitatorAnalysis

EnglishEspañolSimplified EnglishEspañol Fácil

The lawsuit filed by dietary supplement designers and industry groups challenges the FDA's denial of a petition for qualified health claims related to vitamin C and vitamin E. The plaintiffs argue that this denial infringes on their First Amendment rights and seek a declaratory judgment to invalidate the FDA's decision, along with a permanent injunction preventing the FDA from blocking their health claims on product labels. The case involves plaintiffs Durk Pearson and Sandy Shaw, who develop dietary supplement formulations, alongside industry organizations, against several FDA officials. The legal context includes definitions of dietary supplements and health claims as per the Food, Drug, and Cosmetic Act and the Nutrition Labeling and Education Act of 1990, which grant the FDA regulatory authority over such claims. The court is currently reviewing motions for summary judgment from both parties and has decided to grant in part and deny in part these motions while remanding certain claims back to the FDA for further consideration.

The NLEA provides a 'safe harbor' for foods and dietary supplements to avoid being classified as 'drugs' when making health claims, contingent upon the FDA's determination of 'significant scientific agreement' based on available scientific evidence (21 U.S.C. 343(r)(1), 343(r)(3)(B)(i)). Unlike food health claims, dietary supplements lack a specified approval process from Congress, leaving the FDA to establish the necessary procedures (21 U.S.C. 343(r)(5)(D)). The FDA adopted the same standard for dietary supplements as for food claims (21 C.F.R. 101.14(c)). If a dietary supplement makes an unauthorized health claim, it could be classified as misbranded or unapproved, potentially leading to seizures, injunctions, or criminal penalties (21 U.S.C. 343(r)(1)(B), 352(f), 355(a), 331(a), 332, 334, 352(a)).

In Pearson v. Shalala, plaintiffs challenged the FDA's rejection of health claims on First Amendment grounds. The D.C. Circuit ruled against the FDA's review process, highlighting the tension between consumer protection and manufacturers' rights (Pearson I, 164 F.3d at 655-61). The FDA had previously rejected claims by asserting that the evidence did not meet the 'significant scientific agreement' standard, labeling them misleading and outside First Amendment protections (Pearson I, 164 F.3d at 653, 655). The FDA also dismissed the idea of allowing claims with disclaimers, stating it was not obliged to consider this approach if significant scientific agreement was lacking (Id. at 654, 655, 657). The plaintiffs argued that the FDA's standard was unconstitutionally vague and effectively prohibited commercial speech (First Pearson District Court Opinion, 14 F.Supp.2d at 14).

The D.C. Circuit reversed the district court's denial of the supplement designers' motion for summary judgment, applying the commercial speech test from Central Hudson Gas & Electric Corporation v. Public Service Commission of New York. The Court determined that the government's goals of public health protection and consumer fraud prevention were not reasonably aligned with the FDA's rejection of health claims without considering disclaimers. It emphasized a First Amendment preference for disclosure over suppression and for less restrictive regulatory methods. The FDA's failure to evaluate whether disclaimers could mitigate consumer confusion was viewed as a disregard for these principles. The Court remanded the case to the district court, instructing it to have the FDA assess disclaimers' adequacy and specify their content.

Additionally, the Court stated that the FDA must provide a definition for "significant scientific agreement," as arbitrary rejection of claims without explanation violates the Administrative Procedure Act (APA). While the FDA can ban health claims if evidence against them outweighs supporting evidence, it must justify its restrictions on speech adequately. The Court cautioned against mere assertions of misleadingness without substantial justification.

In a subsequent case, Pearson II, plaintiffs challenged the FDA's prohibition of a health claim regarding folic acid after the FDA requested scientific data following Pearson I. Despite the submission of new data and the issuance of guidance on the "significant scientific agreement" standard, the FDA denied the claim as inherently misleading. Plaintiffs contended that this decision misinterpreted the legal standards from Pearson I and violated both the First Amendment and the APA.

Plaintiffs sought a preliminary injunction to prevent the FDA from blocking their use of a folic acid health claim. The district court sided with the plaintiffs, stating that the FDA did not adhere to constitutional guidelines set in Pearson I by failing to justify its conclusion that the claim was "inherently misleading" and unsupported by evidence. Although the court acknowledged the FDA's authority under the Administrative Procedure Act (APA) to assess the claim, it rejected the FDA's evaluation of the scientific data and determined that the claim was not inherently misleading. The court emphasized that the absence of strong evidence against a claim does not equate to negative evidence against it and clarified that credible support for a claim cannot be entirely dismissed unless outweighed by stronger evidence to the contrary. Consequently, the court deemed the FDA's determination arbitrary and capricious, granting the injunction and instructing the FDA to create appropriate disclaimers.

Following this, the FDA requested reconsideration, contending that the court had overemphasized a specific study and set an inconsistent legal standard. The court denied the motion, criticizing the FDA for not accurately representing the evidence. Subsequently, in a related case, Whitaker v. Thompson, plaintiffs challenged the FDA's rejection of an antioxidant claim, claiming a misapplication of standards from Pearson I. The FDA noted insufficient scientific consensus on the antioxidant-cancer relationship, labeling the claim inherently misleading. The district court agreed with the plaintiffs, concluding that the FDA failed to demonstrate that its regulatory approach was the least restrictive means to achieve its objectives, referencing a pertinent Supreme Court decision.

The court found that the FDA did not provide sufficient evidence to justify the denial of a proposed antioxidant claim, even when accompanied by a disclaimer. It emphasized that while deference is given to an agency's scientific evaluations, the court must ensure a reasoned decision-making process. The court analyzed over 150 studies related to antioxidant vitamins and cancer, noting that approximately one-third supported the proposed claim. It determined the FDA had not adhered to its own Guidance Report and had improperly weighed certain studies. The court deemed the FDA's rationale for denying the claim unreasonable, stating it lacked support from the available evidence and the FDA’s own guidelines. Additionally, the court ruled that the FDA failed to demonstrate that a disclaimer would mislead consumers. Consequently, it granted a preliminary injunction against the FDA's decision, asserting that the suppression of the claim conflicted with First Amendment protections favoring disclosure over suppression of commercial speech.

In a related case, Alliance I, the same plaintiffs challenged the FDA's rejection of health claims regarding selenium supplements. The plaintiffs proposed "qualified" claims with disclaimers to mitigate misleading implications, following the FDA's new category established in response to prior court rulings. The FDA later banned some claims entirely, citing a lack of credible scientific support, while allowing modified claims that had some backing. The plaintiffs contended that both the outright bans and the modified claims violated their First Amendment rights by imposing restrictive conditions that distorted the nature of their claims.

The Alliance I court conducted an independent review of the FDA's determinations regarding health claims under the Central Hudson test, emphasizing that it would defer to the FDA’s scientific interpretations only if they were reasoned and not arbitrary. The court noted that under Pearson I, claims supported by credible evidence cannot be entirely banned. It found that while many of the FDA's determinations were reasonable, some were arbitrary, leading to a remand for reevaluation and appropriate disclaimers. For claims allowed in modified form, the court criticized the FDA for not adhering to the Supreme Court’s requirement for a reasonable fit between government goals and commercial speech restrictions, determining that the FDA's modifications negated the original claims' meanings. The court mandated a reconsideration of the scientific literature and the drafting of clear disclaimers. 

The case at hand is similar to Alliance I, involving a challenge to the FDA's June 19, 2009 decision, which denied thirteen health claims and allowed four as modified qualified claims. The plaintiffs are contesting six specific proposed health claims linking vitamins C and E to reduced cancer risks, with evidence described as convincing or persuasive but not conclusive for each claim.

The FDA's decision prohibited Claims 1, 2, 3, and 4, while allowing Claims 5 (regarding Vitamin C and gastric cancer) and 6 (regarding Vitamin E and bladder cancer) as qualified claims with specific language indicating the uncertainty and inconsistency of the supporting studies. The plaintiffs filed suit on August 14, 2009, arguing that the FDA's decision infringed upon their First Amendment rights, asserting that their qualified health claims are backed by credible scientific evidence and are not misleading. The case involves motions for summary judgment from both parties, with the FDA filing an administrative record on December 4, 2009, followed by the plaintiffs' motion on December 30, 2009, and the FDA's cross-motion on February 22, 2010. The Court will grant summary judgment if no material fact disputes exist and if the movants are entitled to judgment as a matter of law. As there are no factual disputes, the Court will make an independent assessment of the constitutional claims without deferring to the FDA's constitutional interpretations, but will defer to the FDA's scientific evaluations, provided they are reasoned and not arbitrary. The D.C. Circuit instructed the FDA to adopt less restrictive speech regulations while ensuring empirical support for its decisions, without dictating how the FDA should evaluate scientific data.

The plaintiffs' health claims regarding vitamin C and vitamin E are classified as commercial speech, requiring evaluation under the Central Hudson framework, as interpreted by the D.C. Circuit in Pearson I and the Supreme Court in Western States. The process begins with determining if the speech involves unlawful or misleading content; if so, it lacks protection. If the speech is lawful and not misleading, or only potentially misleading, the government must demonstrate a substantial interest in regulating it. The regulation must directly advance that interest and not be broader than necessary, requiring a reasonable fit between the government's goals and the means used to achieve them. The government bears the burden to prove that its regulations are not excessively restrictive. The preference is for more disclosure rather than suppression, with disclaimers being favored over outright bans, unless the government can show that a disclaimer would be insufficient to address misleadingness. However, if evidence supporting a claim is significantly weaker than opposing evidence, the FDA may impose a ban without a disclaimer. The plaintiffs argue that the FDA cannot ban a claim without empirical evidence showing that a disclaimer wouldn't suffice, while the FDA contends there is no strict requirement for such evidence. The district court in Alliance I indicated, albeit in dicta, that Pearson I does not mandate empirical evidence for the FDA to prohibit unsupported claims.

The court in Whitaker established that a complete ban on a claim can only occur in limited situations, specifically when there is minimal qualitative evidence supporting the claim and the government can demonstrate that the public would still be misled despite a disclaimer. It concurs with Pearson I, which does not mandate the FDA to empirically prove that disclaimers are ineffective before prohibiting unsupported claims. The D.C. Circuit in Pearson I noted skepticism regarding the government’s ability to empirically show that suggested disclaimers would confuse consumers, yet it acknowledged the possibility without requiring such evidence in cases where the FDA could reasonably determine a disclaimer would be inadequate due to a lack of credible support for the claim. For instance, if evidence contradicts a claim like “Consumption of Vitamin E reduces the risk of Alzheimer's disease,” the FDA might find a disclaimer insufficient. The court clarified that the D.C. Circuit did not imply that an empirical study on disclaimers was necessary before banning unsupported claims. It indicated that claims with weak support could be banned outright, as they might be considered unprotected commercial speech under the Central Hudson analysis. In contrast, claims with some credible support are deemed potentially misleading and are afforded protection, requiring empirical evidence to justify a ban. In the absence of such evidence, the FDA must utilize disclaimers to manage potentially misleading claims.

The FDA has issued a complete ban on four health claims made by the plaintiffs regarding the effects of vitamin C and vitamin E on lung, colon, and gastric cancers. The FDA determined there was "no credible scientific evidence" supporting these claims, aligning with precedents established in Pearson I, which allows for complete bans when evidence against a claim outweighs supporting evidence. The Court's role is limited to assessing whether the FDA's evaluation was consistent with its own standards and not arbitrary or capricious. The FDA employs an evidence-based review system, outlined in its 2009 Guidance Document, to evaluate health claims, assessing the methodological quality and strength of the totality of scientific evidence, ultimately determining if there is significant scientific agreement or credible evidence for a claim. The document emphasizes that while "credible evidence" is not explicitly defined, it plays a critical role in the evaluation process.

The FDA prioritizes certain types of evidence to draw scientific conclusions, emphasizing randomized controlled trials as the most reliable for establishing causal relationships between substances and diseases. In contrast, research synthesis studies and review articles lack sufficient detail on individual studies, while animal and in vitro studies are limited in their applicability to human disease conclusions. The FDA assesses intervention or observational studies based on various factors, including study population, data collection methods, and disease biomarkers. Flawed studies are excluded from review. The FDA's Guidance Document classifies health claims into three tiers: authorized health claims (significant scientific agreement), qualified health claims (credible but not significant evidence), and unsupported claims (deceptive and potentially prohibited). Claims with inconclusive credible evidence may be qualified with disclaimers rather than prohibited outright, reflecting a less restrictive regulatory approach. Plaintiffs argue that the 2009 Guidance Document has reverted to a standard requiring conclusive evidence for claim approval, but the FDA's application shows that inconclusive yet credible evidence has been accepted in several cases, suggesting the agency's criteria for credibility are reasonable.

The Court rejects the plaintiffs' assertion that the FDA's evaluation of scientific evidence is flawed due to its exclusion of certain studies based on methodology. The FDA's Guidance Document specifies criteria for assessing the credibility of health claims related to supplements, indicating that predefined principles help identify studies that lack credible evidence. Specifically, the FDA found no credible support for the plaintiffs' claim linking vitamin C to lung cancer, dismissing five studies cited by the plaintiffs for various reasons. The Comstock and Gackowski studies were discounted because they relied on blood levels of vitamin C as indicators of intake, which the FDA argued are poorly correlated with actual intake due to various physiological factors. The FDA referenced multiple studies to support this reasoning and reinforced that credible evidence requires a strong correlation between intake and biological sample levels, as stated in the 2009 Guidance Document.

The Cho study was also excluded as it was a meta-analysis, which the FDA deems insufficient for evaluating individual study data unless it encompasses all available studies on a substance-disease relationship. The Court found the FDA's exclusion of the Cho study valid, noting it pooled data from only eight studies and concluded that vitamin intake did not reduce lung cancer risk. Overall, the FDA's rationale for excluding the studies aligns with its guidance, warranting affirmation of its decision regarding the lack of credible evidence for the plaintiffs' claim.

The FDA has determined that observational studies linking vitamin intake from dietary sources to lung cancer risk do not provide reliable evidence of the effects of individual nutrient supplements. It discounted studies by Feskanich and Neuhoser due to potential methodological weaknesses, including reliance on self-reported food intake, variability in nutrient content based on food handling, and the inability to isolate the effects of specific nutrients. The FDA's 2009 Guidance Document reiterates that conclusions drawn from observational studies cannot establish a direct relationship between food components and diseases. Plaintiffs challenge this exclusion, referencing the D.C. Circuit's ruling in Pearson I, which suggested that disclaimers could mitigate the lack of significant scientific agreement regarding vitamin supplement claims. However, the FDA argues that the concerns raised in Pearson I are insufficient to validate claims based on observational studies, emphasizing that such studies do not support valid scientific conclusions about single nutrient supplements. The FDA cites evolving understanding since Pearson I, including a 2005 review in the Journal of the American Medical Association, which highlights that nutrients from food may not have the same effects in supplement form and may even increase disease risk.

The Court finds that the FDA has a reasonable basis for concluding that observational studies on food intake do not provide credible scientific evidence for the disease risk effects of single nutrient supplements. While the D.C. Circuit acknowledged that concerns about isolating nutrient effects could be addressed through disclaimers, the FDA has identified additional valid concerns regarding the reliability of such studies. Unlike prior cases where the FDA's standards were unclear, the criteria for evaluating credible evidence were clearly articulated in the 2009 Guidance Document and are deemed scientifically rational. The Court will defer to the FDA's scientific judgment regarding the validity of the methodologies used.

Regarding the plaintiffs' claim that "Vitamin C may reduce the risk of lung cancer," the FDA rejected the studies presented by the plaintiffs as credible evidence. Specifically, the plaintiffs cited five studies to support their vitamin C-colon cancer claim, including the Cahill et al. (1993) study. The FDA dismissed the Cahill study because it did not measure validated surrogate endpoints for colon cancer risk, focusing instead on actual cancer cases or recurrent adenomatous colorectal polyps as the only credible indicators. The Cahill study, which examined vitamin C's effect on colon crypt cell proliferation in ten patients with existing polyps, was found inadequate for supporting claims related to colon cancer risk.

Higher rates of crypt cell proliferation in patients with colon cancer or polyps were noted, but the study did not establish a causal link between this proliferation and cancer risk, nor did it demonstrate that reducing proliferation impacts such risk. The Cahill authors suggested that cell proliferation effects in colon polyps might lower the risk of cancer progression, but this was not validated. Consequently, the Court determined that the FDA's decision to exclude this study was reasonable and aligned with its Guidance Document, which states that scientific conclusions about disease risk cannot be drawn from non-surrogate endpoints. The FDA also excluded four other studies cited by plaintiffs due to their reliance on dietary vitamin intake estimation, which justified exclusion based on the nature of the studies. 

In relation to the plaintiffs’ claim regarding vitamin E and lung cancer, the FDA dismissed the Woodson et al. and Comstock et al. studies for using blood serum vitamin levels, the Knekt et al. study for estimating dietary intake, and the Lonn et al. study for not being specific to lung cancer and potentially biased due to not screening for specific cancers. The Lonn study's main conclusion indicated no benefit of vitamin E in preventing cancer or cardiovascular events and raised concerns about increased heart failure risk. Despite a noted decrease in lung cancer incidence in the vitamin E group, the authors applied stringent statistical rules that undermined this observation.

The study on lung cancer rates did not show a statistically significant difference, leading the authors to conclude that observed differences were likely due to chance. Consequently, the study does not support the plaintiffs' claims regarding health benefits from vitamin E supplements, and the FDA's skepticism about potential biases in the study aligns with this conclusion. The FDA deemed the study, along with others cited by the plaintiffs, as lacking credible evidence to support their claims, a decision found not to be arbitrary or capricious.

For the plaintiffs' vitamin E-gastric cancer claim, they referenced five studies, but the FDA invalidated the Buiatti and Lopez-Carillo studies for estimating vitamin E intake based on food consumption. The You and Jenab studies were discounted for relying on serum or plasma levels, while the Virtamo study was ignored as it was not designed to assess gastric cancer risk and instead focused on urinary tract cancer, concluding no risk reduction from vitamin E.

The FDA recognized that four of the plaintiffs' proposed claims had some credible scientific backing but required rewording to avoid misleading consumers. The revisions for the vitamin E-bladder cancer claim indicated that while one small study suggested a risk reduction, two others did not, leading the FDA to conclude that it is unlikely vitamin E reduces bladder cancer risk. Similarly, for the vitamin C-gastric cancer claim, the FDA noted one weak study and another with inconsistent results, determining that it is highly uncertain vitamin C reduces gastric cancer risk. The Court found that the FDA's rewording of these claims conflicted with the principle that there should be a reasonable fit between governmental objectives and the constraints imposed on commercial speech.

The FDA's disclaimers regarding health claims related to vitamin intake are ineffective, mirroring previous failures identified in Alliance I. The FDA has not provided precise disclaimers that allow for the qualification of plaintiffs' claims in accordance with First Amendment principles favoring disclosure. Instead, the FDA’s modifications completely replace the original claims and obscure the association between vitamin intake and cancer risk, contradicting its own findings that credible evidence supports these claims. The FDA's approach effectively undermines the plaintiffs' claims without justification for not adopting a less restrictive method. In cases where evidence is inconclusive, the First Amendment allows claims to be made without being negated by disclaimers. Consequently, the FDA's substitution of plaintiffs' claims with contradictory language is inconsistent with prior case law (Pearson I). The court has ordered the remand of the vitamin C and vitamin E claims to the FDA for the creation of appropriate disclaimers. The court also granted and denied motions for summary judgment, remanding certain claims while denying other relief sought by the plaintiffs.

The excerpt details a legal discussion regarding the FDA's treatment of health claims related to selenium and its potential effects on cancer risk. Plaintiffs proposed several health claims asserting that selenium may reduce the risk of various cancers, backed by credible yet inconclusive scientific evidence. The FDA rejected some claims, finding them misleading, particularly for not specifying the types of cancer affected. The court criticized the FDA's decision as inconsistent with a prior case (Pearson I), noting that the claims were "literally true" due to existing credible evidence. Consequently, the court remanded the claims to the FDA for possible disclaimers or further justification for the misleading designation.

Additionally, the FDA does not technically "authorize" qualified claims but instead exercises "enforcement discretion" to allow them if they are credible and not misleading. This distinction stems from the FDA's regulatory requirements that necessitate significant scientific agreement for health claims to be authorized. The court emphasized that the First Amendment restricts the FDA from prohibiting all claims lacking significant scientific agreement, thus allowing for qualified claims. Furthermore, the court retains the authority to evaluate federal agency actions for potential constitutional violations, independent of the Administrative Procedure Act (APA), which also allows for a judicial review of constitutional issues related to agency actions.

The standard of review for agency decisions on constitutional questions is de novo, while interpretations of scientific information are afforded deference unless deemed irrational or arbitrary and capricious, as established in Alliance I. In intervention studies, subjects receive a controlled substance, while observational studies assess disease associations in a free-living population without such control. The FDA's Guidance Document leans more towards disclosure than the standard set in Pearson I, where a claim could be rejected if evidence against it outweighed supporting evidence, though a claim could be cured by a disclaimer if supported by credible evidence. The validity of the FDA's proposed language for claims will be addressed separately. The Court has not evaluated the overall reasonableness of all FDA criteria in the Guidance Document but has focused only on the criteria relevant to its decision. A point of contention involves the significance of peer review; while plaintiffs argue that their studies are validated through peer review, the FDA counters that the plaintiffs have not shown that the studies support their specific health claims, emphasizing that credibility hinges on the relevance of the studies to the claims rather than their general scientific merit.

The record review reveals that certain studies cited by the plaintiffs actually undermine their claims regarding the relationship between vitamin C intake and lung cancer risk. Specifically, a 2006 study by Cho et al. found no association between vitamin C intake and lung cancer risk, contradicting the plaintiffs' assertions. The FDA notes that many articles are published for scientific exploration rather than providing conclusive evidence supporting health claims. The Court will only analyze the first sentence of the plaintiffs' claims, disregarding any disclaimers that suggest the evidence is not definitive. The FDA's rejection of these claims was based on insufficient credible evidence linking vitamins to cancer, rather than plaintiffs’ disclaimers.

The plaintiffs argue against the FDA's use of a review article by Lichtenstein and Russell, citing the 2009 Guidance Document, which states that such articles alone do not constitute sufficient evidence for health claims. However, the FDA is permitted to use these articles to provide background and identify relevant studies. The FDA has cited both review articles and individual studies to highlight inconsistencies between observational studies and intervention studies, particularly concerning beta-carotene, which was initially thought to reduce lung cancer risk but later found to increase risk in certain populations. While plaintiffs claim to have submitted 17 studies, they reference more than that in summary form, indicating a broader basis for their assertions.

The Court has conducted a focused review of the FDA's handling of studies cited in the plaintiffs' memorandum. The plaintiffs claim that a study included pre-screening for cancers, referencing follow-up visits that collected data on illnesses and symptoms; however, this screening appears to occur post-study commencement, and the study does not provide findings on gastric cancer. The plaintiffs' claims suggest a potential substance-disease relationship, which the FDA assessed for evidence, while a proposed disclaimer addressed the evidentiary support's strength and nature. The Court clarifies that it is not responsible for drafting disclaimers but references disclaimers from the D.C. Circuit’s Pearson I as examples. Furthermore, the Court remands the claim to the FDA due to the FDA's complete substitution of the plaintiffs' proposed claim, indicating that a comprehensive review of all studies excluded or relied upon by the FDA was unnecessary.