The FDA and FTC intend to require randomized clinical trials (RCTs) in support of promotional claims for dietary supplements and dietary ingredients. That much is clear. The FTC took that position in the recent POM Wonderful case, and the FDA memorialized that position in its Evidence Based Review System (EBRS). In practice, the FDA will not accept less than clinical evidence in support of any “health claim” or “qualified health claim” for a dietary supplement. We previously addressed this issue in prior posts by here and here. Some suggest the FDA is moving towards the EU standards. In an excellent article by the Alliance for Natural Health-US, the ANH reports that the European Food Safety Authority evaluated 2,037 health claims and authorized just 241 for use in commerce. ANH explains, “[t]his free speech issue is echoed in the POM Wonderful case, where the FDA [and FTC] attempted to bar a claim about the health benefits of pomegranate juice unless the company met a standard applicable to drugs.” Id. According to ANH, the FDA now inches closer to the EU model. In this article, we explain why that drug standard fails to the detriment of United States consumers.
The RCT model is impractical for dietary ingredients. Dietary ingredients are not drugs. So evidentiary models tailored for drugs are not always helpful in the nutritional context. Drugs are designed to “treat” conditions, while dietary supplements prevent the onset of health issues or improve general well-being (rather than acute conditions). The science usually suggests that dietary supplements will benefit some fraction of the population. In other words, we know going into a study that the benefit will not be universal. But dietary supplements are generally safe and have limited or no adverse effects when taken daily over time. Thus, even if the ingredient helps five percent of the population avoid a major illness, the claim may be justified.
FDA and FTC would impose the same standards for drugs and dietary supplements. Requiring RCTs for dietary ingredients effectively projects the drug standard into the nutritional context. That is inappropriate and unnecessary for at least several reasons. First, do we really need an RCT to tell us that certain dietary ingredients might provide long-term benefits? Animal and in vitro models often provide useful data, particularly when supported by nutritional science. For dietary ingredients, put simply, we do not need the certainty that we look for in drug products because dietary supplements are offered for preventative purposes or general health. In the POM Wonderful case, the ALJ drew that distinction clearly. The judge explained that FTC should not impose insurmountable evidentiary hurdles when the product is not intended to replace medical therapies (i.e., drugs).
Second, as ANH explained, RCT models are impractical in human nutrition. If someone has a cancerous condition, an RCT can help determine over a short duration whether the drug product improved that condition. The drug might claim to “treat osteoporosis,” for example. So you test patients with that condition, after six months you examine bone density, and the results could help establish efficacy. A dietary supplement may claim to reduce risk of contracting certain cancers. Imagine performing a clinical study that can produce statistically significant results for that claim. How would you test? How large would the population size need to be? How long would the study last? You are testing against the population generally, trying to establish that the dietary ingredient reduces the incidence of cancers over a lifetime.
The problems here should be obvious. We cannot predict with accuracy who will develop cancer, so the population size must be substantial. We would need to test subjects over a lifetime (or at least many years), and that fact alone is financially prohibitive. Even assuming costs were not unreasonably high for studies of that duration, we would need substantial compliance with the protocol over that span. As the duration of a study increases, we lose the ability to control for confounders or variables. We cannot, for example, control a subject’s exact diet over twenty years. In the end, achieving an RCT for a dietary supplement “health claim” is near impossible because: (1) the cost to operate a methodologically sound study would be insurmountable; and (2) FDA could likely find enough bias to reject the data if the agency did not approve of the conclusion.
Again, as ANH noted, many dietary ingredients cannot be patented. What altruistic company has the finances to complete an RCT knowing that the results can be enjoyed by the entire market?
The FDA and FTC also ignore an obvious issue. These placebo-controlled studies may be unethical. When you test a drug, all subjects have the condition already. But dietary supplements are offered to prevent the onset of certain conditions. For example, FDA approved the following qualified health claim: “some scientific evidence suggests that consumption of antioxidant vitamins may reduce the risk of certain forms of cancer.” Assume we have a sound basis to suspect that Vitamin C prevents colorectal cancer. If we organize an RCT to test that theory, and we implement a “control” group (or placebo), then aren’t we deliberately exposing the control group to a heightened probability of colorectal cancer?
FDA could argue that the control group merely reflects the status quo. Without an intervention, the investigators have not affected the control group adversely or positively. But in reality the investigators have deliberately denied the control group a nutrient that may reduce risk of a life-threatening condition. In other words, if the study ultimately produced statistically significant results, that would mean people in the control group actually developed cancer in greater amounts. I guess only that would satisfy our bureaucrats. FDA clearly explains in its Evidence-Based Review System that “[w]hen the substance is provided as a supplement, a placebo should be provided to the control group.” FDA thus downgrades clinical studies of dietary ingredients that lack a placebo controlled group.
FDA might respond, “ doesn’t that same argument apply for drug studies, where the patient with a condition is denied a potentially helpful treatment?” Not quite. “[M]edical interventions are designed to cure a disease not produced by their absence, while nutrients prevent dysfunction that would result from their inadequate intake…” See Blumberg J, et al., “Evidence-based criteria in the nutritional context,” Nutrition Reviews, Vol. 68(8):478-484. In other words, by withholding the nutrients from subjects in a food study, the investigators might actually cause the condition in otherwise healthy subjects:
“In order to conduct a RCT that adequately tests the efficacy of a nutrient for a specific chronic disease, it will usually be important to ensure an adequate contrast in intake between the intervention and the control groups. The control intake is an approximate analog of the placebo control in drug RCTs. However, since sufficiently low intakes are associated with significant disease in some body systems, doing so can lead to serious ethical problems, particularly if the disease outcome is serious and/or irreversible…”
Id. at 480. In Blumberg, et al., the authors discuss the contrasts between evidence-based medicine (EBM) and evidence-based nutrition (EBN):
“EBN thus departs from the situation of EBM, where, for most interventions, the use of a no-intake control group is usually quite appropriate. In EBM, the hypothesis is that adding an intervention ameliorates a disease, whereas in EBN it is that reducing the intake of a nutrient [or limited that intake] causes (or increases the risk of) disease. This distinction is critical. No one proposes in EBM that a disease is caused by the absence of its remedy; whereas for nutrients the hypothesis is precisely that malfunction is caused by deficiency. A hypothesis about disease causation can rarely, if ever, be directly tested in humans using the RCT design. This is because in the RCT the disease/dysfunction occurs in at least some of the study participants, and the investigators must ensure that this will happen. Instead, where EBN must operate is with respect to two related, but different questions: (i) In addition to disease X, does the inadequate intake of nutrient A also contribute to other diseases? And (ii) At what level of intake of nutrient A is risk of all related disease minimized or all related functions optimized?”
Id. In the drug model, investigators merely preserve the status quo in the placebo group. In nutritional studies, the investigators might contribute to the harm.
We need to revisit the FDA’s Evidence-Based Review System (EBRS) as applied to dietary supplements and, particularly, qualified health claims. The FDA must consider animal, in vitro, and mechanistic studies in support of those claims. The agency should review each study for its scientific merit and weigh those studies against the actual risk to consumers from ingestion of the nutrient. If a health claim purports to reduce risk of a significant or life-threatening condition, then the burden of scientific proof should be lowered, or FDA should have an obligation to show that the claim is affirmatively misleading. Under FDA’s current approach the agency rejects everything that is not an RCT. That model puts our citizens at risk. That model is unethical.
The FDA now suggests that the EBRS (which mirrors FTC’s two clinical trial approach) should extend to structure/function claims for dietary supplements. Thus, as ANH explained, the FDA would move us closer to the EU’s system of premarket approval for all claims. After all, few (if any) dietary supplement companies can support structure/function claims with RCTs. That policy would result in a de facto ban for most claims. RCTs are particularly unnecessary for S/F claims, which frequently address the basic chemical or biophysical properties of an ingredient. In other words, do we really need an RCT to tell us Vitamin C helps support the immune system? Why would a ninth-grade biology textbook not be sufficient?
One major difference between the EU and US model is the First Amendment. Freedom of speech is more entrenched in the United States than in the European Union. Free speech in the EU stems from Article 10 of the European Convention on Human Rights. The ECHR specifically allows government intervention where speech is deemed harmful to the public, or when necessary to preserve democracy, prevent disorder or crime, protect the rights of others, or maintain authority. EU and US courts apply a similar legal test when assessing restrictions on commercial speech. But free speech in the EU does not enjoy the same protection as in the US where the right is ingrained in our constitution and considered fundamental.
For example, when the European Parliament and EU Council banned tobacco advertising in 2003, the European Court of Justice upheld the measure. See Case C-380/03, Germany v. Parliament, 2006 E.C.R. I-11573 (holding Directive did not violate freedom of expression). The Court explained: “The legality of a measure adopted in that sphere can be affected only if the measure is manifestly inappropriate having regard to the objective which the competent institutions are seeking to pursue.” Id. (emphasis added). By contrast, we saw the opposite result in the Supreme Court’s 2001 Lorillard Tobacco case, which involved a similar law. See Lorillard Tobacco Co. v. Reilly, 533 U.S. 525 (2001). In that case, applying the Central Hudson test, the Court held that the regulation on advertising was too broad to satisfy the “narrowly tailored” requirement. Id. at 553-61; see also Sean P. Flanagan, Note, Up in Smoke? Commercial Free Speech in the United States and the European Union: Why Comprehensive Tobacco Advertising Bans Work in Europe, but Fail in the United States, 44 Suffolk U. L. Rev. 211, 222 (2011) (comparing EU and US free speech regulation).
The First Amendment likely prevents FDA from adopting a regulatory scheme equivalent to the EU model. If anything, the federal courts have recently been willing to provide heightened protection to commercial speech. See, e.g., Sorrell v. IMS Health, Inc., — U.S. —, 131 S.Ct. 2653 (2011); United States v. Caronia, — F.3d —, 2012 WL 5992141 (2d Cir. Dec. 3, 2012). Our earlier blog post here discussed that shift. As long as that trend continues, FDA will find less support in the law for speech censorship. Meanwhile, if FDA and FTC reliance on RCTs continues in evidence-based nutrition, less valuable information will reach consumers.