What the Label Cannot Say

The bottle cannot tell you what the Cochrane review already says. Around every nutrient, three legal categories operate in nearly every industrial-market jurisdiction (structure-function claim, health claim, therapeutic claim) to guarantee that the seller is barred from citing evidence that would take the product out of its commercial box. A tour of the regulatory landscape, and of the manufactured silence it produces.

The empty label problem

Take riboflavin. High-dose riboflavin, 400 mg per day, for migraine prophylaxis has been in the Cochrane database for two decades. The randomized-trial evidence is not marginal. Yet the bottle of B2 sold in North America or in Europe cannot say that. The label will say “supports energy metabolism”, or “supports normal function of the nervous system”, or, in the North American case, will carry the mandatory disclaimer that the product “is not intended to diagnose, treat, cure, or prevent any disease”. The migraine is not on the bottle. The trial that named the migraine is not on the bottle. The customer is not told.

This is not an accident. It is a structure. In every industrial-market jurisdiction, the same three-tier architecture governs what a supplement label may say: a permissive tier of vague physiological claims, a restricted tier of nutrient-condition claims requiring pre-approval, and a prohibited tier of therapeutic claims reserved for medicines. The three tiers differ in name across jurisdictions but not in function. They exist to guarantee that any specific therapeutic use documented by clinical research remains, on the label, unsayable.

This piece traces that architecture across the western regulatory landscape and Japan, points to the Codex Alimentarius floor from which the national systems derive, and argues that the phrase “insufficient evidence for supplementation”, so familiar in clinical guidelines, is not a scientific verdict about the substrates. It is a manufactured silence produced by the interaction of a research funding architecture that does not fund nutrient trials and a labeling regime that criminalizes the citation of the evidence that has been generated despite the funding.

The common architecture

The three-tier scheme, in generic terms.

Nutrient function claims, or structure-function claims. The most permissive tier. The claim links a nutrient to a normal biological function without asserting an effect on a disease. “Iron contributes to normal cognitive function.” “Vitamin C contributes to normal function of the immune system.” Permissible in most jurisdictions with minimal review, sometimes on the basis of a pre-approved list, sometimes on notification. Physiologically true but clinically evasive. The claim does not tell the customer whether their cognition or their immunity is currently deficient, whether the substrate would address a specific condition, or at what dose.

Health claims, or nutrient-condition claims. The restricted middle tier. The claim links a nutrient to reduction of the risk of a specific disease or to a defined health effect beyond normal function. “Adequate calcium and vitamin D intake, combined with physical activity, may reduce the risk of osteoporosis.” “Consumption of foods high in soluble fibre may reduce the risk of coronary heart disease.” Permissible in principle but subject to pre-approval by a competent regulatory authority. The evidentiary standard is high. The list of approved claims is short.

Therapeutic claims, drug claims, medical claims. The prohibited tier. The claim asserts that the product treats, cures, mitigates, or prevents a specific disease. This is the legal territory of the pharmaceutical, of the registered medicine, of the drug approval process. To make such a claim about a supplement is to reclassify it as an unapproved drug and to invite enforcement action. The migraine, the depression, the neuropathy, the cardiovascular event: all of these belong to the third tier, and none of them may appear on a supplement bottle.

The scheme has a common effect regardless of jurisdiction. What can be said in the first tier is too vague to help the customer choose. What can be said in the second tier is limited to a short list of politically stabilized claims that has not moved substantially in decades. What could be said in the third tier is where the actual clinical literature on the substrates lives, and it cannot be said at all.

1. The United States: DSHEA (1994)

The Dietary Supplement Health and Education Act, passed in 1994, is the founding legislation. It carves out dietary supplements as a distinct regulatory category, neither food nor drug, and sets the terms of what they may say about themselves. The permitted tier is the structure-function claim. Under 21 U.S.C. § 343(r)(6), a manufacturer may claim that a nutrient “supports” or “maintains” a normal biological function without pre-market approval by the Food and Drug Administration, provided the claim is notified to the agency within thirty days of first use and is accompanied by the mandatory disclaimer: “This statement has not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease.”

The disclaimer is the load-bearing sentence of the entire American supplement label. It sits under every claim. It reminds the customer that the label speaks under legal duress. What it does not say, and what the architecture guarantees the label cannot say, is that the reason the claim has not been evaluated is that the substrate is unpatentable and no sponsor exists to fund the drug approval trial. The label reads as if the science had not been done. The label is required to read that way.

Enforcement is the mirror of the permission. Warning letters issued by the FDA to manufacturers whose marketing drifts from “supports cognitive function” into “improves memory in Alzheimer’s disease” are routine, published, and internalized by the market as boundary maintenance. The specific therapeutic claims accumulate in the specialty literature, in the Cochrane reviews, in the clinical guidelines of practitioners willing to reference them, and do not migrate to the bottle.

2. The European Union: Regulation (EC) No 1924/2006

The European framework is more restrictive than the American one, and its restrictiveness is where the “no evidence” verdict is most visibly manufactured. Every health claim in the second tier requires pre-authorization by the European Food Safety Authority (EFSA), whose Panel on Dietetic Products, Nutrition and Allergies evaluates the dossier against a standard modelled on pharmaceutical evidence.

The result of the initial evaluation, running through the late 2000s and early 2010s, is the fact most often cited when the European system is described. Tens of thousands of health claim proposals were submitted. A small percentage were authorized. The bulk were rejected as insufficiently substantiated. Notable rejections included probiotic claims for immune function, glucosamine claims for joint maintenance, cranberry claims for urinary tract health, and a long list of botanical claims left unresolved.

The rejections were not, on inspection, verdicts that the substrates lacked biological effect. They were verdicts that the dossiers, judged against the criteria appropriate to a pharmaceutical registration, did not clear the bar. The criteria are not appropriate to nutrients whose research literature is composed of observational studies, small randomized trials, and mechanism papers, and for which no sponsor has ever assembled a phase-3 pharmaceutical-grade dossier because there is no patent to recoup the investment. What EFSA declared insufficient was not the biology. It was the ability of the nutritional literature to conform to a documentary standard designed for another industry.

At the shelf level, the European supplement bottle is often more silent than its American counterpart. The prevailing communication converges on the short list of authorized function claims consolidated in Regulation (EU) No 432/2012, and prohibits the rest.

3. Canada: Natural Health Products Regulations (2004)

Health Canada operates a distinct regime for dietary supplements, botanicals, homeopathic preparations, and traditional medicines. Every product must carry a Natural Product Number, issued by Health Canada after evidence review. The load-bearing feature of the Canadian system is the monograph. Health Canada publishes pre-approved monographs for common ingredients, establishing the permitted dose ranges and permissible claims. A manufacturer marketing a nutrient with an existing monograph aligns their claim to what the monograph permits, which shortens the approval process; a claim outside the monograph requires a separate dossier and specific evidence review.

The result is a system more restrictive than the American, and more open to traditional and homeopathic categories than the European. It produces its own version of the same silence. What the monograph permits can be said. What the monograph omits cannot. The monograph advances at the pace of regulatory review, which is not the pace of the scientific literature.

4. Australia: the TGA and the Permitted Indications list

The Therapeutic Goods Act (1989) and the Therapeutic Goods Administration run a two-tier system. Listed medicines (AUST L) are low-risk products self-assessed by the sponsor against a pre-approved list of permissible indications. Registered medicines (AUST R) are higher-risk products evaluated in detail. Vitamins, minerals, and most herbal preparations sit in the AUST L tier. A 2018 reform tightened the framework by publishing a formal list of Permitted Indications from which listed medicines must draw their claims. The result is a system in which permissible language is standardized, deviation is prohibited, and specific therapeutic claims supported by clinical literature but absent from the permitted list are legally unsayable on the bottle.

5. Japan: FOSHU (1991) and FFC (2015)

Japan is the western regulatory landscape’s asymmetric neighbor. Foods for Specified Health Uses (FOSHU), established in 1991, was the first national system anywhere to authorize specific health claims on non-medicinal foods, decades before comparable European or American movements. Approval is granted by the Consumer Affairs Agency after scientific review, and approved products bear a distinct seal. A second track, Foods with Function Claims (FFC), was introduced in 2015 to accelerate market entry: sponsors notify the agency of the claim and the underlying scientific evidence without formal pre-approval, and take responsibility for the substantiation. A third track, Foods with Nutrient Function Claims, standardizes the equivalent of European function claims.

The Japanese system permits more direct claims on foods than any other major industrial-market jurisdiction, and the shelves in a Japanese pharmacy read accordingly. It also demonstrates that the western reticence is a regulatory choice, not a scientific inevitability. The evidence base supporting many Japanese FFC claims is the same literature the European rejections cite as insufficient. What differs is the criterion of sufficiency.

6. The Codex Alimentarius floor

Codex Alimentarius, the joint FAO and WHO food standards programme, publishes the Guidelines for Use of Nutrition and Health Claims (CAC/GL 23-1997, subsequently revised). Codex texts are not directly binding on member states, but they are the reference to which national regulators return, and they underpin the World Trade Organization’s Agreement on the Application of Sanitary and Phytosanitary Measures. The Codex framework distinguishes nutrient content claims, comparative claims, nutrient function claims, other function claims, and reduction-of-disease-risk claims, with a general prohibition on therapeutic claims.

It is the source of the tier vocabulary that national regulators localize, and the multilateral instrument that formalizes, at the international level, the exclusion of therapeutic claims from the food and supplement categories. The exclusion is presented as consumer protection. Its effect, for the class of nutrients that will never be developed as pharmaceuticals, is to guarantee that no legal channel exists through which the therapeutic potential documented in the clinical literature may reach the label.

The structural production of “no evidence”

The systems described above share a common design feature. They define what may be said about a substrate by reference to whether the evidence has cleared a bar calibrated to the pharmaceutical registration process. The bar was designed for molecules whose sponsor could recover the cost of the trial through patent-protected pricing. Nutrients cannot. No sponsor has an incentive to spend the tens of millions required to bring the riboflavin migraine indication through a drug approval process, because there is no patent to recoup the investment. The trials that would clear the bar have not been done, and will not be, absent a public funding architecture that has not been built.

The consequence, at the citation level, is the phrase familiar to any reader of clinical guidelines. “Insufficient evidence for supplementation.” The phrase is factually accurate at one level: no phase-3 pharmaceutical-grade dossier exists for most nutrients. It is misleading at every other level. It does not describe the state of the science. It describes the state of the paperwork. The Cochrane reviews, the observational studies, the mechanism papers, the case series, and the small controlled trials that constitute the actual nutritional literature exist and are consultable. They do not appear on the label because they cannot appear on the label. They frequently do not appear in the guideline because the guideline authors are working within the same evidence hierarchy that produced the regulatory bar.

What is called “no evidence” is often shorthand for “no dossier that clears the regulatory bar we set”. Those are not the same claim.

The bar exists for a reason, and the reason is not entirely bad. Consumer protection against fraudulent therapeutic marketing is a legitimate function of the state. The problem is not the existence of the bar. It is that a bar calibrated for the pharmaceutical case is applied to a class of substances the pharmaceutical model was never designed to develop, and the resulting silence is then re-read, by the clinical apparatus that receives it, as if it were an absence of biology.

What this resource says, that no label can

The physiological-rights argument does not ask that the labeling regime be dismantled. It asks that the silence produced by the regime be recognized for what it is: a legal artifact, not a scientific verdict. And it asks that the space the regime creates, the space in which the substrates are systematically undocumented at the point of sale, be occupied by writing that is under no equivalent constraint.

This resource sits in that space. It is not a bottle. It has nothing to sell. It has no incentive to overstate an indication, and no incentive to withhold a documented one. What it can say, that the bottle cannot:

• That riboflavin at 400 mg per day has Cochrane-database evidence for migraine prophylaxis.
• That coenzyme Q10 has Q-SYMBIO trial evidence for symptomatic heart failure, and that the mevalonate pathway means statin therapy depletes it by construction.
• That magnesium has evidence for pre-eclampsia and for migraine, that the reference range for serum magnesium has been challenged by the MaGNet international network for over a decade, and that intracellular magnesium is what actually matters.
• That potassium has intervention-trial evidence for blood pressure and stroke risk, and that no routine assessment of dietary potassium adequacy exists in the standard workup.
• That vitamin B12 in the elderly is masked by folic acid fortification, that the functional markers (holotranscobalamin, methylmalonic acid) reveal deficiencies that serum B12 does not, and that neither is routinely ordered.

Each of these statements is a statement no supplement bottle in the western regulatory landscape is permitted to make about itself. Each is documented in the specialty literature at a level that would justify the claim in a system that respected the actual evidence rather than the shape of the dossier.

The resource does not sell riboflavin, coenzyme Q10, magnesium, potassium, or B12. It has no commercial standing at the threshold of the right. What it has is the freedom to say, in the register of scientific description rather than of commercial claim, what the label is forbidden to say. That freedom is not extraordinary. It is the ordinary freedom of scientific communication when it is unhitched from the act of selling. It is also the substrate on which any coherent physiological-rights framework will have to be built.

A right that cannot be named cannot be claimed. This resource is where the claim starts.