Health claim substantiation is a positioning decision

Health claim substantiation is the requirement that a company hold competent and reliable scientific evidence before it makes a claim, not after. The Federal Trade Commission sets that bar as a two-part test in its health products compliance guidance: “(1) tests, analyses, research, or studies that have been conducted and evaluated in an objective manner by experts in the relevant disease, condition, or function to which the representation relates; and (2) are generally accepted in the profession to yield accurate and reliable results.” The evidence has to come from the right kind of expert review, and it has to be produced using methods the field already treats as reliable. A claim published without both parts already in hand is not a claim with weak backing. It is not substantiated at all.

Most companies treat this as a checkpoint near the end of the process. Copy gets written, a reviewer checks it against the evidence file, and anything unsupported gets softened until it says almost nothing. The evidence never shapes the copy. It only edits it after the fact.

Reverse the order and the evidence becomes a strategy input. What a company can actually prove, and at what strength, sets the boundary of what it can credibly say. Inside that boundary, precision is available to every company willing to look at what it holds. The companies that stand out are not the ones with the deepest evidence file. They are the ones who know exactly what their evidence supports and say exactly that, no more and no less.

Summary

• Substantiation is the evidence a company held before it made a claim, not a check performed after the copy is written. Treating it as an input changes what a company can credibly say.
• The FTC judges the net impression of an entire ad, not just the literal sentence, so images, testimonials, and headlines are all part of the claim.
• A structure/function claim and a disease claim sit on opposite sides of a line that determines whether a product gets regulated as a drug or device.
• The FTC has primary jurisdiction over most advertising; the FDA covers labeling and prescription-drug advertising, under a decades-old division of responsibility.
• A qualifier can narrow an ambiguous claim but cannot contradict one the evidence does not support, and the same evidentiary bar applies to testimonials and reviews as to a company’s own copy.

What does it actually mean to substantiate a health claim?

A claim is substantiated when the evidence existed before publication, not when a reviewer can find supporting material after the fact. The FTC’s plain-language framing of health claims is direct: “Companies must support their advertising claims with solid proof. This is especially true for businesses that market food, over-the-counter drugs, dietary supplements, contact lenses, and other health-related products.”

The bar for that proof rises with the claim. A specific claim about an outcome for a disease or condition needs evidence at the level of a randomized controlled trial, evaluated by experts in that field. A general claim about how a product supports the body sits lower on the same scale, but it still needs something concrete behind it, not an assumption that sounds plausible.

Consider two hypothetical claims a company might put on a homepage: “clinically inspired” and “doctor-formulated.” Neither states a specific outcome, which is exactly why they are common. Neither is free. “Clinically inspired” implies the formulation traces to clinical research; if no such research exists, or if it exists but does not actually support the formulation, the phrase is not backed by anything. “Doctor-formulated” implies a credentialed process; if the doctor in question reviewed a label rather than designed a formulation, the phrase overstates what happened. A founder testing either phrase should be able to name, on the spot, the specific study, data set, or documented process behind it. If the answer is a shrug, the phrase is not a claim with weak evidence. It is not a claim at all, and it should not survive to publication.

Why does a strong product still read like everyone else’s?

This is where most companies lose their edge, and it has little to do with the product. Copy gets drafted first, usually to sound confident and broad. Legal or compliance review arrives last, finds the parts the evidence cannot support, and cuts them. What survives is whatever was vague enough to clear review: “supports overall wellbeing,” “backed by science,” “trusted by experts.” Every competitor’s page can say the same thing, because none of it was ever specific enough to be owned by one company.

The fix is not writing bolder copy and hoping it survives review. The fix is running the evidence question first. Before a single sentence gets drafted, three questions settle what is actually available to say:

1. What tests, data, or documented processes does the company hold today, and who evaluated them?
2. Which of those would meet the FTC’s “competent and reliable” bar on their own, and which are supporting context rather than proof?
3. What specific, narrow claim does that evidence actually support, stated in the plainest possible language?

A company that answers those three questions before writing has a claim nobody else can make, because nobody else holds that exact evidence. A company that answers them after writing is just editing down to whatever survives.

Structure/function claim or disease claim: where does your product sit?

What’s the practical test?

A structure/function claim describes how a product supports the normal structure or function of the body. FDA regulation permits a dietary supplement’s label or labeling to carry this kind of statement, but only when it meets a further set of conditions set out in the same regulation and explained in FDA’s compliance guide on structure/function claims. A disease claim states or implies that a product treats, prevents, cures, or mitigates a disease or condition. The general pattern: “supports” language sits on one side of the line, “treats” language sits on the other.

Crossing that line changes which regulator and which rulebook applies. A disease claim on a product not regulated as a drug can pull the whole product into drug regulation, regardless of what the company calls it on the label. The FDA’s most recent wellness-device guidance draws the same line for software and low-risk devices: “Software that is intended for maintaining or encouraging a healthy lifestyle and is unrelated to the diagnosis, cure, mitigation, prevention, or treatment of a disease or condition” is not a device. The moment the claim implies diagnosis or treatment of a specific condition, that exemption is gone.

A hypothetical health-technology example makes the line concrete: “supports recovery after activity” describes a structure/function role and stays on the permitted side. “Detects early signs of [a named condition]” implies diagnosis and crosses into disease-claim territory, which requires an entirely different evidence standard and regulatory pathway. The words look similar in a sentence. The regulatory consequence is not.

FTC or FDA: who is actually reading your ad copy?

Founders often assume the FDA reviews all health-related advertising. It does not. A decades-old agreement between the two agencies settles the split: “With exception of prescription drugs, the Federal Trade Commission has primary responsibility with respect to the regulation of the truth or falsity of all advertising (other than labeling) of foods, drugs, devices, and cosmetics.” In practice: the FTC reads the ad copy on a website, a social post, or a sales page. The FDA reads the product label and prescription-drug advertising.

The FTC’s enforcement is not theoretical. In 2023 it sent formal notice to almost 700 companies: “certain acts or practices concerning substantiation of product claims are deceptive or unfair and violate the FTC Act.” That notice put those companies on record that unsubstantiated claims going forward could trigger civil penalties.

The clearest illustration of how the evidentiary bar actually gets set comes from litigation, not guidance. In the case widely referred to by advertisers as the POM Wonderful matter, the reviewing court found that the FTC had not adequately justified a categorical requirement of two randomized controlled trials for every disease-adjacent claim, and modified the underlying order to require “at least one RCT” instead. That account comes from a client alert by the law firm Crowell & Moring, which quotes the court’s language; the underlying opinion itself was not the source consulted here, so the alert is the citation, not the case. The detail worth taking from it is the rule, not the case name: one well-designed clinical trial, not two, is the evidentiary bar most practitioners now cite for disease-adjacent claims.

Can you make a claim when your evidence is limited? What “qualified” really means

A qualified claim is one narrowed by a disclosure to match what the evidence actually supports, rather than left to imply more than the evidence shows. The FTC evaluates the whole ad, not just the headline sentence, when it decides what a claim actually communicates: it “should consider each ad as a whole, assessing the ‘net impression’ conveyed by all elements of the ad, including the text, product name, and any charts, graphs, and other images.” A net impression is that whole-ad reading: what a reasonable consumer takes away from the headline, the imagery, the product name, and the fine print, combined.

This is exactly where a disclosure can help and exactly where it cannot. In the FTC’s words: “Although a clear and conspicuous disclosure might be effective to clarify an ambiguous claim that might otherwise be deceptive, it can’t directly contradict a claim.” A footnote can narrow a claim that is genuinely ambiguous. It cannot rescue a headline the evidence flatly contradicts. Founders who add a qualifier expecting it to fix an overreach are solving the wrong problem: the fix belongs in the headline claim itself, not in a disclaimer underneath it.

The same bar extends past a company’s own copy. The FTC’s 2023 update to its Endorsement Guides revised the rules “to reflect the ways advertisers now reach consumers to promote products and services, including through social media and reviews”. A patient testimonial, an influencer post, or a before-and-after image carries the same substantiation burden as a claim written in-house. A company cannot say through a customer’s mouth what it could not say in its own copy.

The cosmetics category carries an added wrinkle worth naming directly, since aesthetics is one of the categories where this line gets crossed most often. As Cohen Healthcare Law Group’s analysis of cosmetics claims puts it, “If a product is marketed with claims that it treats or prevents disease or affects the structure or function of the body, it may be classified as a drug, even if it’s also a cosmetic.” The claim, not the formulation, decides which rulebook applies.

What to do next

1. Inventory the evidence held today: published studies, internal data, documented process, mechanism-level rationale. Write it down before drafting new copy, not after.
2. Sort every current or planned claim into structure/function or disease-adjacent. Flag anything crossing that line for review before launch, not after publication.
3. Read every claim for net impression, not just the sentence. Check what the headline, the imagery, and any testimonial imply together.
4. Where evidence is preliminary, write the qualifier into the claim itself, and confirm it narrows the claim rather than quietly contradicting it.
5. Apply the same evidentiary bar to reviews, testimonials, and endorsement content that applies to in-house copy.
6. Revisit claims when guidance changes. The FDA’s wellness-device policy update in January 2026 is a recent example of a rule moving the line on what is sayable.

None of these steps requires new data. They require an honest look at data the company already has.