A bottle says “supports memory.” The same four syllables run across the top of the website, over a photo of a smiling grandparent, next to the words “clinically shown.”
Those are not the same claim, legally. The words on the package are a labeling question, and the Food and Drug Administration has primary responsibility for labeling. The words in the ad are an advertising question, and the Federal Trade Commission has primary responsibility for advertising. The two agencies apply different tests to identical sentences.
That gap is where the memory, immunity and stress supplement market lives. And the agency on the advertising side wants something most of these products do not have.
Two agencies, two standards, one sentence
The FTC published its Health Products Compliance Guidance on December 20, 2022. The document says it exists to update and replace the agency’s 1998 advertising guide for dietary supplements, a document that had stood for almost a quarter century. The new version also widened the scope beyond supplements to foods, over-the-counter drugs, devices and health-related apps.
The guidance divides jurisdiction in one line each. FDA has primary responsibility for claims that appear in labeling, including the package, product inserts and other promotional materials at the point of sale. The FTC, it says, has primary responsibility for claims in all forms of advertising.
Then it says the thing that reframes the entire subject. After noting that FTC law draws no bright-line distinctions between categories of health-related products or claims, the guidance gives an example: the Dietary Supplement Health and Education Act’s provisions on structure and function claims in labeling do not govern the FTC’s assessment of those same claims in advertising.
Read that twice. A sentence a company may lawfully print on a label under FDA’s labeling rules can simultaneously be an unsubstantiated advertising claim under the FTC Act. Nothing about the wording changes. The reviewing agency changes, and so does the standard. Another article in this library covers how FDA sorts structure and function claims from disease claims. What matters here is that the FTC does not use that sorting.
What the FTC counts as proof
The substantiation standard is the hard part, and the guidance states it directly. Health claims must rest on competent and reliable scientific evidence, which the document describes as tests, analyses, research or studies conducted and evaluated objectively by qualified experts, using procedures generally accepted in the profession to yield accurate and reliable results.
Then it gets specific about what form that takes. The guidance states that, as a general matter, “substantiation of health-related benefits will need to be in the form of randomized, controlled human clinical testing.”
Several practical consequences follow from the surrounding text.
- There is no magic number of studies. Quality matters more than quantity, and the guidance warns that numerous flawed and inadequate studies are unlikely to add up to competent and reliable scientific evidence.
- Independent replication strengthens a case.
- Results must be clinically meaningful, not merely statistically significant.
- A study must run long enough to show the benefit persists.
- Advertising that a specific ingredient is “clinically tested” implies the finished product delivers the benefit, which can require testing of the product itself rather than the ingredient.
- Public health recommendations, foreign monograph approvals and consumer surveys do not substantiate a claim.
- Hedging words such as “may” and “helps” do not rescue a claim built on emerging science.
Set that against the reality of the category. Most memory, immunity and stress products have never been through a randomized controlled trial at all, let alone one of their own finished formula.
The memory case that took almost eight years
The flagship enforcement action in this space is usually described as a 2017 lawsuit. That description is now seven years out of date.
On January 9, 2017, the Federal Trade Commission and the People of the State of New York filed a complaint for permanent injunction and other equitable relief in the US District Court for the Southern District of New York, case number 1:17-cv-00124, against the marketers of a memory supplement. The FTC’s announcement that day said the defendants had made false and unsubstantiated claims that the product improves memory and is clinically shown to work, and that the study they relied on failed to show the product worked better than a placebo on any measure of cognitive function.
What the usual retelling leaves out is everything after that. The docket on the FTC’s case page runs: appellate briefs from both plaintiffs in February, June and July of 2018; a Second Circuit summary order on February 21, 2019; an opinion and order on July 24, 2019; a stipulated order extending deadlines on June 29, 2021; a Memorandum and Judgment on November 18, 2024; and a further order on December 6, 2024. On December 10, 2024 the agency published a statement characterizing the result as a win for the FTC.
Two things are worth taking from that timeline. First, enforcement in this area is real but extraordinarily slow. Nearly eight years passed between complaint and judgment, during which the product remained on sale. Second, the theory of the case was not that memory claims are inherently illegal. It was that the specific evidence offered did not support them. The FTC’s case page does not detail the company’s study or its design, and the underlying complaint was not read for this article, so the specifics of that study are not reported here.
The agency’s cognitive-claims theory is not limited to pills, either. In January 2016 the FTC settled with the makers of a brain-training program for $2 million over claims about cognitive benefits, in a case filed in the Northern District of California. Same standard, different product.
“Immune support” and the line the FTC watches
The phrase “supports immune health” is not defined or restricted anywhere in US law. What is restricted is the implication.
The 2022 guidance has no dedicated immune section. It has something more useful: a worked example. The guidance describes a hypothetical ad for an herbal product promoted as boosting the immune system to help keep the nose and throat healthy through the winter season, with a product name and images of people sneezing and coughing. The guidance concludes that the name, the images and the nose-and-throat reference together likely tell consumers the product helps prevent colds, and that even without the name and images the nose-and-throat reference alone likely conveys a cold prevention claim.
Preventing a cold is a disease claim. A disease claim needs the randomized human trial evidence described above.
The enforcement record shows what happens at scale when that evidence does not exist. In a Business Blog post dated November 12, 2020, FTC attorney Lesley Fair announced 20 more warning letters to marketers making COVID-19 claims, bringing the running total past 330. The claims the post listed included a supplement company promoting two products as a natural strategy for COVID-19, a jewelry seller saying its bracelets boost immunity, and a water bottle maker saying copper can help you avoid flu, colds and other viruses.
The legal basis the letters rested on is the general one, not a COVID-specific rule. It is a violation of the FTC Act to represent, expressly or by implication, that a product can prevent, treat or cure a disease without competent and reliable scientific evidence, which the post says can include well-controlled human clinical studies substantiating the claims at the time they are made. For COVID-19, the post states, no such study was then known to exist for the products named.
FDA runs a parallel track on labeling. A warning letter dated November 3, 2020 to a mushroom supplement seller quoted the company’s own website claims about anti-cancer and antiviral effects and cited the statutory provisions that convert a product carrying disease claims into an unapproved new drug. That letter also invoked a regulation under which citing published studies in labeling can itself be evidence of a disease claim.
What the evidence shows
Memory and cognition
The best single answer comes from a systematic review funded by the Agency for Healthcare Research and Quality and published in Annals of Internal Medicine in 2018. It covered 38 trials of at least six months’ duration, rated low to medium risk of bias, testing omega-3 fatty acids, soy, ginkgo, B vitamins, vitamin D with calcium, vitamin C, beta-carotene, multi-ingredient supplements and other over-the-counter interventions in adults without dementia.
The conclusion: “Evidence is insufficient to recommend any OTC supplement for cognitive protection in adults with normal cognition or MCI.” OTC means over the counter. MCI means mild cognitive impairment, the stage between normal aging and dementia.
Two details inside that review matter. Moderate-strength evidence showed vitamin E had no cognitive benefit, which is a finding rather than an absence of one. And folic acid plus vitamin B12 produced improvements on some memory tests that reached statistical significance but that the reviewers described as of questionable clinical significance, which is exactly the distinction the FTC’s guidance insists on.
The largest trial in the field points the same way. The Ginkgo Evaluation of Memory study enrolled 3,069 community volunteers aged 75 and over at five US academic medical centers, randomized them to 120 milligrams of a ginkgo extract twice daily or placebo, and followed them for a median of 6.1 years. Dementia occurred at 3.3 cases per 100 person-years on ginkgo and 2.9 on placebo. That came to a hazard ratio of 1.12 for dementia of any cause, and the interval around it, 0.94 to 1.33, straddles 1. For Alzheimer disease it was 1.16, and for progression from mild cognitive impairment to dementia 1.13.
None of those reached statistical significance. But notice the direction. Every point estimate sits above 1.0, meaning every one of them nominally favors placebo. That is not a promising signal awaiting a bigger study.
Immunity and colds
The best-studied “immune support” botanical is echinacea, and the Cochrane review published in 2014 is the reference point. It included 24 double-blind trials with 4,631 participants producing 33 comparisons.
Of the 12 prevention comparisons reporting how many people caught at least one cold, not one found a statistically significant difference. Zero out of 12. A post hoc pooling of those same results suggested a relative risk reduction somewhere between 10 and 20 percent, which the reviewers reported while noting the individual trials were not significant. Of seven treatment trials reporting cold duration, one showed a significant effect.
The reviewers’ own summary is that echinacea has not been shown to benefit cold treatment, though a weak benefit from some products is possible, with effects of questionable clinical relevance. They also could not pool the main analysis, because the preparations differed too much from one another to combine.
Name the tier these findings sit in plainly, because advertising copy blurs the tiers on purpose. Supplements for cognitive protection in healthy adults or people with mild cognitive impairment sit in the insufficient evidence tier, which is the formal verdict of a federally funded review of 38 trials. Ginkgo for dementia prevention is stronger than insufficient and weaker than disproven: one adequately powered six-year trial found no benefit, with every estimate pointing the wrong way. Echinacea for cold prevention is insufficient, with 12 of 12 comparisons non-significant. None of this category reaches established evidence, and a product described as clinically studied may be standing on a trial that found nothing.
Look at who paid for all three. The AHRQ review, the ginkgo trial and the Cochrane echinacea review were all publicly funded, two of them with National Institutes of Health support, and all three came back null or near-null. So when an encouraging result turns up in this category, two questions do most of the work: who funded it, and how many people were in it. A federally funded trial of 3,069 people and a 30-person trial run by a company that sells the product are not two equal pieces of evidence, even when a marketing page lists them side by side.
“Adaptogen” is a marketing category, not a pharmacological one
No US statute, regulation, or FDA or FTC document reviewed for this article defines the word “adaptogen.”
It is not a new gap. A 1994 review in Phytomedicine, by one of the founding figures of modern plant pharmacology, opened by stating that the term had not yet been accepted in medicine, attributing that to the difficulty of separating adaptogenic drugs from immune stimulants, anabolic drugs, nootropics and tonics. The review’s framing came from stress physiology: plant drugs that blunt the alarm phase of the stress response or delay the exhaustion phase, at least in animal experiments.
Twenty-seven years later the working definition in circulation is stranger still. A 2021 review in Nutrients defines adaptogens as synthetic compounds, naming bromantane, levamisole, aphobazole and bemethyl, or plant extracts, that can increase the body’s stability against physical loads without raising oxygen consumption. The word is not even specific to botanicals under its own current definition. Bromantane is a synthetic stimulant. The concept came out of Soviet-era military and sports pharmacology, and the same review acknowledges that human clinical trials on plant adaptogens are limited.
So when a stress product calls itself adaptogenic, that word carries no regulatory meaning, no agreed clinical meaning, and no guarantee that the category is even plant-based.
The harm signal in the stress category
Stress supplements overlap heavily with a population already taking prescription medicines, and there is pharmacovigilance data on that overlap, which is rare in this field.
A 2023 retrospective chart review in Frontiers in Pharmacology, from a Polish university pharmacology and psychiatry group, examined 1,816 adverse-event reports. Adaptogens were involved in 9 percent of adverse events associated with taking antidepressants together with other preparations, and the authors identified 30 reports where the pharmacoepidemiological analysis showed a high probability of a causal relationship.
The documented outcomes are not minor. They include upper gastrointestinal bleeding with duloxetine, nosebleeds with paroxetine, vaginal hemorrhage with sertraline, delirium with amitriptyline, seizures with bupropion, arrhythmias with mianserin and low blood sodium with venlafaxine. The authors attribute the pattern to a combination of additive side effects and pharmacokinetic interaction through inhibition of cytochrome P450 enzymes and of P-glycoprotein, a transporter that pumps drugs back out of cells. Ashwagandha, berberine, jiaogulan and maca appear among the preparations named.
A retrospective chart review generates hypotheses rather than proving causation, and the authors say so. It is still the most concrete safety information available on this product class.
When to involve a clinician rather than read a label. If you take an antidepressant and you develop unexplained bruising or bleeding, black or bloody stools, a new seizure, confusion, palpitations or a sudden change in mood or agitation, that is an urgent call, not a wait-and-see. Bring the actual containers. The 2023 review’s whole method depended on someone knowing which preparations a patient was taking, and prescribers often do not. The decision about combining any of these with a psychiatric medicine, an anticoagulant or a transplant medication belongs to a pharmacist or physician who can see the full list.
Reading a claim the way a regulator reads it
Four questions that separate a substantiated claim from a well-produced one.
- Does the ad name an illness, a symptom of one, or a season? “Supports immune function” and “helps you stay well all winter” are different claims. The second is closer to prevention, and the FTC’s own worked example treats seasonal nose-and-throat language as a cold prevention claim.
- Is the cited research on this product, or on an ingredient? The guidance treats “clinically tested” as implying the finished product works, which is a higher bar than one ingredient having a literature.
- Does the claim rest on a mechanism or on an outcome? “Contains antioxidants” and “contains a compound shown to reduce dementia risk” are not the same statement, and only the second needs outcome data.
- Is the qualifier doing the work? “May help support” is not a weaker version of a substantiated claim. The guidance says hedges do not cure inadequate evidence.
The takeaway
The useful thing to carry out of this is not a verdict on any ingredient. It is that the legality of a sentence on a bottle tells you nothing about whether it is true, because labeling and advertising are judged by different agencies under different standards, and only one of those standards asks for randomized human trials.
When you want to know whether a memory or immune or stress claim has evidence behind it, the question is not what the package is allowed to say. It is whether a randomized controlled trial of that product, long enough to matter, with a clinically meaningful result, exists at all. For the three largest and most independent reviews in this field, the answer was no.
Sources
- Federal Trade Commission. “Health Products Compliance Guidance (December 20, 2022).” ftc.gov
- Federal Trade Commission. “Cases and Proceedings: FTC and People of the State of New York v. a memory supplement marketer, FTC Matter No. 152 3206, Civil Action No. 1:17-cv-00124 (S.D.N.Y.).” ftc.gov
- Federal Trade Commission. “Business Blog: 20 more marketers making COVID claims receive FTC warning letters (Lesley Fair, November 12, 2020).” ftc.gov
- Annals of Internal Medicine (PubMed), funded by the Agency for Healthcare Research and Quality. “Butler M, et al. "Over-the-Counter Supplement Interventions to Prevent Cognitive Decline, Mild Cognitive Impairment, and Clinical Alzheimer-Type Dementia: A Systematic Review." Ann Intern Med, 2018;168(1):52-62. PubMed. DOI: 10.7326/M17-1530.” pubmed.ncbi.nlm.nih.gov
- JAMA (PubMed). “DeKosky ST, et al. "Ginkgo biloba for prevention of dementia: a randomized controlled trial." JAMA, 2008. PubMed. DOI: 10.1001/jama.2008.683.” pubmed.ncbi.nlm.nih.gov
- Cochrane Database of Systematic Reviews (PubMed). “Karsch-Volk M, et al. "Echinacea for preventing and treating the common cold." Cochrane Database Syst Rev, 2014. PubMed. DOI: 10.1002/14651858.CD000530.pub3.” pubmed.ncbi.nlm.nih.gov
- Phytomedicine (PubMed). “Wagner H, Norr H, Winterhoff H. "Plant adaptogens." Phytomedicine, 1994. PubMed. DOI: 10.1016/S0944-7113(11)80025-5.” pubmed.ncbi.nlm.nih.gov
- Nutrients (PubMed). “Todorova V, et al. "Plant Adaptogens – History and Future Perspectives." Nutrients, 2021. PubMed. DOI: 10.3390/nu13082861.” pubmed.ncbi.nlm.nih.gov
- Frontiers in Pharmacology (PubMed). “Siwek M, et al. "Harder, better, faster, stronger? Retrospective chart review of adverse events of interactions between adaptogens and antidepressant drugs." Front Pharmacol, 2023. PubMed. DOI: 10.3389/fphar.2023.1271776.” pubmed.ncbi.nlm.nih.gov
- U.S. Food and Drug Administration. “Warning Letter 609440, November 3, 2020 (mushroom dietary supplements, disease claims).” fda.gov
- Federal Trade Commission. “Cases and Proceedings: FTC v. Lumos Labs, Inc., FTC Matter No. 132 3212, Civil Action No. 3:16-cv-00001 (N.D. Cal.).” ftc.gov