In February 2026, thirteen experts published a consensus definition of "gut health" in Nature Reviews Gastroenterology & Hepatology. It took them eighteen months. Their definition: "a state of normal gastrointestinal function without active gastrointestinal disease and gut-related symptoms that affect quality of life." That is: not being sick. The global probiotics market — worth over $100 billion — had been selling this concept for two decades before anyone agreed what it meant.
This is not a story about whether probiotics work. Some do, for specific conditions, at specific doses, with specific strains. This is a story about what happens when an entire market builds itself around an endpoint that was never defined — and what the evidence looks like when you finally try to define it.
The $100 Billion Undefined Endpoint
The ISAPP consensus panel identified six functional areas of gut health: digestive physiology, gut microbiome, intestinal barrier, immune system, gut hormones, and gut-brain connection. They then acknowledged that "reliable tests to measure gut health, particularly the gut microbiome, remain limited." In other words: they defined the concept, identified its components, and admitted you can't measure it.
The Market vs. The Measurement
88% of Americans are familiar with probiotics. 84% say they're more interested in gut health products than before. But 49% think all fermented foods contain probiotics (they don't). 44% think all live cultures qualify as probiotics (they don't). 71% think prebiotics do the same thing as probiotics (they don't). 41% are unaware the gut microbiome exists.
The Regulatory Split
The same science produces opposite regulatory conclusions depending on which continent you're standing on.
In the EU, EFSA assessed 355 probiotic health claims. They approved one. The word "probiotic" itself is considered an unauthorized health claim — a de facto ban on the word on packaging. IPA Europe argued "probiotic" should be a nutrition claim (describing the product), not a health claim. They lost.
In the US, probiotics are sold as dietary supplements. Structure/function claims — "supports digestive health," "promotes gut balance" — require no pre-market approval. The evidentiary burden sits on the FDA to prove a claim is false, not on the manufacturer to prove it's true.
What the Gastroenterologists Actually Recommend
The American Gastroenterological Association published clinical practice guidelines in 2020 using GRADE methodology. After systematic review, they found sufficient evidence for exactly three conditions — all with caveats:
| Condition | AGA Recommendation | Caveat |
|---|---|---|
| C. difficile prevention | Conditional: use | Only while on antibiotics |
| Pouchitis | Conditional: use | Specific 8-strain combo only; very low quality evidence |
| NEC in preterm infants | Conditional: use | Specific strains only; FDA safety warning after infant death |
| IBS | Only in clinical trials | Despite being the #1 marketed use |
| Acute gastroenteritis (children) | Recommended against | Moderate evidence of no benefit |
| General "gut health" | No recommendation | Endpoint undefined; broad supplementation discouraged |
The key principle: effects are strain-specific and combination-specific. Not species-specific. Not genus-specific. The market sells "probiotics." The evidence supports specific strains for specific conditions.
What the Products Actually Claim
Seed DS-01 Daily Synbiotic ($1.67/day, subscription only) contains 24 strains and markets itself for "Whole Body Health" — digestive, skin, heart, immune, and gut barrier benefits. They funded a 350-person RCT showing reduced bloating versus placebo over six weeks. Their own chief medical officer acknowledged the problem: "without rigorous clinical research, it's difficult to distinguish real gastrointestinal benefit from well-packaged claims." The studies were conducted with Seed employees and shareholders.
Align (Procter & Gamble) uses a single strain and markets itself as the "#1 Doctor Recommended Probiotic" for "24/7 Digestive Support." In 2017, P&G settled a class action lawsuit for $30 million over false advertising of digestive health benefits.
Culturelle uses L. rhamnosus GG and markets it as "The Most Scientifically Studied Probiotic Strain" with the tagline "Thank Science." LGG does have evidence for antibiotic-associated diarrhea prevention. But as probiotic expert Yehuda Ringel put it: "If you are a healthy person and keep a healthy way of life and healthy diet, you don't need probiotics to help you get healthier."
Poppi prebiotic soda marketed itself as "gut healthy" — until an $8.9 million settlement in 2025 revealed each can contained only 2 grams of prebiotic fiber, with sugar content that would "offset most, if not all" of the claimed gut health effects.
Each product markets broad "gut health" benefits. The evidence — where it exists at all — supports narrow, strain-specific applications for defined clinical conditions. The gap between what's sold and what's shown is the entire business model.
The Colonization Problem
The Weizmann Institute published a landmark finding in Cell in 2018 that undermines the basic premise of universal supplementation. Using metagenomic characterization of the mucosal-associated gut microbiome, Elinav and Segal found that probiotic colonization is person-specific, region-specific, and strain-specific. "Persisters" and "resisters" — some guts accept probiotics, some reject them.
Stool shedding — finding the probiotic in feces — does not equal gut colonization. The probiotic may pass straight through. And after antibiotic use — the scenario where many people reach for probiotics — the team found probiotics delayed microbiome recovery for months. Autologous fecal microbiota transplant restored it in days.
A BMC Medicine systematic review in January 2026 confirmed: evidence is insufficient to claim probiotics improve gut microbiota diversity in healthy populations.
The Precision Promise
The scientific community's answer to this heterogeneity is "precision probiotics" — strain-specific prescriptions based on individual microbiome profiles. The NIH launched PAR-25-211 in 2025: "Enhancing Mechanistic Research on Precision Probiotic Therapies." Phase 1 (two years): identify which host biological patterns correlate with probiotic response. Phase 2: validate those patterns in rigorous studies.
Translation: the field is still at the stage of figuring out which factors determine whether probiotics work for a given person. Rebyota and Vowst — fecal microbiota products, not traditional probiotics — have been FDA-approved, creating a regulatory pathway. But over 200 active trials later, the vision of personalized probiotic prescriptions remains years from clinical reality.
Where the Evidence Actually Lands
Probiotics are not useless. For antibiotic-associated diarrhea prevention, a 2025 meta-analysis of 15 trials (7,427 adults) shows S. boulardii and multi-strain formulations work. For IBS symptom relief, a 2024 Gastroenterology meta-analysis found probiotics better than placebo, with L. plantarum 299V showing the strongest evidence. For C. difficile prevention, a 2025 Cochrane Review of 47 studies found a "small absolute benefit."
The issue isn't that probiotics never work. It's that the market sells "gut health" — undefined, unmeasurable, unfalsifiable — when the evidence supports only narrow applications of specific strains for specific conditions.
The Mechanism: Undefined Endpoint Exploitation
This maps to a distinct failure mode in my growing taxonomy of knowledge failure. It's different from definition manipulation (antidepressants), where "works" was redefined after the fact. Here, the endpoint was never defined in the first place.
When the Endpoint Is Undefined
Any positive result can be claimed as supporting the product
No negative result can definitively refute it — you were measuring the wrong thing
The market fills the definition vacuum with marketing claims
Regulators who demand specifics (EFSA) find the evidence doesn't exist
Regulators who don't demand specifics (FDA) let the market boom
ISAPP's 2026 definition doesn't solve this. It sharpens it. Defining "gut health" as "not being sick" reduces it to absence of disease — which is already what medicine measures. The six functional areas they identified have no validated composite measurement. You still can't test for "gut health."
The FDA's sole health claim for a probiotic-adjacent food — the 2024 qualified health claim for yogurt and type 2 diabetes — explicitly states the benefit comes from yogurt as a whole food, not from its probiotic component. The probiotic industry's best regulatory win doesn't actually validate probiotics.
The three-sided structure: Industry sells $100B+ for an endpoint no one can define or measure. Regulators split — one demands proof and finds almost none, the other asks no questions. Science finds narrow pockets of real evidence for specific strains and conditions, but the market doesn't sell specifics. It sells "gut health."
The definition came twenty years late. And it defined nothing you can test for.