Probiotics History: From Sour Milk to Strain-Specific Labels
The probiotics story starts with a sour-milk idea and ends with a strain rule. This page separates what PubMed supports, including a 1994 infant trial and a 2012 antibiotic meta-analysis, from the claims a bottle can make.
The history of probiotics starts with sour milk and a long-lived village. Élie Metchnikoff linked the longevity of Bulgarian villagers to sour milk fermented by lactobacilli, according to a 2025 review in *Gut Microbes* ([according to PubMed](https://doi.org/10.1080/19490976.2025.2451088)). The modern question is about strains. A 2012 meta-analysis of 82 eligible trials found that strains were poorly documented across them ([according to PubMed](https://doi.org/10.1001/jama.2012.3507)). ## Who linked fermented milk to health? The 2025 review calls Metchnikoff a Nobel laureate and traces the historical roots of probiotics back to him. It says roots, not first. We have not verified a claim that he was the first to link fermented milk to health, so this page does not make one. ## When did probiotics get a definition? The 2014 ISAPP consensus reaffirmed the FAO/WHO definition of a probiotic as live microorganisms which, when administered in adequate amounts, confer a health benefit on the host ([according to PubMed](https://doi.org/10.1038/nrgastro.2014.66)). The consensus says that definition was 13 years old in 2014, which puts it around 2001. The sources we checked do not date when probiotics became a supplement category, so this page gives no year for that. ## What did the early infant and antibiotic trials show? A 1994 double-blind trial in the *Lancet* fed infants in a chronic medical care hospital either a standard formula or the same formula with *Bifidobacterium bifidum* and *Streptococcus thermophilus*. Diarrhea developed in 2 of 29 infants on the supplemented formula and 8 of 26 on the control formula. Rotavirus shedding fell from 10 of 26 to 3 of 29 ([according to PubMed](https://doi.org/10.1016/s0140-6736(94)91708-6)). The 2012 JAMA meta-analysis pooled 63 trials with 11,811 participants. It found a relative risk of 0.58 for antibiotic-associated diarrhea (95% CI 0.50 to 0.68) ([according to PubMed](https://doi.org/10.1001/jama.2012.3507)). The same paper says the evidence was insufficient to tell whether that association varied by population, antibiotic characteristic or probiotic preparation. ## Which strains show up on labels, and why does the strain matter? Strain names are the unit of evidence. A 2005 review says specific strains confer health benefits, but that evidence cannot be extrapolated to other strains, because the effects are strain-specific ([according to PubMed](https://doi.org/10.1111/j.1469-0691.2005.01228.x)). A 2018 review says probiotic properties are strain-specific, and that data cannot be extrapolated to other probiotics of the same or different species ([according to PubMed](https://doi.org/10.3920/BM2018.0013)). The [probiotics monograph](/ingredients/probiotics) lays out the same strain-by-strain logic in more detail. Good Flora is one example of a label that names its organisms. Its live product page lists *Lactobacillus acidophilus* and *Bifidobacterium* strains with fructooligosaccharides, a prebiotic fiber ([Good Flora](https://mtangelvitamins.com/products/good-flora-supports-healthy-digestion-and-bacteria?utm_source=learn_hub&utm_medium=content&utm_campaign=blog_post&utm_content=history-of-probiotics-from-yogurt-to-lactobacillus)). The page text we checked does not list a CFU count. ## How do probiotic claims differ from the evidence? A probiotic claim usually names a category, while a trial tests a specific strain at a specific dose. The 2014 ISAPP consensus argued for a more precise use of the term, so that clinicians and consumers can tell apart the products on the market. A bottle that says probiotic with no strain named describes the category, not a tested organism. For a newer organism in the same discussion, see the [Akkermansia monograph](/ingredients/akkermansia-muciniphila) and the [Akkermansia buying guide](/reviews/best-akkermansia-probiotics-2026). For the gut side of the story, see [the gut microbiome post](/blog/digestive/digestive-health-and-the-gut-microbiome). ## Bottom line The origin story and the definition are documented. The evidence is strain by strain, so the strain name and the dose tell you more than the word probiotic does. If you have a serious medical condition or are immunocompromised, talk with your clinician before you start. ## Frequently asked questions ### Who linked fermented milk to health? According to a 2025 review in *Gut Microbes*, Élie Metchnikoff linked the longevity of Bulgarian villagers to sour milk fermented by lactobacilli, and the review traces the historical roots of probiotics back to him ([according to PubMed](https://doi.org/10.1080/19490976.2025.2451088)). The review says roots, not first. We have not verified a claim that he was the first to link fermented milk to health, so this page does not make one. ### Why did yogurt become a longevity story in the early 1900s? The sources we checked credit the story to Metchnikoff's work, which the 2025 review says led to the global recognition of probiotics as beneficial supplements. That is the only causal link we could verify. The sources do not explain why the idea caught on when it did, and we did not find one that does, so this page does not offer a reason for the timing or guess at the mechanism. ### When did probiotics become a supplement category? We cannot date that from the sources we checked. The 2014 ISAPP consensus says the FAO/WHO definition of a probiotic was 13 years old at that point, which puts it around 2001. The consensus reinforced that definition rather than writing a new one, so the definition is the date we can document. The category date is not in these sources, so the answer stays open. ### Which strains appear on labels, and why does the strain matter? Strain names are the unit of evidence. A 2005 review says the evidence for specific strains cannot be extrapolated to other strains, and a 2018 review says probiotic properties are strain-specific. A label that names a strain, a species and a dose tells a reader which tested organism a claim belongs to. A label that gives only the word probiotic does not. ### What did the early infant diarrhea and antibiotic trials show? In a 1994 Lancet trial, diarrhea developed in 2 of 29 infants on a formula with Bifidobacterium bifidum and Streptococcus thermophilus, compared with 8 of 26 on the control formula. The 2012 JAMA meta-analysis pooled 63 trials with 11,811 participants and found a relative risk of 0.58 for antibiotic-associated diarrhea. It also found that strains were poorly documented across the eligible trials, which limits what the pooled number can say about any one product. ### How do probiotic claims differ from the clinical evidence? A probiotic claim usually names a category, while a trial tests a specific strain at a specific dose. The 2005 review says the evidence cannot be extrapolated to other strains. The 2014 ISAPP consensus argued for a more precise use of the term probiotic, so that clinicians and consumers can tell apart the products on the market. Until a label names the strain, the trial results cannot be matched to the bottle. ## Next reads - [Probiotics ingredient monograph](/ingredients/probiotics): the strain-by-strain evidence in more depth. - [Digestive health and the gut microbiome](/blog/digestive/digestive-health-and-the-gut-microbiome): the lateral read on the gut-bacteria side of this story. ### Related reading - [the pomegranate origins story](/blog/history/history-of-pomegranate-from-ancient-orchards-to-urolithin-a) - [the prebiotics origins story](/blog/history/history-of-prebiotics-from-chicory-to-fiber-science)