Six servings of fermented food a day reshaped gut microbes in a Stanford trial, so start with one

Key takeaways

• In a 2021 Stanford trial, 18 healthy adults who worked up to about six servings of fermented foods a day gained gut microbe diversity, and 19 inflammation-related blood proteins fell over 10 weeks⁵⁶⁷.
• The trial’s main immune measure didn’t change in either diet group, and it was small, so the finding is promising but early⁵⁷.
• Fermented doesn’t always mean alive: baking kills sourdough’s microbes, and filtering or heat leaves wine, some beers and shelf-stable sauerkraut without live cultures³⁴.
• In March 2024, the FDA said it would not object to yogurt labels claiming that at least 2 cups a week may lower type 2 diabetes risk, while calling the evidence limited⁹¹⁰.
• A global cancer panel judged salt-preserved foods a probable cause of stomach cancer, so lean on lower-salt ferments such as plain yogurt for daily servings⁸¹⁴.

The volunteers started at zero. At the outset of a Stanford University study, 18 healthy adults were eating no fermented foods⁷. Then, step by step, they added yogurt, kefir, kimchi, kombucha and more, until they reached about six servings a day⁶⁷.

By the end of 10 weeks, the variety of microbes in their guts had grown⁶⁷. Blood tests showed 19 proteins tied to inflammation had dropped⁶. A second group, assigned to load up on fiber, saw neither change⁶.

It was a striking result from a tiny study whose main measure came up empty⁷. It also left a puzzle: most of the new gut microbes didn’t seem to come from the food⁷.

Fermented and probiotic aren’t the same thing

In 2021, an expert panel organized by the International Scientific Association for Probiotics and Prebiotics (ISAPP) gave fermented foods a formal definition¹. They are “foods made through desired microbial growth and enzymatic conversions of food components”¹.

The key word is “desired.” Spoilage doesn’t count. Neither do many store-bought pickles and olives, which are simply packed in a salty, acidic brine rather than fermented³.

The microbes don’t have to be alive when you eat the food. Many fermented foods lose them before they reach your plate³⁴:

• Sourdough and other breads. Baking kills the yeasts and bacteria³⁴.
• Wine and some beers. Yeasts are filtered out or killed by heat³.
• Shelf-stable sauerkraut, soy sauce and canned fermented vegetables. Heat treatment makes them keep on a shelf and kills the cultures³.

“Probiotic” is a stricter word. The ISAPP panel said a fermented food earns it only when it holds live microbes that are identified and shown in a study to deliver a health benefit². By that standard, most fermented foods are not probiotics, even when they are full of life³.

The panel’s first author, Maria Marco of the University of California, Davis, put the problem plainly: many people believe fermented foods are good for them, but “the scientific studies required to prove it are limited”².

Inside the Stanford trial

Stanford microbiologists Justin and Erica Sonnenburg and nutrition researcher Christopher Gardner led the study⁶. Graduate student Hannah Wastyk and Gabriela Fragiadakis were lead authors of the July 2021 paper in the journal Cell⁵⁶.

The design. Researchers randomly assigned 36 healthy adults to one of two diets, 18 per group⁶⁷. Everyone went through three weeks of baseline testing, then 10 weeks on the assigned diet, then four weeks eating as they liked⁶.

The two diets. One group built up to about six daily servings of fermented foods⁷. Their menu ran from yogurt, kefir and fermented cottage cheese to kombucha, vegetable brine drinks, kimchi and other fermented vegetables⁶.

The other group roughly doubled its fiber, from about 22 grams a day to 45, using legumes, seeds, whole grains, nuts, vegetables and fruit⁶⁷.

What changed with fermented foods:

• More kinds of gut microbes. Diversity rose over the course of the diet, and the effect was stronger in people who ate larger servings⁶⁷.
• It lasted. Four weeks after the diet ended, diversity was still above where it started, though people were eating fewer fermented foods by then⁷.
• Calmer immune signals. Nineteen inflammation-related proteins in the blood fell⁶. One was interleukin-6, a signaling protein that Stanford tied to chronic stress, type 2 diabetes and rheumatoid arthritis⁶. Four types of immune cells also became less active⁶.

The puzzle. Nine groups of microbes increased, but only a tiny share of them turned up in the foods themselves⁷. Justin Sonnenburg said most came from somewhere else, and nobody knows where⁷. The authors suggested the foods reshaped the gut’s ecosystem rather than simply seeding it⁷.

What happened with fiber. Average diversity stayed flat, and none of the 19 proteins fell⁶. More carbohydrate showed up in stool, suggesting people’s microbes couldn’t fully break down the extra fiber⁶.

The fiber diet did shift the microbes’ capacity to process carbohydrates⁷. According to the paper’s abstract, people’s immune responses on fiber also split into three patterns that tracked how diverse their microbes were at the start⁵. Erica Sonnenburg suggested a longer study, or adding fiber-eating microbes, might be needed to see a payoff⁶.

The limits. The trial’s main measure, an immune score called the cytokine response score, didn’t change in either group⁵⁷. The study had 36 people, all healthy, for 10 weeks, and it measured blood markers rather than illness. Funding came mostly from private donors, foundations, the National Institutes of Health and the National Science Foundation, with no food company listed⁶.

The team planned follow-up studies, including in people with immune or metabolic disease⁶. Until those report, this is one careful study, not a settled answer.

Which fermented foods still hold live microbes

In 2018, a team that included Robert Hutkins of the University of Nebraska–Lincoln pulled together published counts of live microbes in fermented foods²³. The National Dairy Council funded the work, which is worth knowing when you read the dairy numbers.

The counts below are colony-forming units (CFU), a lab measure of live cells, per gram or milliliter. They vary a lot by brand, age and storage³.

Food	Live microbes when you eat it?	What to watch	How to start
Yogurt	Usually; under 10,000 to 1 billion CFU per gram3	Added sugar in flavored cups8	⅔ cup plain yogurt with fruit
Kefir	Usually; cultured dairy runs 100,000 to 1 billion3	Added sugar in flavored bottles8	½ cup plain, alone or in a smoothie
Refrigerated sauerkraut	Often; about 1,000 to 100 million3	Sodium814	1 to 2 tablespoons a few times a week
Kimchi	Often; about 10 million to 100 million3	Sodium814	1 to 2 tablespoons with rice or eggs
Kombucha	Yes in products tested; bacteria and yeasts near 1 million to 10 million per mL3	Sugar; check the label8	4 ounces with a meal
Miso	Varies widely, about 100 to 10 million3	Sodium8	1 teaspoon stirred into soup off the boil
Cheese	From none detected to 1 billion; long-aged hard cheeses lowest3	Histamine and tyramine15	A small piece
Sourdough bread	No, baking kills them34	No live cultures	Enjoy it for flavor
Shelf-stable sauerkraut, soy sauce	No, heat treated3	Sodium8	Use for flavor, not microbes

Labels help a little. The U.S. doesn’t require companies to state microbe counts³. The voluntary “Live and Active Cultures” seal on yogurt means at least 100 million CFU per gram when the product was made³.

Yogurt got an FDA claim, with an asterisk

On March 1, 2024, the Food and Drug Administration said it would not object to a qualified health claim for yogurt⁹. Danone North America had petitioned for it in 2018¹⁰.

The claim lets labels say that eating at least 2 cups of yogurt a week, or three servings, may reduce type 2 diabetes risk⁹¹⁰. The FDA attached a big caveat: the evidence is limited⁹.

The agency’s review shows why. Danone’s petition cited 117 publications, and the FDA weighed 28 observational studies⁹. It rated nine as high quality and found their results mixed. Among moderate-quality studies that tracked new cases of diabetes, only 4 of 10 found lower risk⁹.

Observational studies can’t rule out other explanations, the FDA noted, and the trials it saw weren’t controlled well enough to help⁹.

Sugar didn’t disqualify the claim, because the research linked any benefit to yogurt as a whole food¹⁰. Still, the FDA urged companies to think carefully before putting it on heavily sweetened products, which it said could add empty calories¹⁰.

The 2025–2030 Dietary Guidelines set a tighter bar for snacks. A yogurt snack should carry no more than 2.5 grams of added sugar per two-thirds cup⁸. Check flavored cups against that number, or buy plain and add your own fruit.

Yogurt’s best-supported benefit is plainer. Live cultures help many people digest lactose, the sugar in milk, and the European Union allows that claim on yogurt with at least 100 million live cells per gram³.

Kimchi, kefir and kombucha: big reputations, thin trials

Kimchi. Korean researchers looked at diet and weight in 115,726 adults aged 40 to 69¹¹. Men who ate one to three servings of kimchi a day had roughly 11% to 13% lower odds of obesity than men who ate less than one¹¹.

The pattern was J-shaped, meaning higher intakes lost the advantage, and the authors advised against eating too much¹¹. The study captured one point in time, so it can’t show that kimchi caused anything.

Kefir. This cultured milk often carries high counts of live microbes³, and it was on the Stanford menu⁶. We didn’t find a large, independent trial of kefir alone, so enjoy it as part of a varied mix rather than a cure.

Kombucha. Reviews before 2023 found just one human study, and it had no control group¹³.

That year, a team at MedStar Georgetown University Hospital in Washington tested kombucha in adults with type 2 diabetes¹³. Volunteers drank about a cup a day of kombucha or a look-alike drink with dinner for four weeks, then switched after an eight-week break. Fasting blood sugar fell from 164 to 116 mg/dL on kombucha and from 163 to 142 on the dummy drink¹³.

The gap between the two drinks wasn’t statistically significant, and only seven people made it into the glucose analysis¹³. A kombucha company donated both drinks, and two authors had financial ties to a company called Synbiotic Health¹³. It’s a reason for a bigger trial, not for swapping out diabetes medicine.

Claim	Evidence level	What it rests on
A diet high in fermented foods raises gut microbe diversity and lowers inflammation markers	Promising but early	One 10-week trial of 36 healthy adults; main immune score unchanged57
Live yogurt cultures help digest lactose	Strong evidence	Accepted for an EU health claim at set culture levels3
Eating yogurt regularly lowers type 2 diabetes risk	Promising but early	Observational studies; FDA calls the evidence limited9
Kimchi prevents obesity	Not enough evidence	One cross-sectional study with a J-shaped pattern11
Kombucha lowers blood sugar	Not enough evidence	Pilot trial; seven finishers; no significant gap versus placebo13
Every fermented food delivers live microbes	Evidence says no, or unsafe	Bread, filtered drinks and heat-treated foods have none34
Eating lots of salt-preserved food is harmless	Evidence says no, or unsafe	Judged a probable cause of stomach cancer14

Salt, sugar and histamine: the cautions in the crock

Salt. In 2016, the World Cancer Research Fund and the American Institute for Cancer Research judged that eating foods preserved by salting probably raises the risk of stomach cancer¹⁴. Each extra 20 grams a day of salt-preserved vegetables, under an ounce, was tied to about 9% higher risk¹⁴.

In animal studies, heavy salt changed the stomach’s protective mucus, and it may help Helicobacter pylori, a bacterium that raises cancer risk, take hold¹⁴. Because these foods are eaten mostly in Asia, the panel issued no global rule, but it advised against eating them¹⁴. No study in its analysis accounted for whether people carried H. pylori¹⁴.

That’s a small amount, so daily sides of salty ferments add up. If you want a fermented food every day, plain yogurt or kefir is the lower-salt choice, and salty ferments work best as an occasional condiment. The 2025–2030 Guidelines say people 14 and older should stay under 2,300 milligrams of sodium a day⁸, so check labels.

Sugar. Flavored kombucha, kefir and yogurt can carry added sugar, and the Guidelines advise no more than 10 grams in a meal⁸.

Histamine and tyramine. Bacteria that ferment food can also make histamine, and cheese is the food most often blamed for histamine poisoning after fish, a UK food-safety advisory committee reported in 2015¹⁵. Symptoms include flushing, headache, nausea and a racing heart, and cooking doesn’t destroy the compound¹⁵. Healthy adults tolerate about 50 milligrams per meal, but people with histamine intolerance may react to tiny amounts¹⁵.

Tyramine, a related compound, matters most for people taking monoamine oxidase inhibitors (MAOIs), a class of medicines. Their per-meal limit can fall as low as 6 milligrams¹⁵. If you take an MAOI, ask your pharmacist before adding aged cheese or other fermented foods.

Weakened immunity. NIH’s Office of Dietary Supplements warns that live microbes in probiotics may not be safe for everyone, including seriously ill people and those with weakened immune systems⁴. If that’s you, ask your care team about live-culture foods, too.

When to call a clinician. See a clinician for rectal bleeding, weight loss you can’t explain, low iron, new gut symptoms later in life, or a family history of colon cancer¹². Our article on no-cost preventive care explains colorectal cancer screening coverage.

Home fermenting has its own safety rules, which a coming article covers.

A one-serving-a-day ramp-up plan

You don’t need six servings a day. The Stanford volunteers were trained to work up to that level gradually⁷. This four-week plan is our own design, built around one serving a day.

Week	What to do	Notes
1	A small serving every other day: ⅓ cup plain yogurt or kefir, or 1 tablespoon of sauerkraut or kimchi	Notice any gas or bloating
2	The same small serving every day, at the same meal	Pair it with food you already eat
3	A full serving daily: ⅔ cup yogurt or ½ cup kefir	The FDA claim counts 2 cups as three servings, about ⅔ cup each9
4	Rotate two or three types through the week, keeping salty ferments to small sides	Variety mirrors the Stanford menu6; salt matters14

Choose refrigerated products that say they contain live cultures if live microbes are your goal³, and keep an eye on sodium and sugar⁸.

Old foods, many traditions, and a federal nod

People have eaten fermented foods for thousands of years, long before anyone owned a refrigerator¹. The ISAPP panel wrote its definition to cover the thousands of fermented foods made around the world².

That history belongs to many cultures: kimchi in Korea, sauerkraut in Central Europe, miso in Japan, cultured milks in many lands. These foods carry flavor, memory and skill, and they’re worth eating for those reasons alone.

The 2025–2030 Dietary Guidelines now name sauerkraut, kimchi, kefir and miso among foods that support a diverse gut microbiome⁸. Not every claim is proved, but in broad strokes the old kitchen practice and the new research agree.

Questions readers ask

The bottom line

Fermented foods have deep roots in traditional diets, and a Stanford trial suggests a diet rich in them can broaden gut microbes and calm inflammation markers in healthy adults. That’s promising, but it was one small study, and its main immune measure didn’t change. Yogurt’s diabetes claim rests on limited evidence, and kimchi, kefir and kombucha have even less behind their specific promises.

The practical path is modest: one live-culture serving a day, plain rather than sweetened, with an eye on sodium. Know that many jars and loaves hold no live microbes at all, and check with your pharmacist or clinician if you take an MAOI or have a weakened immune system.

Related reading

• Frozen peas and candy bars are both processed, and the difference matters
• A prebiotic soda has 2 grams of fiber, and half an onion may have more
• Heritage diets hold up in research, and you can keep yours alive
• Same stool, different verdicts: what home gut microbiome tests can tell you
• Why supplement labels say ‘supports’ and never say ‘cures’

Sources

1. Marco ML, Sanders ME, Gänzle M, et al., Hutkins R. “The International Scientific Association for Probiotics and Prebiotics (ISAPP) Consensus Statement on Fermented Foods.” Nature Reviews Gastroenterology & Hepatology, 2021;18(3):196–208 (ISAPP resource page, updated September 24, 2025). isappscience.org
2. NutraIngredients. “Scientists Create International Definition for Fermented Food.” January 7, 2021. nutraingredients.com
3. Rezac S, Kok CR, Heermann M, Hutkins R. “Fermented Foods as a Dietary Source of Live Organisms.” Frontiers in Microbiology, 2018;9:1785. frontiersin.org
4. NIH Office of Dietary Supplements. “Probiotics: Fact Sheet for Consumers.” Updated November 3, 2023. ods.od.nih.gov
5. Wastyk HC, Fragiadakis GK, Perelman D, et al., Gardner CD, Sonnenburg JL. “Gut-Microbiota-Targeted Diets Modulate Human Immune Status.” Cell, 2021;184(16):4137–4153.e14 (eScholarship record). pubmed.ncbi.nlm.nih.gov
6. Stanford Medicine News Center. “Fermented-Food Diet Increases Microbiome Diversity, Decreases Inflammatory Proteins, Study Finds.” July 12, 2021. med.stanford.edu
7. Zafeiropoulou K. “Fermented Foods Show Potential in Increasing Gut Microbial Diversity and Modulating Immune Responses.” Gut Microbiota for Health, European Society of Neurogastroenterology & Motility, November 22, 2021. gutmicrobiotaforhealth.com
8. U.S. Department of Health and Human Services and U.S. Department of Agriculture. “Dietary Guidelines for Americans, 2025–2030.” January 2026. cdn.realfood.gov
9. Healio. “FDA to Allow Limited Claims That Eating Yogurt May Reduce Risk for Type 2 Diabetes.” March 4, 2024. healio.com
10. FoodNavigator. “FDA Weighs Application of New Yogurt Qualified Health Claims on Sugary Products.” March 6, 2024. foodnavigator.com
11. Jung H, Yun YR, Hong SW, Shin S. “Association Between Kimchi Consumption and Obesity Based on BMI and Abdominal Obesity in Korean Adults: A Cross-Sectional Analysis of the Health Examinees Study.” BMJ Open, 2024;14(2) (DOAJ record). pubmed.ncbi.nlm.nih.gov
12. American College of Gastroenterology. “Irritable Bowel Syndrome: Overview.” ACG, updated February 2026. gi.org
13. Mendelson C, Sparkes S, Merenstein DJ, et al., Hutkins R. “Kombucha Tea as an Anti-Hyperglycemic Agent in Humans With Diabetes: A Randomized Controlled Pilot Investigation.” Frontiers in Nutrition, 2023;10. frontiersin.org
14. World Cancer Research Fund and American Institute for Cancer Research. “Preservation and Processing of Foods” (Continuous Update Project evidence, including the 2016 stomach cancer report). WCRF. wcrf.org
15. Advisory Committee on the Microbiological Safety of Food (UK Food Standards Agency secretariat). “Histamine in Cheese” (discussion paper ACM/1193, citing EFSA’s 2011 opinion on biogenic amines). October 2015.