The gut microbiome is real and important, and most of what is sold on the strength of it is not proven in people. Trillions of microbes, most in the colon, ferment the fiber you cannot digest into short-chain fatty acids that feed the colon lining and signal to your metabolism and immune system. That much is solid.
One deliberate change to the microbiome is a clear clinical win: a fecal transplant clears a recurrent C. difficile infection in about 9 in 10 people. Almost everything else sold to consumers, the stool test that scores your health, the probiotic that rebuilds you, the gut fix for weight or mood, runs well ahead of what has been shown in people.
Findings & Outcomes
What The Microbiome Is
Your large intestine holds trillions of bacteria, along with viruses, fungi, and other microbes, most of them in the colon where digestion has largely finished. Together they carry many times more genes than your own cells, and they run chemistry your body cannot. The collection works as a metabolic and immune organ: it digests what you cannot, trains the immune system, and produces molecules that circulate to the rest of the body.
Fermentation is the clearest of these. Fiber that survives your own digestion reaches the colon intact, and resident bacteria ferment it into short-chain fatty acids, mainly acetate, propionate, and butyrate. Butyrate is the main fuel the cells lining your colon burn, and all three act as signals that reach metabolism and the immune system. This is the best-established route by which the microbiome affects the rest of you, and the reason fiber comes up whenever the microbiome is discussed.
What Reliably Shapes It
Diet is the strongest lever, and the best human evidence points at food, not supplements. In a 17-week randomized trial in healthy adults, a diet high in fermented foods raised gut microbial diversity and lowered a panel of inflammatory immune markers. A high-fiber diet in the same trial did not raise diversity across the board; its effect depended on the microbiome each person started with, and it showed up more in what the microbes could do than in their variety. Food changing both the microbiome and the immune system is about as strong as human evidence in this field gets, and it still stops short of a hard clinical outcome.
Swallowing bacteria is a weaker lever than it sounds. When healthy volunteers took a multi-strain probiotic and had their gut lining sampled directly by endoscopy, the strains did not durably colonize the gut in most people. Some guts resisted the strains entirely; others let them establish only while the supplement continued; and stool samples did not reflect what was happening at the gut wall. A probiotic reseeding or permanently rebuilding your microbiome is not what the direct measurements show. Specific strains still have specific evidence for specific situations, weighed claim by claim on the probiotics page.
The One Clear Clinical Win
Deliberately replacing the microbiome is unambiguously effective in exactly one place. Recurrent Clostridioides difficile infection keeps returning after antibiotics, and a fecal transplant, an infusion of a healthy donor's stool, treats it. In the trial that established the approach, a single infusion resolved the infection in 81% of patients and a second brought it to 94%, against about 31% for the antibiotic vancomycin alone. The difference was so large the trial was stopped early. Pooling 45 studies, repeat transplant clears recurrent C. difficile in roughly 91% of people, rated high-quality evidence, which is uncommon in microbiome medicine.
A fecal transplant is proven for recurrent C. difficile and nothing else. It has not become an effective treatment for obesity, autism, or depression, and it is not a reason to reach for a probiotic pill for general health.
The result is narrow because the goal is narrow: restore a whole community at once to crowd out the one organism that has taken over.
Where The Bigger Claims Sit
The most repeated microbiome claims are about obesity, metabolism, mood, and autoimmune disease, and this is where the evidence thins sharply. In animals it can be dramatic. Transplant gut bacteria from a heavier human into germ-free mice and the mice gain more fat than mice given a leaner human's bacteria, and an obese-type community extracts more energy from the same food. These are clean causal experiments, and they are done in mice.
In people the picture is associational and much weaker. When one team reanalyzed 10 human datasets together, the difference in microbial diversity between people with and without obesity was about 2%, most studies were too small to detect even a 5% difference, and a model trained to guess obesity from gut bacteria performed close to a coin flip on new data. The once-popular idea that a high ratio of Firmicutes to Bacteroidetes marks an obese gut did not hold up on reanalysis. The signal is small, inconsistent, and cannot tell an individual anything to act on.
That gap is what a direct-to-consumer stool test sells across. It reports your microbiome against exactly the kind of weak association above, then converts it into a health score and, often, a supplement to buy. The test measures something; the interpretation runs far ahead of the evidence. No validated map exists from a stool-test profile to what you should eat or take, and the same sample sent to different companies can come back with different verdicts. The grounded use of the science today is dull: eat a range of plants and some fermented foods, and do not treat the report as a diagnosis.
The Research & Studies
Everything here is based on the research we have collected and checked, sorted into groups and ordered with the strongest evidence first. Click any claim to open the studies behind it.
How it works
Gut bacteria ferment fiber into short-chain fatty acids that fuel the colon lining
The part of microbiome biology that is solid: gut bacteria ferment the fiber you eat into short-chain fatty acids, and butyrate is the main fuel the lining of your colon runs on.
Koh and colleagues review the pathway from dietary fiber to host physiology. Fiber that reaches the colon undigested is fermented by resident bacteria into short-chain fatty acids, principally acetate, propionate and butyrate. Butyrate is the preferred energy substrate of colonocytes, the cells lining the large intestine. All three molecules also act as signals: they activate G-protein-coupled receptors such as GPR41 and GPR43 and inhibit histone deacetylases, which links them to appetite hormones, glucose handling and the regulation of immune cells. This is the best-characterized route by which what you eat and which microbes you carry reach the rest of the body.
Who this may not transfer to:A review of the fiber-to-SCFA pathway, not a measurement in a defined group of people, so it describes the mechanism rather than an effect size in any population.
The study · 1
Koh et al., from dietary fiber to host physiology: short-chain fatty acids as key bacterial metabolites · Cell 2016;165(6):1332-1345
Swallowed probiotics did not durably colonize the gut in most people
Swallowing a probiotic does not mean it moves in. In a study that looked directly at the gut wall, many people's guts rejected the strains, and where the strains did stick it was temporary.
Zmora and colleagues gave healthy volunteers an 11-strain probiotic and then mapped colonization along the gut using upper and lower endoscopy, rather than relying on stool alone. Colonization was highly person-specific: some individuals were permissive and let the strains establish along the mucosa, while others resisted them, and the pattern could be predicted from a person's own baseline microbiome and host features. Even where strains established, the effect faded after supplementation stopped. Stool composition did not track mucosal colonization, meaning stool testing can misrepresent whether a probiotic reached the gut wall at all.
Who this may not transfer to:Healthy adult volunteers of both sexes, with the gut wall sampled directly; colonization was person-specific, so the pattern is individual rather than universal.
The study · 1
Zmora et al., personalized gut mucosal colonization resistance to empiric probiotics · Cell 2018;174(6):1388-1405
In mice, gut microbes from an obese person drove more fat gain; not shown in people
In mice, transplanting gut bacteria from a heavier human made the mice gain more fat than bacteria from a leaner human, which shows the microbiome can drive weight in an animal. Whether changing your microbiome changes your weight has not been shown in people.
Ridaura and colleagues transplanted fecal microbiota from adult female twin pairs discordant for obesity into germ-free mice. Increased body and fat mass and obesity-associated metabolic traits transferred with the obese-twin communities. When mice carrying obese-twin microbiota were housed with mice carrying lean-twin microbiota, specific Bacteroidetes from the lean community invaded the obese community and prevented the weight gain, and this rescue depended on the mice being fed a diet low in saturated fat and high in fruit and vegetables. The experiment is a clean demonstration that the microbiome can causally influence adiposity in an animal under controlled diet.
Who this may not transfer to:A controlled experiment in germ-free mice given human donor microbes; it demonstrates causation in an animal and has not been reproduced as weight change in people.
The study · 1
Ridaura et al., gut microbiota from twins discordant for obesity modulate metabolism in mice · Science 2013;341(6150):1241214
In mice, an obese-type gut community pulled more energy from food and passed the trait on
Part of why the microbiome is linked to weight: in mice, an obese-type gut community pulls more calories out of the same food and passes that tendency along when it is transplanted.
Turnbaugh and colleagues found that the gut microbiome of genetically obese mice was enriched for genes that harvest energy from the diet, and that this community contained proportionally fewer Bacteroidetes and more Firmicutes. Colonizing germ-free mice with the obese microbiome produced a significantly greater increase in body fat than colonizing them with a lean microbiome, indicating the trait was transmissible with the microbes themselves. It is the founding energy-harvest mechanism behind the idea that gut bacteria influence body weight.
Who this may not transfer to:Measured in obese and lean mice; a plausible mechanism whose size in humans, who eat varied diets, is not established.
The study · 1
Turnbaugh et al., an obesity-associated gut microbiome with increased capacity for energy harvest · Nature 2006;444(7122):1027-1031
Digestion
Fecal transplant cleared recurrent C. difficile in 81% after one infusion, 94% after two, vs 31% on vancomycin
For a gut infection that keeps coming back, a fecal transplant cured about 8 in 10 people after one dose and roughly 9 in 10 after two, while standard antibiotics cured only about 3 in 10. The trial was halted early because the gap was so wide.
Patients with recurrent C. difficile infection were randomly assigned to donor-feces infusion through a nasoduodenal tube after a short vancomycin course and bowel lavage, to standard vancomycin for 14 days, or to vancomycin with bowel lavage. The primary endpoint was resolution of C. difficile-associated diarrhea without relapse at 10 weeks. Of 16 in the infusion group, 13 (81%) resolved after the first infusion and 15 of 16 (94%) after a second, compared with 4 of 13 (31%) on vancomycin alone and 3 of 13 (23%) on vancomycin plus lavage. An interim analysis found the effect so large that the trial was stopped early.
Who this may not transfer to:Adults of both sexes with recurrent C. difficile; the result is specific to that infection and does not extend to fecal transplant for other conditions.
The study · 1
van Nood et al., duodenal infusion of donor feces for recurrent Clostridium difficile · N Engl J Med 2013;368(5):407-415
Across 45 studies, repeat fecal transplant cleared recurrent C. difficile in 91% of people
Pooling 45 studies, a repeated fecal transplant cleared recurrent C. difficile in about 91% of people, and you needed to treat only about 3 people for 2 to be cured who would not have been by antibiotics alone.
This updated systematic review and meta-analysis pooled 45 studies of fecal microbiota transplantation for recurrent C. difficile infection. The clinical resolution rate at week 8 was 91% (95% CI 89 to 94%) after repeat transplant across 24 studies and 1,855 patients, and 84% (80 to 88%) after a single transplant across 43 studies and 2,937 patients. Delivery by lower gastrointestinal endoscopy was superior to other routes. Against vancomycin, the number needed to treat was 1.5 for repeat transplant and 2.9 for single transplant. Repeat transplant was rated high quality of evidence under GRADE, which is uncommon in this field.
Who this may not transfer to:Pooled across 45 studies of both sexes; it applies to recurrent C. difficile, not to fecal transplant for other conditions.
The study · 1
Baunwall et al., faecal microbiota transplantation for recurrent Clostridioides difficile infection: an updated systematic review and meta-analysis · EClinicalMedicine 2020;29-30:100642
Measurement And Diagnosis
The obese-versus-lean microbe-diversity gap was about 2%, and predicting obesity from gut bacteria was near chance
The claim that obese and lean people have clearly different microbiomes is weaker than it sounds. Across 10 studies the difference in microbe variety was about 2%, and a computer trained to guess obesity from gut bacteria was barely better than a coin flip on new data.
Sze and Schloss applied systematic-review methods to identify 10 datasets and reanalyzed them together. Pooling with a random-effects model, they found statistically significant but tiny associations between obesity and diversity measures such as the Shannon index and the number of observed taxa; the diversity difference between non-obese and obese individuals was 2.07%. A power analysis showed only one study was large enough to detect even a 5% difference. The much-cited ratio of Firmicutes to Bacteroidetes was not associated with obesity. When machine-learning classifiers were trained on one dataset and tested on the other nine, median accuracy ranged from 33 to 65%, close to chance. The signal is small, inconsistent, and not usable at the level of an individual person.
Who this may not transfer to:Pooled human datasets where sex was not consistently reported; the association is real but too small and inconsistent to apply to an individual.
The study · 1
Sze & Schloss, looking for a signal in the noise: revisiting obesity and the microbiome · mBio 2016;7(4):e01018-16
Immune Function
A 17-week fermented-foods diet raised microbial diversity and lowered inflammatory markers
Eating more fermented foods like yogurt, kefir and kimchi for a few months raised gut-microbe variety and calmed several markers of inflammation. More fiber helped too, but its effect depended on the microbiome each person started with.
In a 17-week randomized prospective trial, healthy adults were assigned to a diet high in fermented foods or high in plant fiber. The fermented-food arm showed a steady rise in gut microbial diversity and a decrease in a panel of inflammatory immune proteins, including markers such as interleukin-6. The high-fiber arm did not increase diversity uniformly; responses were personalized, driven by each participant's baseline microbiome, and appeared more in the microbes' functional capacity than in raw diversity. This is the strongest kind of human evidence in the field, food changing both the microbiome and immune signalling, and it still stops short of a hard clinical outcome.
Who this may not transfer to:36 healthy adults over 17 weeks; a small trial in healthy people, and it has not been tested against a disease outcome.
The practical read is that ordinary fermented foods are a low-risk way to nudge the microbiome and inflammation, more reliably than most probiotic supplements, while fiber works best matched to your own gut rather than as a single universal prescription. Neither has yet been shown to change a disease endpoint, so treat them as sensible food habits, not a therapy.
The study · 1
Wastyk et al., gut-microbiota-targeted diets modulate human immune status · Cell 2021;184(16):4137-4153
The grades are uneven on purpose. The short-chain fatty acid mechanism and the fecal-transplant result for C. difficile rest on firm ground. The obesity rows are graded preliminary or left without a direction, because the strong causal work is in mice and the human data is a weak association.
Go Deeper
This page describes the ecology. The pages that act on it or sit next to it:
- The gut-brain axis, the specific wiring between the gut and the brain, where the microbiome-to-mood claims are weighed.
- Fiber, the raw material the microbiome ferments into short-chain fatty acids.
- Fermented foods, the food lever with the best human evidence for shifting the microbiome.
- Probiotics, where the evidence for swallowing bacteria is weighed strain by strain.
- Irritable bowel syndrome, the common gut condition where microbiome-based hopes and treatments meet the clinic.
Common Questions
Is the gut microbiome real science or hype?
Both, in different layers. The biology holds: trillions of microbes in the colon ferment fiber into short-chain fatty acids and act as a metabolic and immune organ, and replacing the microbiome cures recurrent C. difficile infection. What outruns the evidence is the jump from that to fixing weight, mood, or immunity by buying a product. The mechanism is established. Most of the widely repeated consumer claims are still animal work or weak associations in people.
Are microbiome stool tests worth it?
Not for guiding what you eat or take. A consumer stool test measures something, but it turns weak and inconsistent associations into a confident-looking health score, there is no validated map from that profile to a personal recommendation, and the same sample can return different results from different companies. It reads as precision it does not have. For persistent digestive symptoms, the useful next step is a clinician, not a test bought online.
Do probiotic supplements rebuild my gut?
Usually not the way the label implies. When researchers looked directly at the gut wall, the strains in a probiotic failed to durably colonize most people, took hold only while the supplement continued when they took hold at all, and did not show up reliably in stool. Particular strains do have evidence for particular situations. Reseeding or permanently rebuilding your microbiome with a capsule is not what the measurements show.
What actually improves my microbiome?
Food, more than supplements. The best human trial found that a diet high in fermented foods raised microbial diversity and lowered inflammatory markers, and that more fiber helped in a way tailored to each person's starting gut. The practical version is well supported: eat a wide range of plants for the fiber, include some fermented foods, and give it weeks. Neither has been shown to change a disease outcome, so treat them as sensible habits.
Can changing my gut bacteria make me lose weight?
Not on the current evidence in people. The dramatic experiments where an obese-type microbiome makes an animal gain fat are done in mice. In humans the link between the microbiome and obesity is a small, inconsistent association, and no one has shown how to use it for lasting weight loss. The microbiome is part of the story of metabolism; it is not yet a way to change your weight.
The Chinese Medicine Reading
Explore Related
Other pages this one connects to, by the evidence they share, the outcomes they touch, and the ground they cover.
All 8 sources on this page independently checked and cross-referenced.
Thomas Dehli, Founder & Editor, Sacred Lotus
Sacred Lotus has published Chinese medicine reference material since 2001. Integrative pages are held to the same standard as the herb and formula library: cite the source, grade the claim at its real strength, and say where the research has not looked. This page is educational and it is not medical advice. Last reviewed and updated August 10, 2026.
Evidence strength
How confidently the research supports a claim. Strength describes the evidence, not our endorsement.