What consumer stool microbiome tests can and cannot tell you

What it is

These are mail-order kits that sequence bacterial DNA from a stool sample, usually by 16S rRNA gene amplicon sequencing or shotgun metagenomics, and return a report of taxa, a diversity score and often diet or supplement suggestions. They are sold without a prescription and sit between regulated medical devices and unregulated wellness products. Two independent evaluations have sent a single standardised or split sample to several companies at once.

What the research says

They are marketed as reading your gut health; the published evaluations say the analytical performance does not yet support that. NIST researchers sent a standardised human faecal material to seven direct-to-consumer services and found differences between providers as large as the biological differences between different donors. A European expert panel sent one faecal sample to six companies and judged the interpretations and recommendations premature and of limited clinical utility. An international consensus statement concluded that evidence for clinical usefulness is scarce.

Evidence grade: Not supported by the research.

What goes wrong

Some services recommended that customers buy supplements the same company sells, for which the evaluators say there is very little clinical evidence of efficacy. (Source 1)

• Survey study, Low certainty.
• Size: 7 evaluated services.
• Who: consumers of direct-to-consumer tests.
• How long: not applicable.
• Result: no quantified harm; the authors describe loss of consumer confidence and delayed medical care as consequences of unreliable reports.
• Funding: no funding statement retrieved.

Additionally, some companies recommended that customers start taking costly supplements (e.g., probiotics) that are sold by the same company and for which there is very little clinical evidence for efficacy.

The evaluators warn that for people with chronic gut conditions who buy these tests, variable results or unmet expectations could delay appropriate medical care. (Source 1)

• Survey study, Low certainty.
• Size: 7 evaluated services.
• Who: consumers.
• How long: not applicable.
• Result: qualitative risk statement, no incidence measured.
• Funding: no funding statement retrieved.

For these individuals, the variability in test results or the lack of expected outcomes from the recommendations provided, could not only lead to a loss of consumer faith in the science but also result in the delay of appropriate medical care.

An international multidisciplinary consensus concluded that evidence supporting the clinical usefulness of gut microbiome diagnostics is scarce and that unregulated consumer tests could waste resources and interfere with patient management. (Source 2)

• Official position, Certainty not rated.
• Size: international multidisciplinary expert panel.
• Who: not applicable.
• How long: not applicable.
• Result: no effect estimate; a framework of minimum requirements was proposed instead.
• Funding: not stated in the abstract retrieved.

An increasing number of commercial providers offer direct-to-consumer microbiome diagnostic tests without any consensus on their regulation or any proven value in clinical practice, which could result in considerable waste of individual and health-care resources and potential drawbacks in the clinical management of patients.

What the evidence supports

There is a legitimate, narrowly defined diagnostic use of microbiota testing, namely detecting a transmissible agent before transplantation, which the European panel placed in the regulated in vitro diagnostic category. (Source 3)

• Official position, Certainty not rated.
• Size: expert panel of 21 experts from 8 countries.
• Who: not applicable.
• How long: not applicable.
• Result: this use case is classed as a potential Class D IVD rather than a wellness product.
• Funding: EU-funded IHMCSA consortium.

As Class D, if the test is designed for specific purposes, such as detection of the presence of a transmissible agent before transplantation.

A review describes gut bacteria as a potential biomarker for inflammatory bowel disease, that is, as something that might one day be used to assess the disease rather than as a test that has been validated for clinical use. (Source 4)

• Expert review, not systematic, Very low certainty.
• Size: narrative review; no pooled sample size stated.
• Who: patients with ulcerative colitis and Crohn's disease compared with healthy individuals.
• How long: not applicable.
• Result: no sensitivity, specificity or cut-off values are reported in the abstract; the named differences are higher Proteobacteria and lower Bacteroides, Eubacterium and Faecalibacterium in patients than in healthy individuals, and the authors frame the clinical use as potential.
• Funding: government and provincial research grants (National Natural Science Foundation of China and others), as stated by the authors.

Limit of this finding: This is a narrative review, not a study of its own, and it reports no accuracy figures at all: no sensitivity, no specificity, no cut-off. Its own wording is that microbial biomarkers have 'the potential to change clinical practice', which is a statement about promise, not evidence that any such test works. One further sentence in the passage recorded from this review - about gut microbiota being crucial and intestinal dysbiosis being associated with the disease - is ungrammatical in the original, and it is deliberately not used as a quotation here.

Understanding the role of microbial biomarkers in the assessment of disease activity and treatment outcomes has the potential to change clinical practice and lead to the development of more personalized therapies.

What the evidence does not support

Seven direct-to-consumer gut microbiome services tested on one standardised faecal material disagreed with each other as much as different people's guts differ. (Source 5)

• Survey study, Moderate certainty.
• Size: 7 companies, three kits ordered per company, one NIST-developed standardised human faecal material.
• Who: not applicable (laboratory evaluation of commercial services)
• How long: single time point.
• Result: major discrepancies within and across providers; between-provider variability on the same scale as biological variability between donors; no method matched the consensus relative abundance range for all 18 taxa analysed.
• Funding: conducted by US National Institute of Standards and Technology researchers with academic co-authors; no funding statement retrieved.

Significantly, we found variability between providers was on the same scale as biological variability between different donors.

No provider reported abundances inside the consensus range for every organism measured. (Source 6)

• Survey study, Moderate certainty.
• Size: 18 taxa across the participating methods.
• Who: commercial testing services.
• How long: single time point.
• Result: zero of the evaluated methods were concordant for all 18 taxa.
• Funding: no funding statement retrieved.

Of note, no method reported concordance with the consensus relative abundance range for all 18 taxa analyzed.

A European expert panel that sent one faecal sample to six companies judged the resulting interpretations and recommendations premature and of limited clinical utility. (Source 7)

• Official position, Certainty not rated.
• Size: 6 companies (five in Europe, one in the USA); panel of 21 experts from 8 countries.
• Who: not applicable.
• How long: single sample, one workshop.
• Result: identified over-promising messages, lack of transparency about methods and lack of reliability of results.
• Funding: part of the EU-funded International Human Microbiome Coordination and Support Action (IHMCSA) consortium.

The experts considered the interpretations and recommendations provided in the different reports to be premature due to the lack of robust scientific evidence and the analyses associated with the reports to be of limited clinical utility.

Even in research settings, taxonomic profiling varies substantially with specimen handling and laboratory protocol. (Source 8)

• Survey study, Moderate certainty.
• Size: blinded specimen sets sequenced by 15 laboratories and analysed with nine bioinformatics protocols.
• Who: human stool, chemostats and artificial microbial communities.
• How long: single baseline study.
• Result: variability attributable in order to biospecimen type and origin, DNA extraction, sample handling environment, then bioinformatics; artificial communities revealed differences in extraction efficiency and classification.
• Funding: US National Institutes of Health (NHGRI, NCI, NIDCR, NIDDK)

Achieving sufficient reproducibility in microbiome research has proven challenging.

The same review states that inflammatory bowel disease still lacks specific markers, and that the established blood and stool markers measure inflammation but are not specific to the disease and have no agreed threshold for it. (Source 4)

• Expert review, not systematic, Very low certainty.
• Size: narrative review; no pooled sample size stated.
• Who: patients with inflammatory bowel disease.
• How long: not applicable.
• Result: no cut-off value is established for C-reactive protein, faecal calprotectin or S100A12 in this indication, per the authors.
• Funding: government and provincial research grants (National Natural Science Foundation of China and others), as stated by the authors.

Limit of this finding: The quoted sentence is awkwardly built in the original: 'difficult to determine an effective cut-off value' is the review's own phrasing, reproduced as printed. What it means is that no agreed threshold has been settled on for C-reactive protein, faecal calprotectin or S100A12 in inflammatory bowel disease, not that a threshold exists and is merely awkward to apply. This is a narrative review summarising other work, so it carries no data of its own.

While traditional biomarkers from plasma and stool, such as C-reactive protein (CRP), fecal calprotectin (FC), and S100A12, can be used to measure inflammation, they are not specific to IBD and difficult to determine an effective cut-off value.

Where the research disagrees

Whether direct-to-consumer microbiome tests have clinical value now

• International consensus panel (Porcari et al., Lancet Gastroenterology & Hepatology, 2025), international multidisciplinary consensus statement: An increasing number of commercial providers offer direct-to-consumer microbiome diagnostic tests without any consensus on their regulation or any proven value in clinical practice (Source 2)
• European workshop experts (Rodriguez et al., Microbiome, 2024), on what a regulated test could legitimately do, expert workshop after sending one faecal sample to six companies: As Class D, if the test is designed for specific purposes, such as detection of the presence of a transmissible agent before transplantation. (Source 3)

A common belief, and what the research shows

The belief: An at-home gut test reads your microbiome and tells you what to eat.

What the research shows: Two independent evaluations sent one sample to multiple companies and got incompatible reports. The NIST group found "major discrepancies, both within and across the different service providers", with between-provider variability on the same scale as the difference between different donors, and the European panel judged the reports' recommendations premature and of limited clinical utility. Some companies then recommended supplements they sell themselves.

Questions and answers

A consumer gut microbiome test is a posted stool kit whose DNA is sequenced by a company that then sends back a report on your bacteria, usually with a diversity score and diet or supplement suggestions. They are sold directly to the public, sometimes by medical laboratories but without any requirement for a prescription. They occupy a regulatory grey zone between medical devices and wellness products. (Source 7)

The test reports which bacterial sequences the company's pipeline detected in the sample it received. What it does not do reliably is measure them the same way another company would: when NIST researchers sent a standardised faecal material to seven services, differences between providers were as large as differences between different people. The report's diet and supplement advice is generated from those numbers. (Source 5)

As a clinical tool the published evaluations are negative: an international consensus panel found evidence for clinical usefulness scarce, and the European panel called the reports' interpretations premature. There is no agreed threshold for what counts as a healthy or good microbiome, so the verdicts printed on a report are not anchored to anything. A narrow regulated use, screening a donor for transmissible agents before transplantation, is different in kind. (Source 2)

Does not apply in the usual sense: this is a test, not a substance. The relevant question is whether buying more tests adds information, and the evidence says repeat results are unstable even within a provider, because three kits per company applied to one standardised material still produced discrepancies within providers as well as between them. (Source 9)

Does not apply to a test. The harm the literature names is not the sample itself but what follows the report: services recommending their own supplements, and the possibility that an unreliable result delays proper medical care. The consensus statement's remedy is regulation and minimum requirements rather than avoidance. (Source 2)

Does not apply directly. The analogous question is why a test might miss organisms that are present, and the Microbiome Quality Control project answers it: detection depends on the specimen itself, then on DNA extraction, the handling environment and the bioinformatics pipeline, which differ between companies. Artificial community specimens exposed differences in extraction efficiency and classification between laboratories. (Source 8)

Does not apply: no food contains or feeds a test. Reports commonly translate their results into food and supplement advice, and that translation is the part the European expert panel called premature, citing a lack of robust scientific evidence behind the recommendations. (Source 7)

Nothing in the literature suggests a person is worse off for not taking one of these tests, because no study shows that acting on the reports improves any health outcome. The published concern runs the other way: consensus authors warn of wasted individual and health-care resources and potential drawbacks in the clinical management of patients. (Source 2)

This topic is the test, and its reliability has now been measured twice by independent groups. Seven services given one standardised faecal material disagreed as much as different donors do, and no method reported all 18 taxa within the consensus range. Six companies given one split sample produced reports the expert panel considered unreliable and of limited clinical utility. (Source 6)

If I send the same sample to several companies, will I get the same answer?

No. NIST researchers ordered three kits from each of seven direct-to-consumer services and analysed one standardised human faecal material: results differed both within and between providers, and the spread between providers was as wide as the biological difference between different people's guts. No method reported all 18 measured taxa within the consensus abundance range. A separate European group sent one faecal sample to six companies and reached the same conclusion about reliability.

Can a consumer microbiome test tell me whether my gut is healthy?

Not on current evidence. The European expert panel states that without a scientific consensus on thresholds it is impossible to say whether a microbiome is healthy, and the 2025 international consensus statement found the evidence for clinical usefulness scarce. Analytical performance comes first: the NIST authors point out that it is a prerequisite for making sound clinical recommendations, and it has not been demonstrated.

Is there anything these tests are legitimately used for?

The European panel identifies detecting a transmissible agent before transplantation as a use that belongs in the regulated in vitro diagnostic category rather than the wellness category. Research uses individual organisms as candidate biomarkers, for example Faecalibacterium prausnitzii to help discriminate ulcerative colitis from Crohn's disease, but that is a proposal in a review, not a validated diagnostic. The consensus statement's aim is to build the framework under which such tests could eventually be developed.

References

1. Communications Biology. Evaluating the analytical performance of direct-to-consumer gut microbiome testing services [Discussion]. 2026. DOI 10.1038/s42003-025-09301-3. Read the source
2. The Lancet Gastroenterology and Hepatology (abstract read from the University of Copenhagen research-profiles repository record). International consensus statement on microbiome testing in clinical practice. 2025. PMID 39647502, DOI 10.1016/S2468-1253(24)00311-X. Read the source
3. Microbiome. Microbiome testing in Europe: navigating analytical, ethical and regulatory challenges [body text, IVD risk classification]. 2024. PMID 39695869, DOI 10.1186/s40168-024-01991-x. Read the source
4. Frontiers in Nutrition. Gut Microbiota Is a Potential Biomarker in Inflammatory Bowel Disease. 2022. DOI 10.3389/fnut.2021.818902. Read the source
5. Communications Biology. Evaluating the analytical performance of direct-to-consumer gut microbiome testing services. 2026. DOI 10.1038/s42003-025-09301-3. Read the source
6. Communications Biology. Evaluating the analytical performance of direct-to-consumer gut microbiome testing services [Results, "Taxonomic profiles varied across companies"]. 2026. DOI 10.1038/s42003-025-09301-3. Read the source
7. Microbiome. Microbiome testing in Europe: navigating analytical, ethical and regulatory challenges. 2024. PMID 39695869, DOI 10.1186/s40168-024-01991-x. Read the source
8. Nature Biotechnology. Assessment of variation in microbial community amplicon sequencing by the Microbiome Quality Control (MBQC) project consortium. 2017. PMID 28967885, DOI 10.1038/nbt.3981. Read the source
9. Communications Biology. Evaluating the analytical performance of direct-to-consumer gut microbiome testing services [Results, "Experimental design strategy"]. 2026. DOI 10.1038/s42003-025-09301-3. Read the source