Heterogeneity and lack of standardisation in gut microbiome testing: a comparative assessment of French medical biology laboratories Authors: Maxime Pichon, Claire Bouleti, Corentine Alauzet, Genevieve Hery-Arnaud, Christophe Burucoa Published: 2026 We read with interest the review by de Vos et al.1 The human gut microbiome plays a pivotal role in the maintenance of health and the development of disease.2 3 With the growing availability of microbiome testing products and publications, some patients feel that their health problems are not fully understood or addressed by doctors using conventional medical approaches.4 Concomitantly with a rise in consumer interest in self-testing, questions have emerged regarding the scientific validity, regulatory status and clinical interpretability of microbiome testing kits.5 6 In this context, we aimed to evaluate, on a nationwide French scale, the methods employed, the nature and quality of the results obtained, and the putative benefits that non-specialists/patients can derive from these readily accessible microbiota analyses. Five French laboratories offering microbiome analysis without prior consultation were selected via standard internet searches by the three microbiology researchers involved in the study, mimicking a patient’s approach in June 2024. The patient was a fictitious 40-year-old man, with a normal BMI according to the WHO, no notable medical or surgical history apart from the reason for carrying out the intestinal microbiome, and no dietary particularities. A homogenised stool sample was prepared from a composite mixture of three male individuals from a cardiovascular cohort (in the acute phase of ST elevation myocardial infarction, characteristics in online supplemental table 1), then split and sent to each laboratory according to their recommendations. Kit components, packaging safety, questionnaires, bioinformatic pipelines and final reports were compared. A high degree of heterogeneity was observed among the selected laboratories. The content of the kits exhibited significant variations, encompassing prescriptions for procedures and instructions for the accurate collection of samples, and a series of promotional materials unrelated to any biomedical activity (online supplemental table 2). Each analysis kit was accompanied by a questionnaire designed to collect demographic, clinical and/or ethical data (table 1), including required consent for research participation. Irrespective of the approach used or the analytical conditions, a multitude of discrepancies were observed between the laboratories in terms of form (report type, length and delay) and the regulatory considerations involved (UN3373 triple-packaging) and in carrying out any medical biology examination in France (identification of a medical biologist by name and indication of subcontracting). Report lengths varied from 9 to 36 pages. No laboratory supplied raw sequencing data and analytical approaches ranged from 16S metabarcoding to shotgun metagenomics, with scant disclosure of pipeline versions or metrics (table 2). Based on alpha-diversity metrics showing biologically implausible variation (ranging from 3.64 to 6.11), interpretations were contradictory, with opposing classifications of ‘eubiosis’ and ‘dysbiosis’ and unsubstantiated health claims. Health or dietary recommendations were often non-personalised, citing dubious or unreferenced sources and were reinforced by the recommendation to consult a specialist. The relationship between the medical laboratory and the pharmaceutical company gives rise to concerns regarding a potential conflict of interest (automatic prescription) as well as a risk to patient health (debated therapeutic solutions). Table 1Composition of the kits receivedLaboratoryABCDE Composition of the kits received CostPresence of advertisingOther information €249.00Yes–€208.50No>6 years old€204.95No>4 years old€199.00Yes>15 years old€199.00No>15 years old General: birth Mode/placeNot determinedNot determinedMode/placeMode/place General: breastfeeding Maternal/artificial/mixedNot determinedNot determinedMaternal/artificial/mixedMaternal/artificial/mixed Clinical information Bristol ScaleNumber of stools per dayType of digestive symptoms (bloating, diarrhoea, constipation…) Type of symptoms (digestive and extradigestive; exhaustive list)Type of digestive symptoms (bloating, diarrhoea, constipation…) Bristol ScaleType of digestive symptoms (bloating, diarrhoea, constipation…) Bristol ScaleType of digestive symptoms (bloating, diarrhoea, constipation…) Familial medical history Yes/no (exhaustive list)–––– Surgical medical history Appendicectomy––Yes/no (exhaustive list)Yes/no (exhaustive list) Personal medical history Yes/no (exhaustive list)Therapeutic consumption (≤3 months ago)COVID-19 history (including vaccination)Tobacco consumption Yes/no (exhaustive list)Current therapeutic consumptionTobacco consumption Yes/no (exhaustive list)Therapeutic consumption (≤6 months ago)Tobacco consumption Yes/no (exhaustive list)Therapeutic consumption (≤1 to 3 months depending on the molecule)Tobacco consumption Yes/no (exhaustive list)Therapeutic consumption (≤1 to 3 months depending on the molecule) Composition of the kits received Cost Cost Presence of advertising Presence of advertising Other information Other information Bristol Scale Bristol Scale Number of stools per day Number of stools per day Type of digestive symptoms (bloating, diarrhoea, constipation…) Type of digestive symptoms (bloating, diarrhoea, constipation…) Bristol Scale Bristol Scale Type of digestive symptoms (bloating, diarrhoea, constipation…) Type of digestive symptoms (bloating, diarrhoea, constipation…) Bristol Scale Bristol Scale Type of digestive symptoms (bloating, diarrhoea, constipation…) Type of digestive symptoms (bloating, diarrhoea, constipation…) Yes/no (exhaustive list) Yes/no (exhaustive list) Therapeutic consumption (≤3 months ago) Therapeutic consumption (≤3 months ago) COVID-19 history (including vaccination) COVID-19 history (including vaccination) Tobacco consumption Tobacco consumption Yes/no (exhaustive list) Yes/no (exhaustive list) Current therapeutic consumption Current therapeutic consumption Tobacco consumption Tobacco consumption Yes/no (exhaustive list) Yes/no (exhaustive list) Therapeutic consumption (≤6 months ago) Therapeutic consumption (≤6 months ago) Tobacco consumption Tobacco consumption Yes/no (exhaustive list) Yes/no (exhaustive list) Therapeutic consumption (≤1 to 3 months depending on the molecule) Therapeutic consumption (≤1 to 3 months depending on the molecule) Tobacco consumption Tobacco consumption Yes/no (exhaustive list) Yes/no (exhaustive list) Therapeutic consumption (≤1 to 3 months depending on the molecule) Therapeutic consumption (≤1 to 3 months depending on the molecule) Table 2ResultsLaboratoryABCDE Global considerations SupportSlidesTextTextTextTextLength (pages)36311299Identification of the medical biologist (MD and/or PharmD)NoMDNoPharmDPharmDTime to results (weeks)64888 Technical considerations Personalisation and adviceYesNoYes (partly)NoNoTechnology approach and analytical versionShotgun metagenomicsShotgun metagenomicsChange of technology: the reports for V.9.5 are not comparable with the results for V.10.2Metabarcoding 16S rDNA V3-V4Metabarcoding 16S rDNA V3-V4Sequencing platformNot availableNot availableNot availableMiSeqMiSeq Quality parameters Total readsNot availableNot availableNot available342 855380 418Bacterial reads (% of the total reads)Not availableNot availableNot available69.46%69.68% Specific considerations Diversity metrics (unit)252 (not determined)3.64 (Shannon Index)6.11 (Shannon Index)3.93 (Shannon Index)3.83 (Shannon Index)Diversity interpretationDysbioticEubioticEubioticEubioticEubioticEnterotype Bacteroides Bacteroides Ruminococcus Ruminococcus Ruminococcus Bacteroides Firmicutes (%)44.3–34.5538.82–51.5552.57–33.6016.71–69.2917.62–68.60Taxonomy (three mains OTUs) Alistipes putredinis Bacteroides caccae Escherichia coli Bacteroides vulgatus Faecalibacterium prausnitzii Alistipes sp Not available Ruminococcaceae Lachnospiraceae Bacteroidaceae Ruminococcaceae Lachnospiraceae Bacteroidaceae UnitNot availableNot availableNot availablePercentagePercentage Escherichia coli Not available2.124.18 (third quartile)Not availableNot available Faecalibacterium prausnitzii 957.616.72 (first quartile)5.025.49 Akkermansia sp 830.091.13 (fourth quartile)0.370.16Data on non-bacterial microorganismsNot availableResearch for protozoa, viruses and fungi: negativeNot availableNot availableNot available Medical interpretation References and advice ’Nutritional quality: optimal’‘Heart and blood vessel capacity: optimal’Recommended diet: 9 pages ‘Positive impact on systems (particularly cardiovascular pathology)’70 references of variable quality‘Reduce consumption of polyphenols and fermented foods’ ‘The test is not intended to diagnose or screen for a disease or pathology’‘Don't hesitate to contact us for a list of the dieticians we work with’ ‘Eubiose without any abnormality, to be checked against the rest of the examination and interpreted by a doctor or health professional’‘Eubiose without any abnormality, to be checked against the rest of the examination and interpreted by a doctor or health professional’Automatic prescriptionYesNot availableYesNoNo Results Alistipes putredinis Alistipes putredinis Bacteroides caccae Bacteroides caccae Escherichia coli Escherichia coli Bacteroides vulgatus Bacteroides vulgatus Faecalibacterium prausnitzii Faecalibacterium prausnitzii Alistipes sp Alistipes sp Ruminococcaceae Ruminococcaceae Lachnospiraceae Lachnospiraceae Bacteroidaceae Bacteroidaceae ’Nutritional quality: optimal’ ’Nutritional quality: optimal’ ‘Heart and blood vessel capacity: optimal’ ‘Heart and blood vessel capacity: optimal’ Recommended diet: 9 pages Recommended diet: 9 pages ‘Positive impact on systems (particularly cardiovascular pathology)’ ‘Positive impact on systems (particularly cardiovascular pathology)’ 70 references of variable quality 70 references of variable quality ‘Reduce consumption of polyphenols and fermented foods’ ‘Reduce consumption of polyphenols and fermented foods’ ‘The test is not intended to diagnose or screen for a disease or pathology’ ‘The test is not intended to diagnose or screen for a disease or pathology’ ‘Don't hesitate to contact us for a list of the dieticians we work with’ ‘Don't hesitate to contact us for a list of the dieticians we work with’ The dual utility of microbiome testing is encouraging for the field, insofar as it demonstrates first, the importance of integrating microbiome considerations into daily clinical practice; second, the need for available information to consumers and medical practitioners.7 Calling into question the reliability and medical utility of these tests, the study highlights a lack of standardisation, transparency and clinical validation in current microbiome testing kits.8 9 The resulting incongruency poses risks of misinformation for consumers and challenges for healthcare providers interpreting such data.10 To address these concerns, enhanced regulatory oversight, method validation and the integration of microbiome science into medical education are imperative.5 The findings of this study call for coordinated efforts between regulators, industry and academia to improve test reliability and foster trust in microbiome diagnostics. Supplementary material