Peer-reviewed publication · Humans · In the living system
A high-fermented-food diet was followed by higher gut microbiota diversity and lower inflammatory markers; the primary outcome did not change.
- No humic or fulvic material tested
- Whole-food diets
The registered primary immune outcome did not change on either diet; in secondary analyses, the fermented-food diet was followed by higher microbiota diversity and lower inflammatory markers.
Wastyk et al., 12 July 2021. Cell 184(16):4137 to 4153.e14.
Scope, material and route
- Studied
- Generally healthy adults recruited locally at Stanford; mean age 51 ± 12 years, 73 percent women, 81 percent Caucasian, 89 percent with a college degree. People already eating over 20 g of fibre or 2 servings of fermented food a day were excluded
- Design
- Randomised, prospective, open-label trial of two diets (ClinicalTrials.gov NCT03275662); no unchanged-diet control arm; no pre-published statistical analysis plan
- What changed
- Fibre arm: at least 20 g a day added to usual intake from legumes, seeds, whole grains, nuts, vegetables and fruit. Fermented-food arm: 6 servings a day, such as yogurt, kefir, fermented cottage cheese, kimchi and other fermented vegetables, vegetable brine drinks and kombucha
- Route
- Diet (whole foods)
- Exposure
- Fibre rose from 21.5 ± 8.0 to 45.1 ± 10.7 g a day; fermented food from 0.4 ± 0.6 to 6.3 ± 2.9 servings a day, baseline to end of maintenance. Study exposure, not application guidance.
- Duration
- A 17-week protocol: 3 weeks baseline, 4 weeks ramp-up, 6 weeks maintenance, 4 weeks of participant choice. The intervention itself lasted 10 weeks
Study design
High-fibre diet
Fibre rose from 21.5 to 45.1 g a day.
18 participantsAnalysed.
- Primary: cytokine response scoreNo change
- Secondary: microbiota diversityNo change
- Secondary: carbohydrate-active enzymes11 rose in relative abundance
High-fermented-food diet
Fermented food rose from 0.4 to 6.3 servings a day.
18 participantsAnalysed.
- Primary: cytokine response scoreNo change
- Secondary: microbiota diversityRose
- Secondary: inflammatory proteins19 of 93 lower
The 17-week protocol, by week
Primary outcome read at protocol week 13, the end of maintenance and week 10 of the diet, marked above the bar.
- Baseline 3 weeks
- Ramp-up 4 weeks
- Maintenance 6 weeks
- Choice 4 weeks
More study details
- Research question
- Do diets aimed at the gut microbiota, high in fibre or in fermented foods, change the microbiome and the immune system of healthy adults?
- Sample size
- 381 assessed; 39 assigned, of whom 37 were randomised and 2 placed in the fibre arm at their own request; 1 withdrew and 2 were excluded after antibiotics; 18 analysed per arm
- Comparator
- The other diet arm and each participant's own baseline
- Material
- No humic or fulvic material was tested
- Outcomes measured
- Primary: change in cytokine response score from baseline to week 10. Secondary and exploratory: stool microbiota composition, genes, proteins and short-chain fatty acids; serum inflammatory proteins; immune cell frequencies and signalling; symptoms; cardiometabolic measures
Findings
Stanford researchers randomised healthy adults to a high-fibre or a high-fermented-food diet within a 17-week protocol: three weeks of baseline, a 10-week intervention of four weeks ramping up and six weeks maintaining, then four weeks of free choice. Eighteen people per arm were analysed. The primary outcome, a cytokine response score, did not change in either arm. In secondary and exploratory analyses, the fermented-food arm, rising from 0.4 to 6.3 servings a day, showed a steady rise in microbiota diversity and lower levels of 19 of 93 circulating inflammatory proteins, including IL-6. The fibre arm, rising from 21.5 to 45.1 g a day, showed no change in diversity, against the authors' hypothesis. There was no control arm and no blinding. No humic or fulvic material was tested.
| Outcome | Comparison or result | Interpretation |
|---|---|---|
| Cytokine response score (primary outcome) | No significant change from baseline to week 10 in either arm | The registered primary outcome was not met |
| Microbiota diversity, fermented-food arm | Rose through the intervention and stayed higher in the choice period (observed ASVs over time P = 0.0023; Shannon diversity P = 0.0014) | Composition, not a health score |
| Microbiota diversity, fibre arm | No change, contrary to the authors' hypothesis | The expected finding did not appear |
| Circulating inflammatory proteins, fermented-food arm | 19 of 93 lower, including IL-6, IL-10 and IL-12b (FDR ≤ 0.05); none of the 19 changed in the fibre arm | Lower markers, not a clinical outcome |
| Immune cell signalling, fermented-food arm | 14 of 60 cell-type-specific signalling responses lower, 1 higher | Consistent with lower resting immune activation |
| Microbial function, fibre arm | 11 carbohydrate-active enzymes increased in relative abundance, none decreased; stool carbohydrate rose with fibre intake (P = 0.0008) | Function shifted without a diversity change; the authors suggest the added fibre was not fully fermented |
| Immune responses, fibre arm | Three response clusters, one with raised immune signalling; a low-inflammation cluster had higher baseline diversity than the high-inflammation cluster (P = 0.037) | Individual responses differed; small subgroups |
| Source of new gut taxa, fermented-food arm | Few new taxa matched those in the foods, at most 5.4 percent | The diversity rise was not mainly the consumed microbes |
| Cardiometabolic measures and symptoms | Glucose, insulin, lipids, blood pressure and waist unchanged (data not shown); bloating rose in the fermented-food arm by the end of ramp-up (P = 0.03), not by the end of maintenance | No change in these health measures |
Only the cytokine response score was the registered primary outcome. All other results are secondary or exploratory, with multiple-testing correction as the authors describe for each analysis. Values are extracted from the source as reported; no raw data were available.
Interpretation and limits
What this record supports
In healthy adults, a 10-week high-fermented-food diet accompanied by rising gut microbiota diversity and lower circulating inflammatory markers, and a high-fibre diet that shifted microbial function without raising diversity.
What remains uncertain
- Design. No control arm, no blinding, 18 analysed per arm, and two fibre-arm participants were not randomised.
- Outcomes. The primary outcome was not met; the diversity and inflammation findings are secondary or exploratory.
- Meaning. Higher diversity and lower inflammatory proteins are biological measurements, not a health score; cardiometabolic measures did not change.
- Durability. The intervention lasted 10 weeks; whether the changes persist, or whether fibre would raise diversity over a longer period, is unknown.
- Relevance. No humic or fulvic material was tested.
Which part of the framework it informs
Environmental conditions
The physical, chemical and management context.
Microbial activity
What microbial communities do within that context.
This record informs this area.Observable function
What can be observed or measured in the host or wider system.
- a Colon wall
- b Plant fibre
- c Bacteria
- d Mucus layer
Source, funding and review
- Source type
- Peer-reviewed publication
- Associated products
- This record does not assess an associated product. No humic or fulvic material was tested.
- Funding and relationships
- Funded by donations to the Center for Human Microbiome Research (Paul and Kathy Klingenstein, the Hand Foundation, Heather Buhr and Jon Feiber, Meredith and John Pasquesi), with seed funding from the Institute for Immunity, Transplantation and Infection and the Sean N. Parker Center for Allergy and Asthma Research; individual authors held NIH and NSF fellowships. Two authors declare they are founders and shareholders of Interface Biosciences; the senior author declares he is a founder, shareholder and advisory board member of January AI and Novome Biotechnologies.
- Review status
- Prepared as an editorial summary. External scientific review has not been completed.
- What was read
- Accepted author manuscript read in full through PubMed Central (PMC9020749). The final published Cell version and the supplementary tables, including Table S4 with the full primary and secondary outcome results, were not read.
- Original title
- Gut-microbiota-targeted diets modulate human immune status
- Authors
- Wastyk HC, Fragiadakis GK, Perelman D, Dahan D, Merrill BD, Yu FB, Topf M, Gonzalez CG, Van Treuren W, Han S, Robinson JL, Elias JE, Sonnenburg ED, Gardner CD, Sonnenburg JL
- Publication
- Cell 184(16):4137 to 4153.e14, 12 July 2021
- DOI
- 10.1016/j.cell.2021.06.019
- Summary prepared by
- Carbon Biome Project
- Date added
- 19 October 2026
- Cite this summary
- Carbon Biome Project. A high-fermented-food diet was followed by higher gut microbiota diversity and lower inflammatory markers; the primary outcome did not change. Evidence library record, added 19 October 2026. carbonbiomeproject.com/evidence/wastyk-2021-fibre-and-fermented-food-diets-immune-status/