Fermented Foods and Gut Microbiome: Stanford RCT Data on Microbiota Diversity and Inflammation

What are fermented foods and how do they differ from probiotics?

Fermented foods are foods that have undergone controlled microbial transformation: lactic, acetic, ethanolic or mixed. They include kefir, live yoghurt, sauerkraut, kimchi, kombucha, certain cheeses and miso. Unlike probiotic capsules containing one or two strains, fermented foods carry a diverse community of live microorganisms, as well as a spectrum of organic acids, peptides and short-chain fatty acids produced during fermentation.

The key condition for preserving biological activity: the product must not undergo heat treatment after fermentation. A pasteurised fermented product contains bioactive metabolites but lacks live cultures.

Study design: Stanford RCT (Cell, 2021)

Wastyk H.C., Fragiadakis G.K., Perelman D. et al. (Cell, 2021, DOI: 10.1016/j.cell.2021.06.019) conducted a randomised controlled trial in 36 healthy adults (18 per group). Design: 3-week run-in period, 10 weeks of intervention, 4-week follow-up. Groups: high fermented food consumption (peak — 6.3 ± 2.9 servings per day) and high fibre consumption (peak — 45.1 ± 10.7 g/day, baseline — 21.5 g/day). Fermented foods consumed by participants included yoghurt, kefir, fermented cottage cheese, kombucha, kimchi, sauerkraut and other fermented vegetables.

Microbiome diversity was assessed by 16S rRNA sequencing. Immune status was evaluated by 19 plasma inflammatory proteins and immune cell activation.

What happened to microbiome diversity?

The fermented food group showed a significant increase in gut microbiome alpha-diversity across all metrics used (observed ASVs, Shannon index, phylogenetic diversity). Fermented food consumption significantly predicted changes in the Shannon index (r = 0.22). Nine amplicon sequence variants (ASVs), predominantly from the phylum Firmicutes (including four Lachnospiraceae species), specifically increased in the fermented food group.

In the high-fibre group, microbiome diversity did not change overall in the cohort. The authors interpret this as the existing microbiome lacking sufficient fermentative capacity to process the sharply increased volume of fibre: participants' stools contained more undigested carbohydrate fragments.

What were the effects on inflammatory markers?

In the fermented food group, 19 plasma inflammatory proteins significantly decreased. These included IL-6, IL-10 and IL-12b. In parallel, activation of four types of immune cells decreased. In the high-fibre group, none of the 19 inflammatory proteins showed a significant decrease over the same period.

The authors emphasise: the association is observational — the study does not address causal mechanisms beyond the measured variables. A sample size of 36 limits generalisability.

Why did a high-fibre diet not produce the same microbiome effect?

Fibre serves as a substrate for microbial enzymatic degradation, but this process requires the microbiome to contain specific taxa with the corresponding enzymes. When a diet is suddenly enriched with fibre against a background of a microbial community with limited fermentative capacity, some fibre passes through unchanged — which is exactly what was observed in the fibre group participants. Fermented foods, in contrast, introduce live microorganisms directly into the gut, bypassing the step of building up fermentative capacity.

These data do not mean fibre is useless — its long-term health effects are supported by an extensive evidence base. They mean that both tools work through different pathways and likely complement each other.

What do the 2025 meta-analyses show?

Mukherjee et al. (Frontiers in Nutrition, 2025, DOI: 10.3389/fnut.2025.1668889) conducted a meta-analysis of 25 studies (19 RCTs, 4,328 participants) on the effects of fermented foods on gastrointestinal function in healthy adults. Results: defecation frequency increased by +0.60 times/day (95% CI: 0.04–1.16, p = 0.04), stool consistency by the Bristol scale improved by +0.25 points (p = 0.03), overall GI symptoms decreased (SMD = −0.60, p = 0.009), and transit time shortened by −13.65 min (p = 0.001). The authors note variability in evidence quality.

Additionally: Caffrey et al. (Advances in Nutrition, 2025, DOI: 10.1016/j.advnut.2025.100412) — a synthesis from the Sonnenburg/Gardner laboratory systematising 12 large clinical trials — confirms that a protocol of 5–6 servings per day consistently reproduces diversity gains and reductions in inflammatory markers.