Fermented Foods and Gut Diversity: What a 10-Week Trial Found

By Hugo Novais, biotechnologist, geneticist and former professional swimmer.

As an Amazon Associate I earn from qualifying purchases. It doesn't change what I recommend.

Fermentation is the part of biotechnology I've worked with longest. It's the oldest biotech there is: microbes doing chemistry on food long before anyone had a lab coat. So when Stanford researchers ran a head-to-head trial asking whether fermented foods could change gut microbiome diversity in real people, I read the paper properly, not the headlines. Here's what it found, what it didn't, and how to use it on an ordinary week.

Why gut diversity gets so much attention

Your large intestine holds hundreds of bacterial species. "Diversity" describes how many different kinds live there and how evenly they're spread. Lower diversity has been linked in observational studies to industrialised diets and to a range of inflammatory conditions. That's a link, not proof of cause, but it's why researchers keep asking a simple question: can food push diversity up?

The obvious candidate has always been fibre. Fermented foods were the wildcard.

The Stanford trial: fermented foods vs high fibre

In a randomised study published in Cell in 2021, 36 healthy adults were split into two groups. One group gradually increased fibre (legumes, wholegrains, seeds, vegetables, fruit). The other gradually increased fermented foods such as yoghurt, kefir, fermented cottage cheese, kombucha, kimchi, sauerkraut and vegetable brine drinks, building up to roughly six serves a day by the end of the 10-week diet phase (Wastyk et al., 2021).

The results surprised a lot of people, including the researchers:

• Fermented-food group: microbiome diversity rose steadily across the intervention, and 19 inflammatory proteins measured in the blood went down, including interleukin-6 (Stanford Medicine).
• High-fibre group: diversity stayed broadly stable. Their gut bacteria did gain more fibre-digesting enzymes, and immune responses split into different patterns depending on how diverse their microbiome was at the start.

The part I find most interesting as a biotechnologist: the researchers reported that most of the extra diversity didn't seem to come straight from the microbes in the foods. Fermented foods appear to change the gut environment in a way that lets other species establish, rather than simply "seeding" the gut with what's in the jar.

The myth to drop: "fermented = probiotic"

An international expert panel defines fermented foods as "foods made through desired microbial growth and enzymatic conversions of food components" (Marco et al., 2021). That's broader than probiotics. Sourdough bread, most beer and wine, and shelf-stable sauerkraut are all fermented, but baking, filtering or pasteurising means few or no live microbes are left.

So if live microbes are what you're after, look for products kept in the fridge and labelled "live cultures" or "unpasteurised". I covered why the number on a probiotic label means less than you'd think in The Probiotic CFU Myth.

What the wider research adds

• Large-scale data: across 6,811 people in the American Gut Project, regular fermented-food eaters had subtle but measurable differences in their gut microbiome and metabolites compared with non-consumers (Taylor et al., 2020). Observational, so it can't show cause.
• Digestive comfort: a 2025 meta-analysis of 19 randomised trials in healthy adults found fermented foods modestly improved bowel-movement frequency, stool consistency and some digestive symptoms, but the certainty of evidence was mostly low (Mukherjee et al., 2025).
• Kefir specifically: a 2026 systematic review of eight human studies found modest, mixed changes, with Bifidobacterium increasing in three of four studies that measured it (Hamsho et al., 2026).

If you want to see how this ties into immunity more broadly, read The Gut-Immune Axis.

The honest limitations

• Small trial. 18 people per group, all healthy adults in California. It needs repeating in larger, more varied groups.
• Diversity isn't a health outcome. It's a marker. Nobody has shown that raising it this way lowers disease risk.
• Six serves a day is a lot. Most people won't sustain that, and we don't know the minimum dose that still helps.
• Not for everyone. Kimchi and sauerkraut can be high in salt. Some people with IBS or histamine intolerance feel worse on large amounts. If you're pregnant or immunocompromised, avoid unpasteurised products unless your doctor says otherwise.

General information, not medical advice. If you have a gut condition, talk to your GP before making big changes.

How to use this: the one-serve ramp

You don't need six serves. Start with one and build slowly so your gut can adjust:

1. Week 1: one serve a day, such as a small tub of plain yoghurt or a glass of kefir with breakfast.
2. Week 2: add a second serve at another meal, like two forkfuls of sauerkraut or kimchi with lunch.
3. Week 3 onwards: rotate types (dairy, vegetable, drinks) rather than doubling one. Variety is the point.
4. Keep your fibre up too. The trial compared the two, but in real life they work well together.

Kefir is the easiest high-protein option for shift workers: it's drinkable, needs no prep and survives a cooler bag. If you'd like to make your own, a milk kefir starter is the simplest place to begin. Kombucha counts too, though watch the sugar and acidity. I went through the human evidence in Is Kombucha Healthy?

Recipe: 10-minute fridge-door sauerkraut

This is the first ferment I recommend to anyone. Two ingredients, and the bacteria do the rest.

• 1 kg green cabbage (about half a large head)
• 20 g non-iodised salt (2% of the cabbage weight; weigh it)
• Optional: 1 tsp caraway seeds or a grated carrot

1. Remove the outer leaves, then finely shred the cabbage.
2. Sprinkle with salt and massage with clean hands for 5–8 minutes until it's limp and sitting in its own liquid.
3. Pack tightly into a clean 1-litre jar, pressing down until the brine covers the cabbage. A glass fermentation weight keeps it submerged, which is the single most important step.
4. Cover with a lid that lets gas escape (an airlock fermenting lid makes this easy), or loosen a normal lid once a day.
5. Leave at room temperature, out of direct sun, for 1–3 weeks. Taste from day 7. When it's sour enough for you, move it to the fridge.

Safety: anything above the brine can grow mould. If you see fuzzy mould, or it smells rotten rather than sour, throw it out. Homemade ferments vary, so they're a food, not a measured dose.

The takeaway

The best trial we have suggests fermented foods can raise gut microbiome diversity and lower some inflammation markers in healthy adults, faster than fibre alone did in that study. It's early, small and not a cure for anything. But one serve a day is cheap, tasty and easy to try.

Not sure which foods suit your body and your roster? Start with my free Biotype Guide. It takes a few minutes and gives you a clear first step. And if you'd like a plan built around your shifts, your first consultation is free.

Sources

1. Wastyk HC, et al. Gut-microbiota-targeted diets modulate human immune status. Cell. 2021;184(16):4137–4153. doi:10.1016/j.cell.2021.06.019 (PubMed 34256014)
2. Stanford Medicine News. Fermented-food diet increases microbiome diversity, decreases inflammatory proteins. July 2021. med.stanford.edu
3. Marco ML, et al. The ISAPP consensus statement on fermented foods. Nat Rev Gastroenterol Hepatol. 2021;18(3):196–208. doi:10.1038/s41575-020-00390-5 (PubMed 33398112)
4. Taylor BC, et al. Consumption of fermented foods is associated with systematic differences in the gut microbiome and metabolome. mSystems. 2020;5(2):e00901-19. doi:10.1128/mSystems.00901-19 (PubMed 32184365)
5. Mukherjee A, et al. Impact of fermented foods consumption on gastrointestinal wellbeing in healthy adults: a systematic review and meta-analysis. Front Nutr. 2025;12:1668889. doi:10.3389/fnut.2025.1668889 (PubMed 41141260)
6. Hamsho M, et al. Kefir: a potential gut microbiota modulator. A systematic review of human interventional studies. MicrobiologyOpen. 2026;15(3):e70297. doi:10.1002/mbo3.70297 (PubMed 42036973)