Fermented Foods

TL;DR. Fermentation is controlled spoilage: you deliberately encourage microbes that make the food inhospitable to the ones that would ruin it. It predates refrigeration by thousands of years and every cuisine on Earth has it. The health evidence improved substantially with a 2021 Stanford randomised trial that found a high-fermented-food diet increased microbiome diversity and reduced inflammatory markers, where a high-fibre diet did not. Note that many fermented foods contain no live microbes by the time you eat them, because they were baked, pasteurised, or filtered.

1. What fermentation is

Microbes consuming sugars and producing acid, alcohol, or gas, plus a large number of flavour compounds. The result is preserved because the pH drops, alcohol accumulates, or the niche is occupied by organisms you want.

TypeOrganismProductExamples
Lactic acidLactobacillus, Leuconostoc and relativesLactic acidSauerkraut, kimchi, yoghurt, sourdough, pickles, salami
AlcoholicSaccharomyces yeastEthanol, CO₂Beer, wine, bread
AceticAcetobacterAcetic acidVinegar, kombucha
AlkalineBacillusAmmoniaNatto, some African seed ferments
MouldAspergillus, Rhizopus, PenicilliumEnzymes, flavourMiso, soy sauce, tempeh, blue cheese, koji

2. Why it was invented

Preservation, before refrigeration. Lactic fermentation drops the pH below about 4.5, which stops most pathogens including Clostridium botulinum. That is the whole reason sauerkraut, kimchi, salami, and cheese exist.

Along the way it does four other useful things:

  • Predigestion. Microbes break down complex molecules: lactose in yoghurt, phytate in sourdough, oligosaccharides in tempeh, proteins into savoury peptides in miso and soy sauce.
  • Nutrient synthesis. Some ferments generate B vitamins and vitamin K2. Natto is the richest known dietary source of menaquinone-7, a long-acting form of vitamin K2.
  • Toxin reduction. Cassava fermentation removes cyanogenic glycosides (Chapter 42); sourdough degrades phytate and fructans; some ferments reduce mycotoxins.
  • Flavour. Umami, sourness, and complexity that cannot be produced any other way. Fermentation is the origin of most of the deeply savoury ingredients in world cooking: soy sauce, fish sauce, miso, Worcestershire sauce, cheese, vinegar, and cured meat.

3. The live and the not-live

This distinction decides whether a food can affect your microbiome directly, and labels rarely make it clear.

Contains live microbesDoes not, by the time you eat it
Live yoghurt, kefirBread and sourdough (baked)
Unpasteurised sauerkraut and kimchi (chilled section)Most shelf-stable sauerkraut and pickles (pasteurised)
Miso (if not boiled)Beer and wine (filtered), most soy sauce
KombuchaVinegar-pickled vegetables (not fermented at all)
Aged cheeses, especially raw milkTempeh and tofu once cooked
Kvass, some traditional picklesChocolate, coffee, vanilla (all fermented, none live)

Vinegar pickles are not fermented. Cucumbers in vinegar are preserved by acid you added, not by microbes. Traditional brine pickles, fermented in salt water, are.

The shelf position is the practical tell: live ferments are almost always refrigerated, because heat treatment is what makes a product shelf-stable.

4. What the evidence shows

Established: fermentation predigests lactose (yoghurt is tolerated by most lactose non-persistent people), degrades phytate in sourdough, reduces FODMAPs in long-fermented bread, and preserves food safely.

Strong, and this is the notable recent development: the 2021 Stanford trial by Wastyk, Sonnenburg and colleagues, published in Cell, randomised 36 adults to a high-fermented-food diet (around six servings a day of yoghurt, kefir, kimchi, kombucha, and vegetable brine drinks) or a high-fibre diet, for ten weeks. The fermented food group showed:

  • Increased gut microbiome diversity, which is unusual: diversity is normally very stable in adults.
  • Decreased levels of 19 inflammatory proteins, including interleukin-6.

The high-fibre group did not show either effect over that period, though fibre remains well-evidenced for other outcomes (Chapter 14). The trial was small and its design was strong, and it is currently the best single piece of evidence for fermented food as a category.

Also reasonably supported:

  • Yoghurt intake is consistently associated with lower type 2 diabetes risk in cohort studies.
  • Fermented dairy is not associated with the cardiovascular risk its saturated fat content would predict (Chapter 54).
  • Natto and vitamin K2 for bone density, with Japanese observational data and some trials.
  • Kefir has more and more diverse organisms than yoghurt (typically 30 to 60 strains including yeasts, against 2 to 7 in yoghurt) and better evidence for lactose digestion.

Thin: kombucha for essentially all of its marketed claims. Sourdough as a cure for gluten sensitivity, though the fructan reduction is real and probably explains why some people tolerate it (Chapter 51). Any specific fermented food as a treatment for a named disease.

5. Making it at home, and doing it safely

Vegetable fermentation is genuinely easy and genuinely safe if you follow the rules, because the acid environment excludes pathogens.

Sauerkraut and kimchi:

  1. Salt at 2 to 2.5 percent of the vegetable weight. This is the critical number: enough salt selects for lactic acid bacteria and suppresses spoilage organisms. Weigh it; do not guess.
  2. Keep everything below the brine. Exposure to air is what causes mould and kahm yeast. Use a weight.
  3. Ferment at room temperature (18 to 22 °C) for 3 to 14 days, then refrigerate.
  4. Expect bubbling and cloudiness. That is the fermentation working.
  5. Discard if it smells putrid, is slimy, or has fuzzy coloured mould. A white film on the surface (kahm yeast) is harmless and can be skimmed; fuzzy blue, black, or green growth means discard.

Why it is safe: the pH drops below 4.5 within days, and no known foodborne pathogen, including C. botulinum, grows below that. Documented illness from properly made vegetable ferments is very rare.

Where the real risks are:

  • Low-acid, oxygen-free preparations: home-canned vegetables, garlic in oil, and improperly fermented fish and meat. Botulism cases in the Arctic and in Alaska come from traditionally fermented marine mammal and fish preparations made in modern airtight plastic containers rather than traditional permeable ones.
  • Home-cured meat and salami, where nitrite curing salts and controlled humidity are doing real safety work and improvisation is dangerous.
  • Kombucha made in ceramic or lead-glazed vessels (acid leaches lead) or fermented too long (excessive acidity). Alcohol content can also exceed 0.5 percent, which matters for some people.
  • Raw milk products, which carry genuine listeria, E. coli, and campylobacter risk and are specifically not recommended in pregnancy.

6. Who should be careful

  • Histamine intolerance. Fermented foods are among the highest-histamine foods there are: aged cheese, cured meat, sauerkraut, soy sauce, wine, and vinegar. In people with reduced diamine oxidase activity this causes headaches, flushing, itching, and gut symptoms. It is a real, under-diagnosed, dose-dependent phenomenon.
  • MAOI antidepressants and tyramine. This is a serious drug interaction: aged cheese, cured meat, soy sauce, sauerkraut, and draught beer contain tyramine, which normally gets broken down by monoamine oxidase. On an MAOI, tyramine accumulates and can trigger a hypertensive crisis. This is the "cheese effect," it is potentially fatal, and it is the reason MAOIs come with strict dietary rules (Chapter 77).
  • Immunosuppression. Live-culture and raw-milk foods carry infection risk; discuss with your team.
  • Pregnancy. Avoid unpasteurised soft cheeses (listeria); pasteurised live yoghurt and cooked ferments are fine.
  • Sodium. Kimchi, sauerkraut, miso, and soy sauce are all salty by design. A serving of miso soup can carry 1 g of salt.
  • IBS. Some ferments are better tolerated (yoghurt, hard cheese, long-fermented sourdough, tempeh) and some worse (kombucha, kefir in quantity, garlicky kimchi).

7. Practical guidance

  • A serving or two a day is roughly what the Stanford trial's benefit built on over ten weeks, and it started from a much higher dose (six servings), tapering after an introduction period.
  • Start small and build. Going from none to six servings a day produces bloating.
  • Buy from the fridge if you want live cultures. Check for "live cultures" or "unpasteurised" on the label.
  • Rotate the sources. Different ferments carry different organisms, and diversity is the point.
  • Do not boil miso. Add it off the heat to keep the organisms and the aroma.
  • Cheap and effective options: plain live yoghurt, kefir, unpasteurised sauerkraut, kimchi, miso, tempeh. Sauerkraut made at home costs almost nothing.
  • Sourdough for the phytate and fructan reduction, even though it contains no live culture after baking.

8. The bottom line

  • Fermentation is controlled spoilage, invented for preservation, and it also predigests, adds nutrients, removes toxins, and creates most of the deeply savoury flavours in world cooking.
  • Many fermented foods contain no live microbes by the time you eat them. The refrigerated shelf is where the live ones are.
  • The best current evidence is the 2021 Stanford randomised trial: a high-fermented-food diet increased microbiome diversity and lowered inflammatory markers over ten weeks, where a high-fibre diet did not.
  • Home vegetable fermentation is safe if you use 2 to 2.5 percent salt and keep everything under the brine. Home-cured meat, garlic in oil, and low-acid canning are where the real danger is.
  • Two groups must be careful: people with histamine intolerance, and anyone taking an MAOI, for whom tyramine in aged and fermented foods can cause a hypertensive crisis.

Sources and notes

Fermentation microbiology and preservation mechanisms follow standard food microbiology texts. The randomised comparison of high-fermented-food versus high-fibre diets, finding increased microbiome diversity and reduced inflammatory markers, is Wastyk et al., Cell, 2021, from Stanford. Yoghurt and type 2 diabetes follows Chen et al., BMC Medicine, 2014. Kefir microbial diversity and lactose digestion follow Hertzler and Clancy's work. Natto and vitamin K2 (MK-7) follow Japanese observational and intervention studies on bone density. Sourdough phytate degradation follows Lopez et al. Cassava fermentation and cyanide reduction follows Bradbury's work. Home fermentation safety, the 2 to 2.5 percent salt rule, and pH below 4.5 excluding pathogens follow the National Center for Home Food Preservation and Katz, The Art of Fermentation. Arctic botulism from modern airtight containers follows CDC Alaska reporting. Histamine intolerance and diamine oxidase follow Maintz and Novak, American Journal of Clinical Nutrition, 2007. The MAOI tyramine hypertensive crisis follows standard psychopharmacology references and BNF guidance.

Open questions. The Stanford fermented food trial is small and short, and whether the microbiome and inflammatory changes persist or translate into outcomes is untested. Which organisms in fermented foods matter, if any, is unknown.

👉 Next: salt, sugar, and sweeteners.