Fermented foods are marketed for almost everything. Whether one is “worth it” comes down to three questions: is it actually fermented, does it still contain live microbes when you eat it, and has anyone tested the claim in people? A useful starting point is the expert definition of a fermented food – one “made through desired microbial growth and enzymatic conversions of food components” – which is not the same thing as a probiotic (Marco et al., 2021). A probiotic is a defined, live microorganism shown to benefit health in an adequate amount (Hill et al., 2014). Most fermented foods are not tested probiotics, and some – baked sourdough, most beer and wine, pasteurized sauerkraut – contain no live microbes at all, though many others, such as yogurt, kefir and raw kimchi, can carry large numbers (Rezac et al., 2018).
Good Evidence
- Yogurt and kefir if lactose is a problem. The European Food Safety Authority accepted that live yogurt cultures improve the digestion of the lactose in yogurt in people who digest lactose poorly (EFSA, 2010). In a trial of adults with lactose maldigestion, plain kefir and plain yogurt both produced about a third of the breath hydrogen that milk did and reduced flatulence (Hertzler & Clancy, 2003).
- Eating more – and more varied – fermented foods. In a Stanford trial, healthy adults who built up to about six servings a day of foods such as yogurt, kefir, kimchi, sauerkraut and kombucha over 10 weeks ended with more diverse gut microbiota and lower markers of inflammation. It was small, with 18 people per group (Wastyk et al., 2021), but it’s the best direct test so far. See how much fermented food per day for sensible amounts.
- Fermented milks and cheese, long term. In a 2026 meta-analysis of 50 cohort studies with more than three million people, higher intake of fermented milks (including yogurt) and cheese was associated with lower all-cause and cardiovascular mortality (Matalas et al., 2026). These are observational associations, not proof of cause, but they’re consistent.
Promising, but the Evidence Is Thin
- Kimchi. In a crossover trial of 22 overweight adults, four weeks of fermented kimchi improved blood pressure, body fat, fasting glucose and cholesterol more than fresh kimchi did (Kim et al., 2011). Encouraging, but small and short. Raw, refrigerated kimchi keeps its live bacteria; see kimchi.
- Apple cider vinegar. A 2025 meta-analysis of 10 randomised trials (789 people) found modest reductions in body weight and BMI with daily vinegar, mostly in people who were overweight or had type 2 diabetes (Castagna et al., 2025); another found lower fasting glucose and HbA1c in seven trials in type 2 diabetes (Arjmandfard et al., 2025). The trials are small and varied. Vinegar is a fermented product but not a probiotic: its benefits, if real, come from acetic acid, not live bacteria. Drink it diluted, and talk to your doctor if you take diabetes medication.
- Natto and other fermented soy. In a Japanese cohort of about 80,000 adults, women who ate the most fermented soy, natto included, had a lower risk of cardiovascular disease; there was no association in men, or with cancer (Nozue et al., 2021). Natto is very rich in vitamin K, which can interfere with warfarin (Homma et al., 2006); see natto.
Hype Ahead of the Evidence
- Kombucha. A 2019 systematic review found just one human study of kombucha’s health effects (Kapp & Sumner, 2019). Since then, a pilot trial in 12 adults with type 2 diabetes reported lower fasting glucose after four weeks; the kombucha used contained 1.5% alcohol (Mendelson et al., 2023). In the US, kombucha that reaches 0.5% alcohol is regulated as an alcoholic drink (TTB). Enjoy it as a drink you like, not as medicine, and check the label for sugar; see how to make kombucha.
- Miso. Long-aged, unpasteurized miso is a superb seasoning, but in the same 2026 cohort meta-analysis miso showed no association with mortality (Matalas et al., 2026), and it’s salty. Use it for flavour.
- “Probiotic” claims on foods without live microbes. Shelf-stable, heat-treated or vinegar-pickled versions of fermented foods may taste similar but have no live cultures (Marco et al., 2021). A product making probiotic claims should name its strains and the number of live organisms.
How to Choose
- For live cultures, buy refrigerated, unpasteurized products and look for “raw”, “unpasteurized” or “live cultures” on the label – our guide to store-bought fermented foods goes through brands.
- Prefer plain versions; sweetened fermented drinks and yogurts can carry a lot of added sugar.
- Watch the salt in kimchi, sauerkraut, miso and pickles, especially if you have high blood pressure.
- If you’re immunocompromised, pregnant, or take warfarin, ask your doctor before making big changes.
