Gut health has moved from a laboratory term into a product category. Kefir and kimchi share shelf space with capsules printed with strain codes, and the promise on the packaging is usually the same one: a better gut, and through it, better everything else.

September brought two systematic reviews that speak almost directly to that promise. Both looked at supplements designed to modify the gut microbiome in adults over 60 — one at cognition and muscle outcomes, the other at cognition alone. Neither is a single study with a tidy result; both are syntheses of randomized, placebo-controlled trials. Read together with the federal evidence summaries, they describe a field that is genuinely promising and genuinely unfinished.

What the two September reviews measured

The first, published in Frontiers in Nutrition, screened randomized placebo-controlled trials of probiotics, synbiotics or fecal microbiota transplantation in populations with a mean age of at least 60. Twenty-four trials covering 1,809 participants met the criteria, and 11 contributed to at least one pooled analysis.

On the Mini-Mental State Examination the pooled estimate was positive — a Hedges’ g of 1.02, with a 95% confidence interval of 0.36 to 1.68. That number comes with a caveat attached to it: statistical heterogeneity was extreme, at roughly 91%, meaning the trials disagreed with one another to a degree that makes an average hard to trust. A second cognitive battery, the Repeatable Battery for the Assessment of Neuropsychological Status, showed no statistically significant difference (g = 0.46; 95% CI −0.50 to 1.41). Handgrip strength improved modestly (g = 0.46; 0.06 to 0.85), but removing one trial from the pool flattened that result to non-significance. The authors’ own conclusion is measured: preliminary short-term signals, a small evidence base, short follow-up, and no firm conclusions about efficacy.

The second review, in Age and Ageing, included 14 studies of older adults — 13 randomized trials and one other interventional study, split across 10 probiotic, 2 prebiotic and 2 synbiotic interventions. Nine reported statistically significant improvement in at least one cognitive domain, most often global cognition, memory, attention and executive function. A subset meta-analysis put the improvement in global cognition scores at a mean difference of 1.28 points (95% CI 0.15 to 2.40). Here, too, the authors pair their positive reading with the limits of the literature: heterogeneity, small samples, and what they call a paucity of high-quality studies.

The stronger evidence is narrower

If the September papers are the frontier, the settled ground is somewhere else entirely. NCCIH, the NIH center that evaluates complementary and integrative health claims, maintains a condition-by-condition review of probiotic research, and what it documents is a set of specific findings tied to specific products — not a general wellness effect.

A short survey of that record:

  • Antibiotic-associated diarrhea. A 2017 review of 17 studies with 3,631 participants found that taking probiotics alongside antibiotics was associated with roughly half the likelihood of diarrhea. NCCIH calls the conclusion tentative, because study quality was only moderate, and notes that a benefit has not been demonstrated in elderly people.
  • C. difficile infection. A 2017 analysis of 31 studies and 8,672 patients concluded it is moderately certain that probiotics reduce the risk of C. difficile diarrhea in adults and children receiving antibiotics, most of them hospital patients.
  • Preterm infants. A 2017 review of 23 studies (7,325 infants) found probiotics helped prevent necrotizing enterocolitis, and a separate review of 37 studies (9,416 participants) found a reduced risk of sepsis.
  • Irritable bowel syndrome. A 2018 review of 53 studies and 5,545 participants found probiotics may help global symptoms and abdominal pain, but could not identify which species, strains or combinations are most likely to help.
  • Ulcerative colitis and Crohn’s disease. A 2014 review of 21 studies (1,700 participants) found that adding probiotics, prebiotics or synbiotics to conventional treatment could help induce or maintain remission in ulcerative colitis; in Crohn’s disease, 14 studies and 746 participants showed no such evidence.
  • Traveler’s diarrhea and urinary tract infections. A 2017 clinical practice guideline from the International Society of Travel Medicine found insufficient evidence to recommend probiotics for travelers, and a 2015 review of nine studies (735 participants) found no evidence of benefit for preventing urinary tract infections.

The pattern is consistent across dozens of trials: each result is attached to a named strain and a named clinical situation. NCCIH’s summary of its own review is blunt — for most instances, it says, we still don’t know which probiotics are helpful and which are not, how much a person would have to take, or who is most likely to benefit.

There is a reason the strain keeps appearing in the research and rarely on the front of the package. As NCCIH explains, different probiotics have different effects: if one kind of Lactobacillus helps prevent an illness, that says nothing about another Lactobacillus, or about any Bifidobacterium. A blanket “supports gut health” claim is not organised around the variable the science is organised around.

How the products are regulated

That gap is partly a labelling question. NCCIH notes that many probiotics are sold as dietary supplements, which do not require FDA approval before they are marketed. Supplement labels may carry claims about how a product affects the structure or function of the body without FDA approval, but they may not make health claims — statements about lowering the risk of a disease — without the agency’s consent. A probiotic marketed as a drug for treating a disease has to clear a stricter bar: proven safe and effective in clinical trials, and approved by the FDA, before it can be sold.

What the food evidence points to

Diet, meanwhile, is the lever with the longest track record. Harvard’s T.H. Chan School of Public Health notes that diet plays a large role in determining which microbes live in the colon. Dietary fibre can only be broken down by enzymes from those colonic microbes; the fermentation produces short-chain fatty acids, which lower the pH of the colon and, in turn, limit the growth of some harmful bacteria — C. difficile among them.

Harvard lists the foods that feed this system: raw garlic, onions, leeks, asparagus, Jerusalem artichokes, dandelion greens, bananas and seaweed are the richest sources of prebiotic fibre, with fruits, vegetables, beans and whole grains close behind. The school adds a practical warning that most supplement marketing skips — a sudden jump in prebiotic intake can cause gas and bloating, so people with a sensitive gut should start small and build up. A fibre-rich diet built on vegetables is the everyday version of the same idea.

Fermented foods represent the other route. Harvard points to kefir, yogurt with live active cultures, kimchi, sauerkraut, miso, tempeh and kombucha — while noting that not every fermented food still carries live microorganisms, since pasteurising or heating after fermentation destroys them. Examples include sourdough bread and heat-treated fermented vegetables. Harvard frames the open question plainly: whether a relatively healthy person gains anything from probiotic supplementation, or whether a fibre-rich diet is enough on its own.

Safety notes the packaging leaves out

Probiotics have a long history of apparently safe use, particularly in healthy people, but NCCIH is careful about how much that is worth: few studies have examined safety in detail, so the frequency and severity of side effects are not well documented. The risk of harm is greater in people who are severely ill or have compromised immune systems. The FDA warned health care providers in 2023 that premature infants given probiotics are at risk of severe, potentially fatal infections, and possible harms listed by NCCIH include infections, the production of harmful substances by the microbes, and the transfer of antibiotic-resistance genes. Some products have been found to contain organisms other than the ones on the label.

Where the pitch outruns the evidence

Three moves are common on the shelf and none of them follow from the sources above. The first is turning a strain-specific, condition-specific finding into a general promise about gut health or immunity — the same leap that ran through our look at the cold-and-flu supplement aisle. The second is selling a category rather than a strain: the research is organised around named organisms and doses, and most labels are not. The third is treating a capsule as a substitute for eating pattern, when the dietary evidence is the older and steadier part of the story.

What the sources support is undramatic. A varied, fibre-rich diet and fermented foods are the practical, best-evidenced route to a well-fed microbiome. For a diagnosed condition, probiotics are a conversation with a clinician — not a self-prescribed product, and not a reason to put off care. Nothing here is a diagnosis or a treatment plan; that position is set out in our medical disclaimer.