Probiotics: uses, evidence, dosing and safety

Probiotics are sold for almost everything, and the honest version of this page is that they have one solid result and a long tail of maybes.

The solid one is worth knowing about. So is the detail Cochrane recorded about who paid for the trials.

What probiotics are

NCCIH, the US body for complementary health research, defines them as live microorganisms that are intended to have health benefits when consumed or applied to the body.

They are not exotic. They can be found in yogurt and other fermented foods, dietary supplements, and beauty products, and the organisms involved are largely ones already living in us: many of the microorganisms in probiotic products are the same as or similar to microorganisms that naturally live in our bodies.

Two families dominate the shelf. The most common are bacteria that belong to groups called Lactobacillus and Bifidobacterium, alongside yeasts such as Saccharomyces boulardii.

How probiotics work

The gut holds a large population of bacteria that help digest food and occupy space that less welcome organisms might use. Antibiotics disturb that population, which is where the strongest evidence on this page comes from.

The single most important thing to understand is that probiotic is a category, not an ingredient.

NCCIH states the consequence directly: if a specific kind of Lactobacillus helps prevent an illness, that doesn’t necessarily mean that another kind of Lactobacillus or any of the Bifidobacterium probiotics would do the same thing.

That is why most sections below carry low certainty. The reviews pool many different organisms at many different doses, and a positive average across a mixed bag does not tell you what to buy.

What the evidence says about probiotics and antibiotic-related diarrhea

This is the best-supported use, and the numbers are worth stating precisely.

Antibiotics can clear out enough of the normal gut population to let a bacterium called Clostridioides difficile take hold, causing a diarrhea that ranges from unpleasant to dangerous.

A 2025 Cochrane review pooled 38 trials (13,179 participants). It found that probiotics may result in a small absolute risk reduction (ARR) of 1.6%: an incidence of 1.6% in the probiotic group versus 3.2% in the control group. That is half the rate, and the exact ratio is in the summary box above.

Halving the rate sounds dramatic and the absolute figure keeps it in proportion: roughly 16 fewer cases per thousand people treated.

Safety in this setting was reassuring. Probiotics were not associated with adverse effects when used in patient populations receiving systemic antibiotics who are not immunocompromised.

The part of the review nobody quotes

Cochrane graded that result low-certainty evidence, and its reasons are unusually specific.

Around 30% of studies lacked a published protocol or clinical trial registration, meaning nobody can check what those trials set out to measure before they measured it.

And then the line worth reading twice: the reviewers noted authors’ affiliation with a probiotic company and financial support received from probiotic companies in 28 studies, out of 47 included in total.

That does not make the result wrong. It is a reason the certainty is low rather than high, and it is the kind of detail that rarely survives into a product page.

What the evidence says about probiotics and irritable bowel syndrome

The use most people actually buy them for, and the evidence is genuinely unsettled.

A meta-analysis comparing probiotics against drug treatments for irritable bowel syndrome, or IBS, concluded that comparing the overall efficacy of both intervention types is inherently complex and such results should be interpreted with care, because of the large diversity of probiotic- and drug types and doses and the different subtypes of the condition.

A separate review of probiotic fungi found the partial response and the relative safe of probiotic fungi for IBS treatment have been demonstrated, with mild complications in up to half of participants in some trials and no serious ones. Its own verdict: it is premature to eventually declare the practical effects.

So: probably something, for some people, with some products. Both sets of reviewers ask for the same thing, which is trials of a specific strain in a specific subtype rather than more pooling.

What the evidence says about probiotics and depression

A modest signal, better than most on this page.

A 2026 meta-analysis included 13 studies involving 710 patients with MDD, where MDD is major depressive disorder. The pooled effect size was 0.38, favouring probiotics use, which is small to moderate.

The reviewers looked for what explained the variation between trials and could not find it: subgroup analyses and meta-regression identified no significant moderators.

Their conclusion is appropriately hedged. Probiotics appear to reduce depressive symptoms, but factors contributing to heterogeneity remain uncertain.

For a condition where the placebo response is large and the trials are small, that is a signal worth following rather than a treatment.

What the evidence says about probiotics and cognition

Better-rated than most sections here, and thinner than the rating suggests.

An umbrella review found that the beneficial associations between probiotics and cognitive health outcomes are supported by moderate- to high-quality evidence.

The qualification matters more than the grade. Much of what was pooled were laboratory measures rather than how people think or function. Of the outcomes examined, the reviewers report that 10 (21.3%) focused on oxidative stress markers and a further share on inflammatory markers, rather than on cognition itself.

They call for exactly what is missing: additional clinical trials are needed to investigate the effects of different probiotic strains, doses, and intervention durations on clinical outcomes.

What the evidence says about probiotics in cancer care

One of the stronger positive findings, and like omega-3, it is in hospital rather than on a shelf.

An umbrella meta-analysis of cancer treatment complications found probiotics reduced diarrhea and oral mucositis, the painful mouth ulceration caused by chemotherapy and radiotherapy, with the clearest effect on the severe form.

It also found a difference between products that supports the strain argument. Multi strain probiotic preparations were more efficacious than single strain ones.

The reviewers describe their own findings as strong evidence that probiotic supplementation can reduce cancer complications, which is unusually confident language for this field.

Anyone in cancer treatment should still raise it with their oncology team first, because live organisms and a suppressed immune system are a combination that needs supervising.

What the evidence says about probiotics and eczema in babies

An instructive negative, because it overturns what earlier reviews reported.

A 2026 review found that current evidence does not support routine probiotic supplementation initiated only during infancy for the prevention of atopic dermatitis, atopic dermatitis being infant eczema.

The interesting part is why earlier reviews disagreed. Positive preventive effects reported in previous meta-analyses are largely driven by perinatal (maternal-plus-infant) regimens, whereas infant-only supplementation has failed to demonstrate consistent benefit.

In other words, the timing was doing the work. Courses that began with the mother before birth showed the benefit; courses that started with the baby did not, and pooling them together made it look as though the baby’s supplement was working.

What the evidence says about probiotics and acne

A small, specific result with an honest caveat about the condition itself.

A 2026 review found probiotics did not demonstrate a significant reduction in acne lesion counts. What did improve was the severity rating doctors assign.

The effect concentrated in a particular group: it was most prominent in mild-to-moderate acne patients who received oral probiotics containing Lactobacillus species combined with other strains for a minimum of 12 weeks.

The reviewers add a caution specific to skin. These findings should be interpreted with caution given the heterogeneity across studies and the fluctuating natural course of acne, which waxes and wanes on its own and makes any twelve-week trial hard to read.

What the evidence says about probiotics and fertility treatment

Modest associations at weak certainty, in an area where hope runs ahead of evidence.

A 2026 review in women with infertility found oral probiotics associated with higher clinical pregnancy rates and higher live-birth rates, and vaginal probiotics associated with a reduced risk of miscarriage only.

Live birth is the outcome that matters, and it moved, which is more than most supplements manage.

The reviewers’ own summary keeps it in proportion: there was a modest association, and the quality of evidence was low to very low. Anyone in fertility treatment should be raising this with their clinic rather than acting on it alone.

What the evidence says about probiotics and blood sugar

Markers improve, certainty is low, and the reason is the one running through this whole page.

A 2024 meta-analysis in type 2 diabetes found supplementation with probiotics seems to improve indices of glycemic control, meaning blood sugar management, including fasting blood sugar, insulin and a measure of insulin resistance. Longer interventions produced larger reductions in fasting blood sugar.

Then the caveat, which names the problem exactly: caution must be considered because of the notable heterogeneity (with regard to dosage, duration, and the species/strains used) and the low quality of evidence.

What the evidence says about probiotics and inflammation

The mechanism section, and one of the few places a pattern emerged rather than an average.

A 2025 review in prediabetes and type 2 diabetes found probiotic and synbiotic supplementation significantly reduces inflammatory cytokines including the standard inflammation marker, with the strongest effects observed in those with higher baseline inflammation and longer intervention durations.

Two useful details sit inside that. The benefit concentrated in people who started more inflamed, which is the deficiency-correction pattern seen across this site. And it needed twelve weeks or more.

These are blood markers, not symptoms. They are here because the anti-inflammatory claim underwrites much of the marketing above.

What the evidence says about probiotics and weight

Frequently claimed, and the umbrella review that supports it also undermines it.

The review suggests that existing evidence supports the potential benefits of probiotics in managing obesity and overweight.

Its very next sentence is the one to read: it is essential to acknowledge that the credibility of this evidence is somewhat limited due to the inclusion of studies with poor-quality designs and relatively small participant numbers.

Reviewers rarely rebut themselves that directly. Take the second sentence as the finding.

What the evidence says about probiotics and hay fever

A quality-of-life effect without a convincing biological one.

A 2026 umbrella review in allergic rhinitis found probiotics may improve quality of life and nasal symptoms while reducing IgE, the antibody involved in allergic reactions. Several other measures, including eosinophil count and eye symptoms, did not move.

The conclusion is explicit about how far that goes: routine clinical recommendation is not supported by high-certainty evidence, with the reviewers citing high heterogeneity, fragility of the IgE finding, and pooling of diverse strains.

What the evidence says about probiotics and heart failure

Short, because the answer is short.

A 2025 review of probiotics and polyphenols in adults with heart failure found the probiotics or polyphenols from the included studies did not demonstrate significant changes in the health indicators analyzed compared with placebo.

The reviewers still describe dietary management for heart failure as promising and call for more research. On what exists, there is nothing here.

Myths about probiotics, and what the evidence says

“Probiotics are all much the same.” NCCIH says the opposite: one kind of Lactobacillus helping does not mean another kind, or a Bifidobacterium, will do anything.

“They are proven for gut health.” One use is well supported, C. difficile diarrhea alongside antibiotics, and even that is rated low certainty. Irritable bowel syndrome remains unsettled.

“They are completely safe.” For healthy adults, largely. NCCIH reports severe or fatal infections in premature infants given probiotics, and an FDA warning to providers.

“More strains means a better product.” In cancer care, multi-strain did outperform single-strain. Nowhere else on this page was that tested, and a longer ingredient list is mostly a marketing decision.

“Probiotics prevent eczema in babies.” Starting them in infancy did not. The benefit in earlier reviews came from courses that began with the mother during pregnancy.

“They help you lose weight.” The umbrella review supporting this describes its own evidence base as poor-quality and small.

Dosing and forms of probiotics

Doses are given in colony-forming units, or CFU, usually in the billions. Trial doses vary enormously and there is no established general amount, because the right dose depends on the organism.

The practical advice that falls out of this page is unusual for a supplement: match the product to the use. Where a trial found a benefit, it found it for a named strain at a named dose, and the shelf equivalent is whichever product states the same strain rather than the highest CFU count.

Refrigeration and expiry matter more than for most supplements, because the organisms have to be alive on arrival.

Safety, side effects and who should be careful

Most healthy adults get nothing worse than bloating and wind in the first days.

The serious warning is narrow and important. NCCIH reports that cases of severe or fatal infections have been reported in premature infants who were given probiotics, and that the US Food and Drug Administration has warned health care providers about this risk.

The same logic extends to anyone whose immune system is suppressed: by chemotherapy, by transplant medication, or by illness. Live organisms behave differently in a body that cannot police them, and Cochrane’s safety reassurance applies specifically to people who are not immunocompromised.

Anyone with a central line, short bowel syndrome or serious illness should ask a doctor before taking live organisms.

Interactions

Probiotics have no significant interaction with most medicines, and the one that matters is timing rather than chemistry: taken alongside antibiotics they are usually spaced a couple of hours apart, so the antibiotic does not kill the organisms as they arrive.

The more meaningful caution is with immunosuppressant drugs, where the safety concern above applies.

Bottom line on probiotics

There is one use here supported well enough to act on: alongside a course of antibiotics, to reduce the chance of C. difficile diarrhea. Even that sits at low certainty, with a 1.6 percentage point absolute benefit and industry funding across most of the trial base.

Everything else ranges from modest to unproven, and almost every review reaches the same explanation for why: they pooled dozens of different organisms and called the result probiotics.

Until trials test named strains for named conditions, the honest position is that this is a promising category with one good answer in it. Fermented food is a reasonable everyday choice and a cheap one. A specific product for a specific condition is a question for a doctor, not a label.