Best Probiotics for IBS: The Strains With Real Evidence

Ask five gastroenterologists which probiotic to try for IBS and you'll get five different answers, or a shrug. That isn't because the field is a scam. It's because "probiotic" isn't a molecule; it's a category that includes dozens of species and hundreds of strains, each with its own behavior in the gut. The 2021 American College of Gastroenterology guideline conditionally recommends against probiotics for global IBS symptoms based on very low quality evidence. The American Gastroenterological Association agrees. But dig into the trial data by strain, and a few products keep showing up on the winning side, especially when you sort patients by IBS subtype.

This is the strain-level guide the guidelines don't give you. What follows is what actually has evidence for constipation-predominant IBS (IBS-C), diarrhea-predominant IBS (IBS-D), and the mixed subtype (IBS-M), which doses were used, and how to run an 8-week trial that tells you if it's working for you.

Why IBS subtype changes the answer

IBS is not one disease. Rome IV criteria split it into four subtypes based on stool form: IBS-C (constipation-predominant, hard or lumpy stools on more than a quarter of days), IBS-D (diarrhea-predominant, loose or watery on more than a quarter of days), IBS-M (mixed, both types on more than a quarter of days), and IBS-U (unclassified). The NIDDK uses the same framework.

This isn't administrative box-ticking. A probiotic that relaxes bowel transit and softens stool is helpful in IBS-C and potentially disastrous in IBS-D. A strain that firms things up is the opposite. Meta-analyses that lump every IBS patient together average these opposite effects and land close to zero, which is a big part of why the field looks so muddled. The 2021 strain-specific systematic review of 42 RCTs makes the point cleanly: benefit is real, but only when you match the strain to the outcome you care about.

Before you spend money on a probiotic, know which subtype you have. If you're not sure, tracking stool form on the Bristol Stool Chart for two weeks will usually make it obvious. Types 1 to 2 dominating means IBS-C, types 6 to 7 dominating means IBS-D, both showing up regularly means IBS-M.

Best probiotics for IBS-D (diarrhea-predominant)

Saccharomyces boulardii CNCM I-745

This is technically a yeast, not a bacterium, which is part of why it works so well in diarrhea. It's not affected by antibiotics, it doesn't colonize the gut long-term, and it has decades of trial data. A Cochrane-adjacent systematic review of 22 trials found S. boulardii reduced the duration and severity of acute infectious and antibiotic-associated diarrhea. For IBS-D specifically, a 2025 clinical review confirmed it as one of the most effective options for managing diarrhea, with significant improvements in abdominal pain intensity and bowel movement frequency.

Dose used in trials: 250 to 500 mg twice daily (roughly 5 to 10 billion CFU).

What to look for on the label: the strain designation "CNCM I-745" (also sold as Florastor in the US). Generic S. boulardii without a strain code has less consistent data. If it just says "Saccharomyces boulardii" with no strain, it might work, but you're gambling on a bottle-by-bottle basis.

Bacillus coagulans MTCC 5856

The dark horse of IBS probiotics. A 2016 double-blind randomized trial in IBS patients found this specific strain, at 2 billion CFU once daily, significantly improved abdominal pain, bloating, straining, and stool consistency versus placebo over 90 days. A more recent network meta-analysis of 43 randomized controlled trials covering 5,531 patients ranked Bacillus coagulans first overall for IBS symptom improvement, with the biggest effects on IBS-D.

Bacillus species are spore-formers, which means they survive stomach acid intact and don't need refrigeration. That's operationally convenient, but it also means the "50 billion CFU" advertised on rival products doesn't translate directly - most non-spore probiotics lose a huge fraction of viable cells before reaching the intestine.

Dose used in trials: 2 x 10⁹ CFU once daily, taken with a meal.

Lactobacillus plantarum 299v

An RCT of 214 IBS patients found this strain, at 10 billion CFU daily for four weeks, produced significant improvement in bloating, abdominal pain, and stool frequency compared with placebo, with the biggest effects in IBS-D. The strain is patented and sold as GoodBelly and a handful of European brands. It doesn't have the volume of data S. boulardii does, but it has the cleanest single-strain IBS-D signal outside the yeast.

Dose used in trials: 10 billion CFU once daily.

Best probiotics for IBS-C (constipation-predominant)

Bifidobacterium longum W11

The most current and specifically IBS-C targeted evidence. A 2025 clinical evaluation published in Frontiers in Medicine found B. longum W11 significantly improved IBS-C symptom severity, stool consistency toward Bristol type 3 to 4, and quality of life scores without notable side effects. Effects showed up around week 4 and strengthened through week 8.

W11 is a lactose-fermenting strain that produces short-chain fatty acids the colon uses as fuel. Because it also seems tolerant of antibiotics, it's been tested alongside rifaximin without losing viability, which is useful for people whose IBS-C overlaps with methane-predominant SIBO. If that's you, our SIBO vs IBS guide is worth reading before you buy anything.

Dose used in trials: 5 x 10⁹ CFU daily.

Bifidobacterium lactis HN019

Not exclusively an IBS strain, but has decent transit-acceleration data. A double-blind trial of adults with functional constipation found HN019 reduced whole-gut transit time from about 49 hours down to 21 hours over two weeks, with improvements in stool frequency and consistency. That's a bigger effect than most fiber supplements produce. It's not IBS-specific, but if your IBS-C is dominated by slow transit rather than pain, it's a defensible pick.

Dose used in trials: 1.8 x 10¹⁰ CFU daily.

Best probiotics for IBS-M and unclassified

Bifidobacterium infantis 35624

The most-studied single strain in IBS overall. The landmark 2006 trial in women with IBS tested three doses and found 1 x 10⁸ CFU daily outperformed placebo across abdominal pain, bloating, incomplete evacuation, and gas. Higher and lower doses failed. That specific dose matters - most modern products bury this strain in a multi-strain formula at unclear individual counts, which is a problem if you're trying to reproduce the trial.

Sold as Align in the US. A 2017 strain-specificity review pulled together the subsequent literature and concluded the benefit is genuine but modest, most reliable for bloating, and less consistent for stool form. It's a reasonable starting point for IBS-M or when you don't have a clean subtype.

Dose used in trials: 1 x 10⁸ CFU daily.

Multi-strain formulations (VSL#3 / Visbiome)

An eight-strain combination originally sold as VSL#3 (formulation now sold as Visbiome after a trademark dispute). A controlled trial in children with IBS found the combination significantly improved abdominal pain and bloating over six weeks. Adult data are more mixed but the AGA conditionally recommends this specific combination for maintenance of remission in pouchitis, which is one of the few probiotic indications with any professional-society endorsement.

Very expensive. The dose used in the pouchitis trials (900 billion CFU daily) is not the dose used in the IBS trials (450 billion CFU), and running a real course costs several hundred dollars. Worth it if simpler single-strain options haven't moved the needle for you. Not worth it as a first try.

The 8-week trial protocol

The single biggest mistake with probiotics is buying a bottle, taking it inconsistently for two weeks, and concluding "it didn't work." Trials that show benefit run for 4 to 12 weeks with daily dosing at a consistent time. If you're going to try one, do it properly.

Pick one product. Match strain to subtype using the guide above. Do not stack two or three probiotics simultaneously; you won't know which one moved anything.

Hold everything else constant for two weeks before you start. Same diet, same fiber intake, same sleep schedule, same coffee habit. Baseline your symptoms during this period.

Start the probiotic at the trial-tested dose. Take it at the same time of day. Most strains prefer with food, especially Bacillus and Bifidobacterium species. S. boulardii works fine on an empty stomach.

Track daily. Stool form on the Bristol scale, bowel movement frequency, bloating on a 0 to 10 scale, and pain if you have it. This is exactly the loop the Number Two app is built for, but a paper log works too - the discipline is the point.

Evaluate at week 4 and week 8. If nothing has changed by week 4, keep going but expect the probiotic isn't going to be your answer. If nothing has changed by week 8, stop taking it. If your key symptom is at least 30% better than baseline, that's a real effect worth continuing.

Reassess yearly. The gut microbiome shifts. A probiotic that worked at 32 may not work at 42. Trials rarely run longer than 12 weeks, so we don't actually have great data on what sustained benefit looks like.

How to read a probiotic label without getting fleeced

The supplement aisle plays on the fact that most buyers can't distinguish strain-level science from marketing.

Genus, species, AND strain must all be listed. "Lactobacillus plantarum" tells you almost nothing. "Lactobacillus plantarum 299v" tells you which strain has been studied. If the label only lists genus and species, you cannot match the product to any clinical trial. Move on.

CFU count is not quality. Higher is not better. Bifidobacterium infantis 35624 at 10⁸ beat 10¹⁰ in the original trial. Bacillus coagulans works at 2 billion, not the 50 billion the label brags about. Buy the dose that matches the trial, not the biggest number on the shelf.

"Through expiration date" not "at time of manufacture". Live bacteria die on the shelf. Reputable manufacturers guarantee CFU counts through the printed expiration date. The alternative wording lets the product be functionally empty by the time you buy it. Mayo Clinic flags this as the single most useful quality check for a consumer.

Third-party testing. USP, NSF, or ConsumerLab seals mean the product actually contains what the label claims. The FDA does not pre-approve supplement labels, which is a real problem in this category. A 2019 audit of 22 commercial probiotics found frequent discrepancies between labeled and actual strain content.

Skip the additives-heavy formulas. If the product includes prebiotics, herbs, digestive enzymes, and "detox complex" alongside the strain, you can't attribute any effect to the probiotic itself. That's fine if you don't care why it worked, but bad if you want to know whether to keep buying it.

When probiotics can make IBS worse

Probiotics have a low-but-not-zero risk profile and a specific failure mode in a subset of patients. If your IBS is actually SIBO (small intestinal bacterial overgrowth) masquerading as IBS, adding more bacteria to a small intestine that already has too many will predictably increase bloating, gas, and cramping. Somewhere between 4 and 78% of IBS patients also meet SIBO criteria depending on the diagnostic test used, so this is not a rare edge case.

If a strain-appropriate, trial-dose probiotic makes your symptoms worse instead of better in the first two weeks, stop it. Consider whether SIBO fits your pattern - our SIBO symptoms guide covers the presentation and how it differs from garden-variety IBS.

Probiotics can also cause bloodstream infections in immunocompromised patients, people with central venous catheters, short-bowel syndrome, or acute pancreatitis. The FDA-mandated warning is not decorative - if you fall into one of these groups, do not start a probiotic without your doctor.

What the pillar of the evidence really says

The professional societies are right that we don't have the trial base to recommend probiotics for IBS as a class. Individual strains, matched to subtype, have modest but replicable effects. The question isn't "do probiotics work for IBS." It's "does the strain I'm buying have trial data for the subtype I have at a dose I can actually hit?" If the answer is yes, an 8-week trial is a reasonable, low-risk experiment. If the answer is no, save the money and start with diet - the low FODMAP diet and identifying your personal food triggers move the needle for more people than any probiotic will.