The best probiotic for IBS, and whether the dose is on the label

Two strains carry the IBS grade on this site, and getting either of them at the amount that was actually tested is a different problem from finding them on a shelf. Lactiplantibacillus plantarum 299v (DSM 9843) and Bacillus coagulans Unique IS2 (MTCC 5260) are the strains behind that row. Both are printed on labels in this audit. For 299v there is no published trial amount to match a panel against, because the trial that carries the strain reported "one capsule" and no CFU figure — so "is this the dose that worked" has no answer for it, on any label, at any price.

What that leaves is a narrower question a label can actually answer: does the panel print an amount for this organism, or is it inside a blend total? A blend total is not a small disclosure problem here. It means the strain doing the marketing could be present as a trace, and the panel would look the same. That is the axis this page ranks on, because it is what the label gives a buyer to work with.

This is the buying page. It reads the audited labels against the strains that earn a grade, and it sends you to the trial record rather than reprinting it.

What the grade covers, and what it does not

The grade on this site is a statement about the strength of the evidence for a named strain, and it is not a statement that any product does anything. A Grade C is one positive randomised trial of a named strain with a material design defect in it. That row, its wording and the trials under it are argued on the evidence page, which is where every number on this page comes from and where you should go if you want the confidence intervals rather than the shopping.

The row of the 22-row outcome table this page shops against, reproduced unchanged. One row: as numbered in the full table. Sortable columns are marked; sorting needs JavaScript, the order above does not.
# Outcome Best grade at strain level The strain, and the trial behind it Buyable at the dose that was studied?
3 IBS pain and bloating (diagnosed IBS) Grade C for IBS pain and bloating (diagnosed IBS) — one positive trial with a material design defect Lactiplantibacillus plantarum 299v (DSM 9843); Bacillus coagulans Unique IS2 (MTCC 5260) Partly — 299v sits in one product inside a proprietary blend, so its dose is unknowable; Unique IS2 is in none

That row is assigned and argued on the IBS evidence page — the trials, the sample sizes, the sex splits, who funded them, and the guideline bodies that decline to recommend any of it. Every number this page uses comes from there or from the label records below. If you want the literature rather than the shopping, that is the page to read, and this one will keep pointing you at it.

IBS-D, IBS-C, IBS-M — what this audit can tell you, and what it cannot

The guides that rank on this question are organised by subtype — all-around, then IBS-D, then IBS-C, then IBS-M — and it is a reasonable shape, because subtype is the thing a diagnosed reader already knows about themselves. This audit cannot honestly publish that shape, and the reason is worth more to a buyer than the shape would be.

Searching every organism note, red-flag summary and lead finding in the teardown collection for a named IBS subtype returns the counts below. Almost everywhere a subtype appears, it appears as the population a trial recruited — "a trial in diarrhoea-predominant IBS" — which is not the same claim at all. A trial run in one subtype tells you who was enrolled. It does not tell you that the strain works better in that subtype than in another, because nobody randomised the comparison.

The count of records that go further and report a result for a subtype is below too, and it is the honest ceiling on the segmentation this category sells. The record it points at is a subgroup sentence, from a trial of one brand's own formula, with an author affiliated to that brand — and the subgroup it favours is "IBS-D and female", on a product whose page title reads "Best Daily Probiotic for Men". A subgroup that was not the trial's primary comparison is a hypothesis; sold as a segmentation, it is a claim nobody tested.

Teardowns naming a diarrhoea-predominant population
7 of 166
Teardowns naming a constipation-predominant population
0 of 166
Teardowns naming a mixed-type population
0 of 166
Teardowns whose record reports a result for an IBS subtype
1 of 166

Scope, stated because these counts are a published claim: they search the lead finding, the organism-slot notes and the red-flag summaries of all 166 teardowns in this audit. They report where this audit's own records name a subtype. They are not a survey of the IBS literature, and a zero above means “absent from these records”, never “no such trial exists”.

The two strains behind the grade, and what their labels print

What follows is the buying branch as it was published on the home page, unedited, and then the labels. The branch states two facts about panels; both are re-checked against the collection on every build, and a change in either fails the build rather than shipping a recommendation the records no longer support.

If you have diagnosed IBS, two strains are worth asking about, and neither is cleanly buyable. L. plantarum 299v (DSM 9843) has the largest male-heavy trial in the digestive literature — n=214, 70.6% male, pain severity 0.68±0.53 vs 0.92±0.57, P<0.05 (PMID 22912552 (opens in a new tab)). B. coagulans Unique IS2 (MTCC 5260) has the single most male-heavy digestive RCT on record — n=136 randomised, 72.22% male, ≥50% pain responders 84.91% vs 12.73%, p<0.001 (PMID 31434935 (opens in a new tab)), though it was fully funded by its manufacturer with all authors or their organisations financially compensated. 299v appears in Nature's Bounty Probiotic 10 (Ultra Strength) under Probi licence, inside a proprietary blend, so the per-strain dose is not knowable — and the source trial's own abstract gives only "one capsule" with no CFU figure, so even the trial dose is unstated. Unique IS2 is in none of the 27 products in the frozen founding cohort — though it does appear, with a disclosed per-organism dose, on labels torn down since; the branch below names them.

L. plantarum 299v (DSM 9843)

5 of the 166 labels in this audit print this designator, and 4 of those print an amount for this organism. The remaining one buries it in a blend total, where the amount cannot be read at all.

B. coagulans Unique IS2 (MTCC 5260)

3 of the 166 labels in this audit print this designator, and 2 of those print an amount for this organism. The remaining one buries it in a blend total, where the amount cannot be read at all.

Where a label and a trial can actually be compared

Of the strains on this page, one can be laid beside its trial and compared, and it is not either of the two above. B. coagulans MTCC 5856, sold as LactoSpore, was tested at 2×10⁹ CFU a day — a published amount, which is the thing 299v does not have, and some of the labels here print a per-organism count for it — so for once a reader can check the number on the box against the number in the trial instead of guessing. The count of labels that actually do is below, and it is smaller than the count that prints the strain.

8 of the 166 labels print this designator. 3 print a count of organisms for it — the unit the trial dosed in — and those are the only ones that can be held against the trial at all. 3 print milligrams and no count, and 2 bury the organism in a blend total. Compare each printed count below against the 2×10⁹ CFU a day the trial used, and read a milligram figure as what it is: a weight of powder, disclosed honestly, and silent on the question you are asking.

What the shelf sells for IBS instead, and how many bottles carry it

These four are the strains the shelf actually sells for IBS and digestion, and none of them is behind the graded row above. They are here because a buyer will meet them, priced as though they were the answer, and because the count of labels printing each one shows how far the marketing has travelled from the trials. The trial record for every one of them is on the evidence page; what this page adds is how many bottles carry them, and what those bottles disclose.

B. longum subsp. longum 35624
5 labels print it; 0 print a count of organisms for it 5 print milligrams and no count.
L. rhamnosus GG (ATCC 53103)
36 labels print it; 9 print a count of organisms for it 1 prints milligrams and no count. 26 bury the organism inside a blend total.
B. animalis subsp. lactis HN019
16 labels print it; 7 print a count of organisms for it 1 prints milligrams and no count. 8 bury the organism inside a blend total.
B. animalis subsp. lactis BB-12
3 labels print it; 0 print a count of organisms for it 3 bury the organism inside a blend total.

What each of those strains has behind it — including the two whose largest trials missed the endpoints they are sold on — is set out strain by strain on the IBS evidence page. If bloating and gas are the reason you are reading this, the two strains sold hardest for that symptom are taken apart on the bloating and gas page, which is where their failed endpoints are read against the bottles that carry them.

The product in this audit sold for IBS by name

One product in this audit puts the condition in its own name. It is worth reading against its panel, because the gap between the two is the whole argument of this page in a single SKU: the marketing is strain-specific and the label is not. Its own practitioner FAQ opens by saying that the health effects of probiotics vary between species and strains — and then the panel names three organisms at species level, inside one proprietary blend, with no designator against any of them and no per-organism amount.

1 of the 166 teardowns carries the condition in its product name: Klaire Labs Ther-Biotic Pro IBS Relief, 21 capsules (medical food; UPC 709616310878). Its panel names 3 organisms and prints 0 strain designators between them. Disclosure score 7.3, rank 152.

What to do with all of that

If you are going to try one anyway — which is a defensible thing to do, and there are guideline positions that support trying one inside a time limit — then buy on the two things a label can settle. Buy a panel that prints the designator, and buy a panel that prints an amount for the organism you are buying it for. Everything else on the box is a claim about a product; those two are facts about a label, and they are the part of this purchase you can check before you swallow it.

Then set an end date before you start, and hold to it. The guideline positions on how long to try, and the bodies that decline to recommend probiotics for IBS at all, are set out with their wording and their certainty ratings on the evidence page. They are not reprinted here, because a shopping page that quietly rewrites a guideline is how the wording drifts.

Scope for every count on this page: the organism slots transcribed from 156 of the 166 labels in this audit. A panel that is not fully transcribed can hide a line, so “not printed on any panel this audit has transcribed” means exactly that and not “not sold anywhere”. A panel printing a designator is a fact about the panel: it is not evidence that the product was tested, and not evidence that it delivers the amount that was studied.

The buying branch above was published on the home page under #buy-ibs and is relocated here unchanged. The counts around it are computed from the product collection on every build; nothing on this page is typed across from another one.