Many probiotic bottles can’t be traced to a study, so check the strain code

Key takeaways

• In a 2019 Georgetown study of 93 probiotic bottles, only 33 could be linked to human research, and more strains, bigger doses or higher prices didn’t make a bottle more likely to have evidence².
• The FDA still requires probiotics to be listed by weight, which counts dead cells too; live counts in CFUs are voluntary under a 2018 draft guidance that remains a draft⁵⁶⁷.
• A count “at time of manufacture” isn’t what you swallow months later; look for a count promised through the end of shelf life³⁵.
• Small independent lab tests in 2025 and 2026 found products with no live bacteria, missing species or counts below the label¹⁸.
• Use the label decoder and five-step checklist below, and ask a clinician first if you’re seriously ill, immune-weakened or buying for a baby.

In a study published in 2025, scientists at Newcastle University in England bought probiotic products online and grew their contents on lab plates¹. One human supplement promised 200 billion live bacteria per gram¹.

Nothing grew. The team found no detectable live bacteria in it at all¹.

That’s an extreme case from a small study. A more common problem is quieter. In a study published in 2019, Georgetown University researchers examined 93 probiotic bottles bought from four big national chains around Washington, D.C.². They could link only 33 of them, about 35%, to any human study showing a benefit².

The rest might have had research behind them. The labels were simply too vague to tell².

Probiotics are living microbes, and natural-health shoppers often choose them for exactly that reason. This guide shows you how to read the label like a buyer who expects the living part to show up. Our probiotics article covers which strains help with what; this one covers how to pick the bottle.

The strain code is the product’s real name

A probiotic’s full name has three parts: genus, species and strain. ISAPP, the International Scientific Association for Probiotics and Prebiotics, says reputable makers list all three⁴.

The strain code is the string of letters and numbers after the species name. It matters because research is done on strains, not on whole species, and ISAPP notes that different strains can deliver different benefits⁴.

Many bottles skip it. In surveys of Washington, D.C.-area stores, 45% of supplements gave no strain designation at all³. Without that code, neither you nor your doctor can look up the study.

The Georgetown team counted a bottle as evidence-backed only if three things lined up²:

• the label named the strain,
• the strain came at a dose shown to help, and
• at least one controlled human study of it existed.

Even the 33 bottles that passed never mentioned their studies on the label². You have to do that legwork yourself.

“Billions” on the front, milligrams on the back

Flip the bottle over. On the Supplement Facts panel, federal rules require only the total weight of the microbes, in milligrams⁵. That weight includes dead cells along with live ones, so it tells you nothing about how many are alive⁵⁶.

In September 2018, the Food and Drug Administration (FDA) offered a fix. Its draft guidance said companies could also list live microbes in colony-forming units (CFUs), a lab count of cells able to grow⁶. The weight must still come first, with the CFU number in parentheses or on its own line⁶.

The draft sets conditions. The CFU figure must count only live microbes, and it must be accurate and not misleading⁶. When we checked the FDA’s guidance page in October 2026, it still carried the label “draft” and “not for implementation”⁷. Companies follow it voluntarily.

Big numbers sell, but they don’t prove much. The NIH Office of Dietary Supplements (ODS) says many products hold 1 billion to 10 billion CFUs per dose, some hold 50 billion or more, and higher counts aren’t necessarily more effective⁵.

Proprietary blends. Some products list a dozen or more strains under one blend name with a single total. ISAPP notes that multi-strain products can carry up to 30 strains, and one total count can’t tell you whether each strain is present in a useful amount³. The FDA draft asks that microbes in a blend be listed from most to least by weight⁶. Our article on the Supplement Facts panel explains how blends hide doses in general.

A count “at time of manufacture” isn’t what you swallow

Live microbes die slowly on the shelf. ISAPP says most strains lose cells over a product’s shelf life, so the mix near the expiration date differs from the mix at the factory³.

That’s why the phrase “at time of manufacture” matters. It describes a count taken before shipping, warehouse time and months in your cabinet. ODS advises looking for the CFU count at the end of shelf life instead⁵.

The same surveys found that just as many supplements, 45%, gave no count that held through shelf life³. International guidelines from the United Nations food and health agencies call for a minimum live count at the end of shelf life³.

How do makers hit a shelf-life number? They add extra cells at the start, called an overage, so enough survive¹³¹⁴. That’s a normal, honest practice. A label promising the count “through expiration” is the company telling you it did that math.

Independent labs keep finding gaps

How often do products miss? We found no large, recent independent census of U.S. products, so the best data come from small tests.

England, 2025. The Newcastle team tested 12 products, five for people and seven for poultry¹. Four of the five human products tested below their declared counts, though all but one stayed inside the margin of error the team allowed¹. The one outside it was the bottle with no live bacteria¹.

In two other human products, the team couldn’t grow one of the species on the label: Bifidobacterium bifidum in one and Lactobacillus bulgaricus in the other¹. The work was funded by an Iraqi government scholarship, and the authors reported no conflicts¹.

Poland, 2026. Scientists at Poland’s National Medicines Institute, a government lab, tested 15 probiotics bought in pharmacies⁸. On the first round of plate counts, 8 of the 15 were significantly below their declared counts⁸. When they retested products later, all but one of those retested had fewer live cells than the first time⁸.

They also made 11 detections of species that weren’t on the labels, including baker’s yeast in two products that listed no yeast⁸. None held the disease-causing germs they tested for, and most of the contamination traced to one maker⁸.

United states and Canada, 2022. ConsumerLab, a private testing company that also sells a certification program to brands, tested 30 probiotic products⁹. Most held at least their listed amounts, and none were contaminated with harmful germs. One held hardly any live organisms despite listing many strains⁹.

Keep the limits in view. These are small samples, often one batch per product, from different countries¹⁸. Plate counts also miss some live cells that won’t grow in a lab, and the Polish team found higher counts with a second method⁸. Still, the pattern repeats: many products come close, and some miss badly.

Seals check the bottle, not the benefit

A third-party seal tells you an outside group checked what’s in the product. It doesn’t tell you the product works. Our article on third-party testing explains the difference in depth.

Verifying live microbes has proved slow. In 2019, a trade publication reported what it called the industry’s first probiotics verified by the United States Pharmacopeia (USP), a nonprofit standards group¹³. An executive at the maker, a California contract manufacturer, said the work took about three years, much of it stability testing to set the overage needed to meet the label at the end of shelf life¹³.

Even then, DNA tests couldn’t count live cells strain by strain, because DNA can’t tell living cells from dead ones¹³.

So a seal is a real plus for honesty about contents. It’s not a substitute for matching the strain to a study.

A free guide rates products by their studies, with a funding footnote

You don’t have to search medical databases alone. The Clinical Guide to Probiotic Products Available in USA lists specific branded products by health use, with the dose and the level of evidence for each¹⁰¹¹.

It’s produced by the Alliance for Education on Probiotics (AEProbio), with Dragana Skokovic-Sunjic as lead author and an expert review board¹¹¹². ISAPP links to the U.S. guide from its resource pages but says it doesn’t endorse products¹⁰.

The rules for getting in are useful. The Canadian edition requires published clinical evidence for the exact strains in a product. A multi-strain product needs studies of that specific combination, not borrowed evidence for each strain alone¹².

Evidence comes in three levels. Level I means at least one well-designed trial, Level II covers other controlled and observational studies, and Level III rests on expert opinion or descriptive studies¹².

Now the footnote. The 2024 Canadian edition says it was made possible by an unrestricted education grant from AEProbio, the same alliance that produces it¹². It has pages for alliance members in Canada and the U.S., but the text we could read doesn’t name them¹². The guide says its work is independent, that listed products needn’t come from members, and that it isn’t a systematic review or a clinical practice guideline¹².

Use it the way you’d use any sponsored tool: as a starting list. A Level I rating can rest on a single trial, so open that study and check whether its volunteers looked like you.

Storage: follow the label, then keep it cool and dry

ODS says some probiotics need refrigeration and others are fine at room temperature, so the label is your guide⁵. Most supplements sit on unrefrigerated shelves, notes the International Probiotics Association, a nonprofit that promotes probiotic use¹⁴.

Its guidance warns that heat and moisture both cut survival¹⁴. It cites advice to keep fragile strains, including most Bifidobacterium, cold, says chilling a bottle once it’s open won’t hurt, and notes that some labels give no storage directions¹⁴.

A few habits follow from that:

• If the label says refrigerate, buy it from a refrigerated case and get it home cold.
• Skip the steamy bathroom cabinet and the hot car.
• Keep the lid tight and any desiccant packet inside.
• Order refrigerated products online only if the seller ships them cold.

Price buys a count, not proof

Prices vary wildly. In ConsumerLab’s 2022 tests, most products cost between 1 cent and 61 cents per billion live cells⁹. The one that barely held any live organisms worked out to more than $600 per billion⁹.

By our math, a 10 billion CFU daily dose at that common range would run about 10 cents to $6.10 a day, or roughly $3 to $183 a month.

Paying more didn’t buy better evidence in the Georgetown survey. The cheaper, simpler bottles were the ones the team could more often match to a study². Co-author Mary Ellen Sanders, who also wrote the ISAPP label guide cited here, consults for probiotic companies and sits on industry advisory boards, which the university disclosed²³.

Marketing deserves the same scrutiny, even from household names. In 2010, the Federal Trade Commission (FTC) charged that Dannon’s ads called Activia “clinically proven to help regulate your digestive system in two weeks” without adequate support¹⁵. Dannon settled, agreed that ads must say three servings a day were needed, and paid $21 million to 39 state attorneys general¹⁵.

Five steps to a bottle that lives up to its label

Natural-health values point in a clear direction here. You want what’s on the label to be what’s in the bottle. You want proof tied to the specific thing you’re buying, not to a cousin strain or a borrowed study. You’d ask the same of a drug company.

Those standards aren’t anti-supplement. They reward the companies that do the hard work: naming strains, testing stability and publishing trials. This is how to find them.

1. Name your goal. Probiotics help with a few specific jobs. Check our probiotics article or the clinical guide for strains tested for your goal¹⁰.
2. Match the strain code exactly. The label should show the same genus, species and code as the study³⁴.
3. Check the count and when it’s promised. Look for at least the studied dose, guaranteed through the end of shelf life³⁵.
4. Check storage and dates. Buy well before expiration, and store it the way the label says⁴¹⁴.
5. Price it per day, then give it a fair trial. Compare cost at the studied dose, not per bottle. ISAPP suggests giving a product about a month; if you see no benefit, it may not be the right one for you⁴.

How strong is the evidence?

Claim	Evidence level	What it rests on
Many U.S. labels lack the strain or shelf-life count needed to match research	Strong evidence	Washington, D.C.-area store surveys: 45% missing each item; 35% traceable to studies23
Store-bought probiotics reliably hold the live count on the label	Promising but early	Small tests with mixed results: most passed in one, 8 of 15 ran low in another189
A higher CFU count reliably means more benefit	Evidence says no, or unsafe	ODS: higher counts not necessarily better; no link in the Georgetown survey25
A pricier bottle is more likely to be backed by research	Evidence says no, or unsafe	In the Georgetown survey, cheaper bottles were traced to studies more often2
Refrigerated products always beat shelf-stable ones	Not enough evidence	Depends on strain and packaging; follow the label514
A quality seal shows the probiotic works	Evidence says no, or unsafe	Seals verify contents, not health effects13

Questions readers ask

The bottom line

In the Georgetown survey, most probiotic bottles couldn’t be matched to a study because the label left out the strain code or a count that lasts through shelf life. Small independent tests show that many products come close to their stated counts while some miss badly, and more strains, bigger numbers and higher prices don’t buy better evidence.

Start from a goal, find the tested strain, and buy only a bottle that names that strain, promises the studied dose through expiration and tells you how to store it. Treat seals and clinical guides as helpful starting points, and talk with a clinician first if you’re seriously ill, immune-weakened or buying for a baby.

Related reading

• Probiotics work for a few jobs, so match the strain or skip the bottle
• Third-party tested says what’s in the bottle, not that it works
• The supplement facts panel and the blend that hides your dose
• The supplement math that changes which bottle wins
• How to tell a real certification from a logo

Sources

1. Taha MW, Fenwick DJC, Marrs ECL, Chaudhry AS. “Assessing Bacterial Viability and Label Accuracy in Human and Poultry Probiotics Sold in the United Kingdom.” Microorganisms, 2025;13:1933. mdpi-res.com
2. Georgetown University Medical Center. “Researchers Find Labels of U.S. Probiotic Products Lacking” (on Merenstein D, Guzzi J, Sanders ME, Journal of General Internal Medicine, 2019). June 4, 2019. gumc.georgetown.edu
3. Sanders ME. “Decoding a Probiotic Product Label.” International Scientific Association for Probiotics and Prebiotics, February 11, 2022, updated June 21, 2025. isappscience.org
4. International Scientific Association for Probiotics and Prebiotics. “How to Choose a Probiotic.” ISAPP, updated September 5, 2025. isappscience.org
5. NIH Office of Dietary Supplements. “Probiotics: Fact Sheet for Health Professionals.” Updated March 25, 2025. ods.od.nih.gov
6. U.S. Food and Drug Administration. “Policy Regarding Quantitative Labeling of Dietary Supplements Containing Live Microbials: Guidance for Industry” (draft). FDA, September 2018. fda.gov
7. U.S. Food and Drug Administration. “Draft Guidance for Industry: Policy Regarding Quantitative Labeling of Dietary Supplements Containing Live Microbials” (guidance page, docket FDA-2011-D-0376). FDA, accessed October 2026. fda.gov
8. Zawistowska-Rojek A, Rybak J, Smoleń P, et al. “Evaluating the Quality of Selected Commercial Probiotic Products, Both Dietary Supplements and Foods for Special Medical Purposes.” Foods, 2026;15:373. mdpi-res.com
9. ConsumerLab.com. “Best Probiotics Identified” (probiotic supplements review news release). November 10, 2022. consumerlab.com
10. International Scientific Association for Probiotics and Prebiotics. “Clinical Guide to Probiotic Products Available in USA” (resource listing). ISAPP, updated March 31, 2026. isappscience.org
11. Campbell K. “Evidence-Based Guidelines for Probiotic Products Now Available for U.S. Doctors.” Gut Microbiota for Health, May 30, 2016. gutmicrobiotaforhealth.com
12. Skokovic-Sunjic D. “Clinical Guide to Probiotic Products Available in Canada,” 16th edition. Alliance for Education on Probiotics, 2024. cfmws.ca
13. NutraIngredients. “MeriCal’s USP Probiotics Verification Was Years in Making, Exec Says.” November 14, 2019. nutraingredients.com
14. Fleishman C. “Probiotic Supplements: Refrigerate or Not?” International Probiotics Association, July 27, 2021. ipa-biotics.org
15. Federal Trade Commission. “Dannon Agrees to Drop Exaggerated Health Claims for Activia Yogurt and DanActive Dairy Drink.” Press release, December 15, 2010. ftc.gov