Labels

Three Billion, Five Billion: What A CFU Count Means

Reviewed by Dr. Mark Johnston, MD

One tub carries two different live-culture figures, and the larger one is qualified by four words that are doing real work: at the time of manufacture. NCCIH’s review of the probiotic evidence explains why the count on its own settles less than it looks like it settles.

What a colony-forming unit counts

A colony-forming unit is exactly what the name says — one unit capable of forming a colony on a plate. It is a count of viable organisms, not a weight and not an activity. Five billion CFU says that when somebody plated a sample and counted, five billion things grew.

That is a useful measurement and it is the standard one in this category. What it is not is a statement about what happens afterwards, either in the tub or in a person. NCCIH’s summary of the probiotic evidence is worth reading beside any label carrying a CFU figure, and the how-it-works page uses it the same way.

Two numbers on one tub

This is not a discrepancy between a seller and a critic. It is a discrepancy inside the seller’s own artwork, and the honest thing is to print both.

Two different numbers for the same product, on the same seller’s own artwork. Picking one and printing it alone would hide the more interesting fact, which is that they disagree.

Neither figure is chosen silently anywhere on this website. The panel’s figure is the one with a qualifier attached, so it is the one quoted whenever the measurement condition matters, and the tub’s figure is printed beside it. The panel page and the home page both carry the pair rather than a preference. The labelling rule does not require the two to be reconciled, which is part of why they are not.

Four words doing real work

At the time of manufacture is a measurement condition, and it is the most informative phrase on the probiotic line. Live cultures decline. Heat, time and moisture all take a share — and the share is not printed anywhere on this label.

Why the phrase is used at all

Counting at manufacture is the easiest count to make and the easiest to guarantee, because it happens before the tub has been through a warehouse, a courier and a kitchen cupboard. A count at the end of shelf life is a harder promise and a more useful one. This panel makes the first kind.

What that means for the buyer

It means the number on the panel is a ceiling in the same way a blend total is a ceiling. The count in the glass on a given morning is somewhere at or below it, and nothing on the label says where. Part 111, the manufacturing rule, requires a maker to meet its own established specifications for identity, purity, strength and composition; it does not require a shelf-life count to be printed on the front of a tub. The testing page goes through which seals describe a process and which describe a result.

The strains this panel does name

Credit where it is due: many labels in this category print a genus, a species and a count and stop. This one names strains, and strain codes are the level at which the published research is actually done.

• Bacillus coagulans BC99, given as 20 per cent of the blend.
• Lactobacillus acidophilus LA85, 30 per cent.
• Bifidobacterium lactis BLa80, 30 per cent.
• Lactobacillus reuteri LR08, 20 per cent.
• A further probiotic powder line naming L. casei, L. acidophilus, L. rhamnosus, Clostridium butyricum and Streptococcus thermophilus.
• A separate 2 mg Akkermansia muciniphila AH39 line, outside all four blends.

The percentages are shares of 52.5 mg rather than colony counts — so they cannot be converted into a CFU figure for any one strain. The evidence reviews NCCIH summarises are organised by strain and condition, which is why that conversion would be the useful one and why the panel not supporting it is worth saying. The ingredients page lists the strains in full.

What a count does not settle

This is the section that costs the number some of its authority, and it is drawn from the published reviews rather than from scepticism.

The specific findings behind the fifth row are worth stating. A 2018 review of 53 studies in 5,545 people on irritable bowel syndrome could not identify which species, strains or combinations of probiotics are most likely to be helpful. A 2014 review of 14 studies in 1,182 people on constipation in adults showed some evidence of benefit, especially for Bifidobacterium lactis, which is one of the strains on this panel. A 2017 evaluation of nine studies in 778 elderly people found a small but meaningful benefit. That is a mixed and honest picture, and the benefits page stays inside it. Structure and function wording is all a supplement label is permitted in any case.

What the buyer actually controls

Some of the decline between the factory and the glass is yours to slow down, and none of it requires anything unusual.

1. Keep the lid closed. Moisture is the enemy of a dry culture and the reason silicon dioxide is in the formula.
2. Keep the tub out of direct sun and away from the hob, the kettle and the top of the fridge.
3. Keep the scoop dry, and never put a wet scoop back into the tub.
4. Use the tub inside the supply it was sold as rather than eking it out over a year.
5. Check the date and lot printed on the tub you received. Lot FIN-26/GE-7289 is what this desk holds, and the batch page explains what a lot code can and cannot confirm.

Fifty-two and a half milligrams of it

The Probiotic Blend weighs 52.5 mg. Set that beside the 1000 mg Polyphenol Blend and the 755 mg Gut Health blend and the scale becomes clear — live cultures are a small part of a 3.2 g scoop by weight, which is normal, because organisms are counted rather than weighed.

It is still worth noticing that the marketing weight and the physical weight point in different directions. The front of the tub leads with a CFU figure; the panel gives that line the smallest total of the four. FDA’s own consumer page on what approval means is the right frame for the seals sitting next to it: registration of a facility is not an approval of a product. The testing page goes through each seal on the tub.

What a better live-culture label prints

Three additions, none of which gives away a formula.

• A count at the end of shelf life, not only at manufacture. It is a harder promise, which is exactly why it means more.
• A CFU figure for each named strain, rather than a percentage of a blend weight.
• One figure, not two. If the front of the tub and the panel disagree, the front should follow the panel.

None of that is required by the panel rule or by the manufacturing rule, and that is the point: a label that prints them has chosen to. The testing page keeps a running list of what this one does and does not publish.

What the probiotic line is still worth

Ending on the caveats would misrepresent the label. This panel names strains with codes, gives the blend a weight, gives the Akkermansia line a weight of its own, and attaches a measurement condition to its count. A great many tubs in this category do none of those things.

The honest summary is that the number is real and the qualifier is real — and the conclusion a buyer can draw from the pair is narrower than the front of the tub implies. That is a reasonable basis on which to try something. The reviews page is where the buyers on this desk’s record say what a month of it was actually like, and the timeline article puts their days on a calendar.