The Oasis Health Journal · Submitted July 20, 2026 · 8:16 AM ET

Your body does not really absorb vitamin B12. It processes it, through four departments, each with its own paperwork.

Here is the actual route. B12 shows up bound to protein in your food. Saliva hands it off to a carrier called haptocorrin. Stomach acid and gastric protease pry the rest of it loose from the food matrix, and haptocorrin takes custody of that too. Down in the duodenum, digestive enzymes strip the haptocorrin back off, and a different protein called intrinsic factor, secreted way back up in your stomach by parietal cells, takes over. Only then, at the far end of the small intestine, does the ileum recognize that complex and let it through the door.

Four handoffs. One vitamin. It is the most bureaucratic thing your gut does all day. And there is a reason the standard blood test is bad at auditing it.

The warehouse and the truck

Once B12 is in your bloodstream, it does not all count the same. Roughly 70 to 90 percent of it is bound to haptocorrin, which is basically long-term storage. Your cells cannot touch it. Only the 20 to 30 percent riding on a protein called transcobalamin can actually be pulled inside a cell. That fraction has a name, holotranscobalamin, and a nickname, active B12.

A standard serum B12 test counts both. It counts the pallets parked in the warehouse and the boxes on the delivery truck, adds them together, and prints one number. Your brain only ever sees the truck.

231 people whose lab work all came back fine

In 2025, a UCSF-led team published a study in Annals of Neurology that leaned hard on that distinction. They enrolled 231 healthy older adults from the Brain Aging Network for Cognitive Health cohort. Average age 71. Nobody had dementia. Nobody had mild cognitive impairment. Their average total B12 was 414.8 pmol/L, which is close to triple the US deficiency cutoff of 148 pmol/L. On paper this was a group with nothing to discuss.

Then the researchers measured the active fraction instead of the total. After adjusting for age, sex, education and cardiovascular risk factors, participants with lower active B12 had slower processing speed on cognitive testing, and the effect got stronger with age. They had delayed visual evoked potentials, meaning the signal took longer to travel. And on MRI they had greater volumes of white matter hyperintensities, the small areas of injury in the brain's wiring that track with cognitive decline and stroke risk.

Senior author Ari Green put it plainly: revisiting how we define B12 deficiency to include functional biomarkers could allow earlier intervention. Which is a very restrained way of saying the current cutoff may be measuring the wrong thing.

Now the part where I take some of that back

This was an observational study. It found associations, not causes. Nobody was given B12 and watched to see if their white matter improved, and low active B12 could easily be a marker of something else going on rather than the thing doing the damage.

The supplementation evidence is genuinely mixed. A 2025 systematic review in Nutrition Reviews pooled 17 randomized trials covering 5,275 older adults who took B6, B9 or B12 for at least 26 weeks. Global cognitive function did improve, with a Hedges' g of 0.423, but the analysis carried considerable heterogeneity and the authors rated the certainty of evidence as very low. Separately, a Mendelian randomization study found no clear evidence that genetically higher B12 protects the general population from cognitive impairment, though it used total serum B12, which is the exact measure the UCSF work argues is too blunt.

So the honest summary: filling a real gap matters, topping off a full tank probably does not, and we do not yet have a trial that settles it.

Whose supply chain is already leaking

Four handoffs means four places to fail, and some people fail more of them. Depending on the definition used, between 3 and 43 percent of community-dwelling older adults are B12 deficient by serum levels, with atrophic gastritis a common culprit because it cuts the stomach acid that starts the whole process. Proton pump inhibitors and H2 blockers do the same thing on purpose. Metformin can reduce absorption and lower serum B12. Pernicious anemia takes out intrinsic factor directly. Gastrointestinal surgery can remove the stretch of ileum where the final handoff happens. And because natural B12 is limited to animal foods, vegans and vegetarians start with less arriving at the loading dock in the first place, which is why a B12 formula is usually the first thing recommended on a plant-based diet.

What to ask for at the lab

If a total B12 result lands between 150 and 399 pg/mL, the NIH's own guidance is that it should not be the end of the conversation. Methylmalonic acid is the most sensitive marker of B12 status, and experts suggest checking it in exactly that gray zone. Homocysteine rises as B12 falls, but it is a noisy signal that also responds to low folate and declining kidney function.

Active B12 testing does exist. In the UK, NICE guideline NG239 sets thresholds for it, and at least one NHS trust treats results under 25 pmol/L as consistent with deficiency and over 70 pmol/L as making deficiency unlikely. The catch is availability. That same trust runs the assay only for pregnant patients, twice a week, in batches. The better test is real. It is just not sitting on the shelf next to the standard one.

The dose math is stranger than it looks

Here is the part that confuses people. Absorption through the intrinsic factor route is about 50 percent, but only at doses under 1 to 2 mcg, because that is where intrinsic factor's binding capacity taps out. At 500 mcg you absorb around 2 percent. At 1,000 mcg, about 1.3 percent.

That sounds like a scam until you do the arithmetic. 1.3 percent of 1,000 mcg is 13 mcg, which is more than five times the adult RDA of 2.4 mcg, and it gets in through passive diffusion that does not need intrinsic factor at all. That is the entire logic behind high-dose oral B12, and it is why formulas that pair B12 with intrinsic factor exist for people whose own supply has gone missing. One more honest note: the NIH says there is no evidence that absorption rates differ between forms, so methylcobalamin versus cyanocobalamin is a smaller argument than the internet makes it.

B vitamins are also fragile, which is where our sourcing habit earns its keep. We order professional-grade formulas fresh per order rather than parking inventory on a shelf to slowly lose potency. It makes shipping slower. It makes what shows up more like what the label promised, whether that is a standalone B12 or the B12 inside a multivitamin.

Your body built a four-step chain of custody for one molecule. The least we can do is not test it with a headcount of the warehouse.

This article is for education only. It is not medical advice, and it is not a substitute for talking to your own clinician, especially about lab results, medications or a suspected deficiency.

Your B12 Test Counts the Whole Warehouse. Your Cells Only Get What's on the Truck.

The warehouse and the truck

231 people whose lab work all came back fine

Now the part where I take some of that back

Whose supply chain is already leaking

What to ask for at the lab

The dose math is stranger than it looks

Sources

Leave a comment

Save 10% off your first purchase

The warehouse and the truck

231 people whose lab work all came back fine

Now the part where I take some of that back

Whose supply chain is already leaking

What to ask for at the lab

The dose math is stranger than it looks

Sources