A blood test comes back showing vitamin B12 in the normal range, and the conversation moves on. But for older adults, that reassurance may be premature. A study of 231 cognitively healthy seniors found that those whose B12 sat at the lower end of the accepted reference range still showed measurable signs of brain trouble: slower cognitive processing, more white-matter damage on MRI, and elevated markers of nerve cell injury in spinal fluid. The findings, published in May 2026 in Annals of Neurology by a team led by researchers at the University of California, San Francisco, raise an uncomfortable question for clinicians: are the lab thresholds meant to flag B12 problems actually catching the people whose brains are quietly paying a price?
What the UCSF team found
The researchers recruited older adults who had no diagnosis of cognitive impairment and whose B12 levels fell within the conventional normal range, typically defined as roughly 200 to 900 pg/mL depending on the laboratory. Even within that band, differences mattered. Participants whose levels clustered toward the lower end performed worse on timed cognitive tasks that measure processing speed, one of the earliest capacities to erode in age-related neurodegeneration.
Brain imaging told a parallel story. Lower B12 was associated with greater white-matter hyperintensity burden, a radiologic signature of small-vessel damage and myelin breakdown that accumulates with aging but accelerates in the presence of vascular risk and nutrient shortfalls. Cerebrospinal fluid analysis added a third dimension: concentrations of neurofilament light chain, a protein released when nerve fibers are damaged, were higher in participants with lower B12. The convergence of cognitive, structural, and biochemical signals in the same cohort is what gives the study its weight. A single abnormal measure might be noise. Three pointing in the same direction is harder to dismiss.
The paper, published under DOI 10.1002/ana.27200, did not emerge in isolation. Earlier work from the Health, Aging, and Body Composition (Health ABC) cohort had already linked B12 status to peripheral nerve conduction velocity in older adults, showing the vitamin’s reach extends beyond the brain to the body’s entire wiring system. And the Singapore Longitudinal Aging Study, which followed cognitively normal older adults over time, found that low B12 combined with elevated homocysteine predicted incident cognitive impairment or dementia years later. The UCSF analysis slots into a growing body of evidence from multiple countries and study designs, all pointing toward the same conclusion: B12 status and nervous system integrity are linked, and the link does not switch off just because a lab report says “normal.”
Why the standard lab test may mislead
Part of the problem is what the test actually measures. The total serum B12 assay, the one ordered in most clinical settings, captures both active and inactive forms of the vitamin circulating in blood. Research on holotranscobalamin, the fraction of B12 that cells can actually use, has shown that total B12 can misclassify patients whose functional B12 status is genuinely insufficient. A person might register 300 pg/mL on a standard panel and still have inadequate active B12 reaching the brain.
The UCSF team did not report holotranscobalamin measurements, so it remains unclear whether a more targeted assay would have sharpened the associations they observed or identified a subgroup at particular risk. Functional markers like methylmalonic acid and homocysteine, which rise when B12-dependent metabolic pathways stall, were not the primary focus of the analysis either. That leaves a diagnostic gap: clinicians who want to act on these findings lack a consensus-backed test to identify which “normal” patients are actually running low where it counts.
Aging itself compounds the issue. After about age 50, many people produce less stomach acid, which is essential for liberating B12 from food proteins. Chronic use of proton pump inhibitors for acid reflux further suppresses absorption. Metformin, one of the most widely prescribed diabetes medications among older adults, is a well-documented driver of B12 depletion. And strictly plant-based diets, which contain no natural B12, leave anyone who does not supplement reliant on fortified foods alone. For older adults juggling several of these risk factors simultaneously, a B12 level that looks adequate on paper may represent a precarious balance.
The limits of what this study can tell us
The most important caveat is structural. The UCSF analysis is cross-sectional: it captured a single time point rather than following the same people over years. It can show that lower B12 and worse brain biomarkers tend to appear together, but it cannot prove that one caused the other. It is plausible, for instance, that early, undetected neurodegeneration changes eating habits or nutrient absorption in ways that drag B12 levels down, reversing the assumed direction of the arrow.
The headline framing of “faster decline” also deserves scrutiny. No longitudinal cognitive trajectories or incident dementia counts were reported for this cohort. The inference of decline rests on cross-sectional proxies: processing speed differences and white-matter lesion volumes that are known to worsen with age. The Singapore study does supply incidence-based outcomes over time, but it examined a different population with different methods, so its results cannot be directly mapped onto the UCSF data.
Then there is the supplementation question. If low-normal B12 is genuinely harmful, can raising it help? A randomized controlled trial that gave B12 supplements to older adults with moderate deficiency (but no anemia) found limited neurologic or cognitive improvement over roughly 12 months. The trial was small and short, so it does not rule out benefits from earlier intervention, higher doses, or longer follow-up. But it does mean no one can currently point to rigorous trial data proving that pushing B12 from the low-normal zone into the mid-normal zone prevents cognitive decline. The hypothesis is plausible. It is not proven.
Generalizability is another open question. The UCSF participants were described as healthy, often well-educated, and with good access to medical care. In populations with higher vascular risk, poorer nutrition, or different genetic backgrounds, the relationship between B12 and brain health could be stronger, weaker, or shaped by factors this study did not measure.
What this means at the doctor’s office
For older adults and the clinicians who care for them, the practical message is narrow but worth hearing. A B12 result that clears the lab’s reference range is not necessarily a clean bill of neurologic health. In patients with unexplained neuropathy, gait problems, or cognitive complaints, checking B12 alongside functional markers like methylmalonic acid or homocysteine can help unmask subtle insufficiency that total B12 alone might miss. Treating frank deficiency remains standard of care and is uncontroversial.
The gray zone is trickier. For individuals whose B12 hovers near the lower boundary of normal, clinicians may reasonably consider dietary counseling (meat, fish, eggs, and fortified foods are the primary sources), closer monitoring, or a trial of supplementation, while being honest that the cognitive benefits of doing so are not yet firmly established. Blanket high-dose supplementation for every older person with a B12 of 250 pg/mL is not supported by the current evidence, nor is promising cognitive protection as a guaranteed outcome.
For researchers and guideline panels, the UCSF findings sharpen the case for large, longitudinal studies that track B12 status alongside advanced imaging, fluid biomarkers, and clinical outcomes over many years. Randomized trials that intervene earlier and follow participants longer are equally needed. Only with that kind of evidence will it become clear whether shifting the definition of “normal” B12 upward, or routinely treating low-normal values, meaningfully changes the trajectory of brain aging.
A signal worth watching, not yet a prescription
Vitamin B12 is cheap, widely available, and carries minimal risk at standard supplemental doses. That profile makes it tempting to leap from “associated with brain changes” to “everyone over 65 should take more.” The science is not there yet. What the UCSF analysis and its supporting evidence do establish is that the line between adequate and inadequate B12 for the aging brain may not sit where laboratory reference ranges currently draw it. Recognizing that signal, investigating it further, and resisting the urge to oversimplify it in either direction is the balance the evidence demands as of June 2026.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
