A study of 400,000 patients found that your white blood cell ratio predicts Alzheimer’s risk long before any symptom appears

The warning sign for Alzheimer’s disease has been sitting in your routine bloodwork for years. Nobody was looking at it the right way.

A landmark study published in Alzheimer’s & Dementia in April 2026 by researchers at NYU Langone Health and the Veterans Health Administration analyzed nearly 400,000 patients and found that a standard metric already present in a complete blood count, the ratio of neutrophils to lymphocytes, can predict elevated risk of Alzheimer’s and related dementias years before a single cognitive symptom appears. The data was sitting inside existing medical records the entire time. The question was never whether the signal existed. The question was whether anyone was measuring it with the right question in mind.

What a Neutrophil Actually Is

Neutrophils are the immune system’s first responders. When your body detects infection, injury, or inflammation, neutrophils are the cells that arrive fastest and in the greatest numbers. They make up the largest share of white blood cells circulating in human blood at any given moment, and their count fluctuates rapidly in response to biological stress.

Doctors already measure neutrophils routinely. A complete blood count, one of the most commonly ordered laboratory tests in medicine, captures their levels automatically. From those numbers, clinicians calculate a value called the neutrophil-to-lymphocyte ratio, or NLR: the number of neutrophils divided by the number of lymphocytes. The NLR has been used for years as a general marker of systemic inflammation and infection severity. What nobody had tested at population scale was whether a chronically elevated NLR, measured years before any diagnosis, could predict who would eventually develop dementia.

The Scale of the Evidence

The NYU Langone team, led by Dr. Jaime Ramos-Cejudo and data scientist Tianshe He, addressed this gap using electronic health records from two large and demographically distinct populations: NYU Langone Hospitals in New York and the Veterans Health Administration.

Patients entered the analysis after age 55, before any recorded Alzheimer’s or dementia diagnosis. The researchers then followed those patients forward through time, tracking who eventually received a dementia diagnosis and working backward to see what their earlier NLR values had looked like. After statistical adjustment for confounding factors, the results were consistent across both health systems. Higher NLR values were associated with a 7% higher recorded dementia risk in the New York cohort and a 21% higher risk in the veterans population. The association held for both short-term and long-term dementia risk, meaning the signal appeared across different time horizons before diagnosis.

This is the first large-scale investigation to demonstrate that neutrophil metrics are associated with increased dementia risk in humans. Previous work had found neutrophil-related markers in Alzheimer’s brain lesions and noted associations between neutrophil levels and molecular markers of the disease in smaller samples. This study put that biological hypothesis in front of nearly 400,000 real patients and it held.

The Signal Is Not Equal for Everyone

One of the more medically significant findings involved who the signal was strongest for. Subgroup analysis revealed that the NLR association with dementia risk was notably stronger in female patients and Hispanic patients across both health systems.

The reasons behind this disparity remain unsettled. Biology, differences in healthcare access, the presence of other conditions, and stress-related inflammatory burden could all contribute. The researchers were clear that the data could not separate these factors from one another. But the finding matters because it suggests the NLR may be a particularly useful screening tool in populations that are already at elevated dementia risk and historically underserved by early diagnostic protocols.

The Question the Data Cannot Yet Answer

The most important scientific question the study leaves open is also the one with the most treatment implications: are neutrophils a marker of Alzheimer’s risk, or are they actively causing it?

These are meaningfully different possibilities. If neutrophils are simply a passive signal, rising in response to whatever biological process is already underway, then the NLR is a useful detection tool but not a therapeutic target. If neutrophils are actively participating in the damage, crawling into blood vessel walls, triggering neuroinflammation, and accelerating the tissue destruction that drives cognitive decline, then they become something far more important: a potential point of intervention.

Evidence from animal studies points toward active participation. Neutrophil-driven inflammation has been observed in the brains of Alzheimer’s patients, and in animal models blocking neutrophil activity has reduced some of the vascular and neurological damage associated with the disease. The human data from this study does not prove causation, but it is consistent with the hypothesis that the immune system is not merely watching the disease unfold. It may be helping to build it.

Why This Hasn’t Changed Clinical Practice Yet

A legitimate question arising from this research is why, if the NLR is already measurable in routine bloodwork, it isn’t already being used for Alzheimer’s screening. The answer involves both the nature of the signal and the state of the treatment landscape.

An elevated NLR on its own is not sufficient to predict dementia with clinical certainty. Neutrophil levels fluctuate in response to dozens of conditions: infection, physical stress, chronic disease, medications. A single elevated reading means almost nothing in isolation. What this study demonstrates is that a pattern of persistently elevated NLR, tracked over time in patients over 55 and combined with other known risk factors, could meaningfully improve the identification of people who should receive more comprehensive neurological evaluation before symptoms emerge.

Dr. Ramos-Cejudo framed the practical application clearly: the NLR could contribute to gateway diagnostic tools for people at risk, enabling more in-depth testing and early interventions long before cognitive decline becomes visible. In a disease where the pathological process begins a decade or more before any symptom appears, that window is everything.

What Early Detection Actually Buys

The reason early detection matters for Alzheimer’s has changed significantly in recent years. For most of medical history, catching Alzheimer’s early was a psychological and logistical benefit, giving patients and families time to plan, but it offered no biological advantage because no treatment could alter the disease course.

That calculus is shifting. Disease-modifying therapies targeting amyloid and neuroinflammation are advancing. The window in which these interventions appear most effective is early, before substantial neuronal loss has occurred. A screening tool that identifies elevated risk in people who are cognitively intact and over 55 creates exactly the population these therapies need: people early enough in the biological timeline to potentially benefit.

The data sitting inside existing blood tests is not a diagnosis. It is a door. And for the first time, researchers have shown it was there all along.