What Can an Alzheimer's Blood Test Tell You?

Blood tests can now pick up signs of Alzheimer's disease that once needed a PET brain scan or a lumbar puncture. That raises a harder question: if you feel completely well at 65 or 70, can a blood test tell you whether you will go on to develop problems with memory and thinking? A 2026 study of 2,684 cognitively healthy older adults moves us closer to an answer, and shows why healthy people should not rush out to be tested yet.

TL;DR

• Alzheimer's biology can be detected in blood. One of the leading biomarkers is phosphorylated tau 217, or p-tau217.
• A large JAMA study tested prediction. It followed 2,684 cognitively unimpaired adults, and over a median 5.4 years, 478 developed cognitive impairment.
• Higher p-tau217 meant higher risk. Each standard-deviation increase was associated with a hazard ratio of 1.38 for progression to cognitive impairment.
• The headline numbers are absolute risks. Very high p-tau217 was linked to an estimated 38% five-year risk and a modelled 78% ten-year risk, though the ten-year figure rests on more limited data.
• They do not transfer to the general population. Participants came from selected research cohorts, and 43% were already amyloid-positive at the start.
• A positive biomarker is not a dementia diagnosis. It changes estimated risk but cannot tell an individual whether or when dementia will develop.
• The NHS is testing blood biomarkers for diagnosis. Current UK trials focus on people referred to memory services, not on screening healthy adults.

What Does an Alzheimer's Blood Test Measure?

It does not detect forgetfulness; it detects biology. Alzheimer's disease is associated with abnormal changes in two proteins in the brain, amyloid beta and tau, and these can begin years before obvious symptoms. One of the most promising blood biomarkers is p-tau217, a phosphorylated form of tau closely linked to Alzheimer's pathology. Measuring it in blood could make those changes far easier to detect than an expensive PET scan or a sample of cerebrospinal fluid taken by lumbar puncture.

That leaves two separate questions: is Alzheimer's pathology present, and will this person develop cognitive impairment? The evidence for each is at a different stage.

Blood Tests Can Already Detect Alzheimer's Pathology

In May 2025 the US Food and Drug Administration cleared the first blood test to aid Alzheimer's diagnosis. The Lumipulse G pTau217/β-Amyloid 1-42 Plasma Ratio measures p-tau217 alongside amyloid. In 499 samples assessed by the FDA, 91.7% of people with a positive result had amyloid plaques confirmed by PET or cerebrospinal fluid testing, and 97.3% of those with a negative result were negative on confirmatory testing.

The test was cleared for people aged 55 or over who already show signs and symptoms of cognitive decline. It is not intended to screen healthy people or to diagnose Alzheimer's on its own.

Can It Predict What Happens Next?

That is the question the new research set out to answer. The study was published in JAMA on 14 July 2026 and presented at the Alzheimer's Association International Conference the following day. It pooled 2,684 cognitively unimpaired adults from six observational and clinical trial cohorts, with a median age of 69.6, and 63% were women.

Researchers measured p-tau217 at the start and followed participants for a median 5.4 years. In that time, 478 developed cognitive impairment, defined as a diagnosis of mild cognitive impairment or dementia, or two consecutive global clinical dementia rating scores of 0.5 or higher. Each standard-deviation increase in p-tau217 was associated with a hazard ratio of 1.38 for progression. After accounting for amyloid PET results, the hazard ratio was 1.32, which suggests p-tau217 carries information about future risk beyond whether amyloid is already present in the brain.

The more striking numbers came from grouping participants by level. Among people with very high p-tau217, the estimated absolute risk of cognitive impairment was 38% within five years, and the modelled ten-year risk reached 78%. Those figures sound close to a predictive test, but they are not one yet.

Why 38% Doesn't Mean You Have a 38% Risk

The study population matters. At baseline, 43% of participants were amyloid-positive. They were not randomly selected older adults but members of research and trial cohorts set up to study Alzheimer's disease, so the 38% five-year risk cannot simply be applied to a random 70-year-old with the same blood result. The authors say the estimates need validating in population-representative cohorts before they can be used for individual prognosis.

They also place more confidence in the five-year estimates than the ten-year figures, because long-term data were sparse and concentrated in a subset of the cohorts. And the endpoint was cognitive impairment of any cause, not Alzheimer's dementia alone. People can develop cognitive problems through vascular disease and other neurodegenerative conditions that p-tau217 does not capture. A high result does not mean "you will get Alzheimer's". It means Alzheimer's-related biology is present at a level associated with a higher probability of later cognitive problems.

How Accurate Is the Test in Healthy People?

A 2025 study in JAMA Neurology examined p-tau217 in 2,916 cognitively unimpaired people across 12 international cohorts. As a stand-alone test for amyloid pathology, its positive predictive value was 79%, so around one in five positive results did not match amyloid on PET or spinal fluid testing. When positive blood results were confirmed with PET or cerebrospinal fluid testing in a two-step approach, the positive predictive value rose to 91%. How well a test performs depends on how common the condition is in the group being tested, so results from specialist cohorts or memory clinics do not translate directly to millions of symptom-free adults.

Why Not Test Everyone at 60?

Alzheimer's biology can begin years before dementia, so waiting for memory problems can seem illogical. The difficulty is that earlier detection is only clearly useful when there is something to do with the result. Disease-modifying treatments such as lecanemab and donanemab exist for selected people with early symptomatic disease, but neither is currently recommended for routine NHS use in England, with NICE's appraisals still in progress. No treatment is established for cognitively healthy people simply because p-tau217 suggests preclinical Alzheimer's pathology.

Trials are testing whether intervening before symptoms can change the course of the disease. Until they report, screening could leave a healthy person knowing they carry elevated biological risk without being able to say whether dementia will develop, when, or what treatment should follow. The authors note that current Alzheimer's Association guidelines recommend against testing cognitively unimpaired adults outside research studies or clinical trials.

What About the NHS?

The UK is investigating how Alzheimer's blood tests could enter routine care, but the focus is diagnosis rather than population screening. The ADAPT trial aims to recruit around 1,100 people through about 20 NHS memory services. Participants already have cognitive concerns, and researchers are testing whether giving clinicians p-tau217 results earlier improves diagnosis, decisions on further investigation and treatment, and cost-effectiveness.

A second trial, READ-OUT, is evaluating several blood biomarkers for Alzheimer's and other forms of dementia. Both belong to the Blood Biomarker Challenge, a programme designed to build the evidence for using blood tests in NHS dementia services.

What This Actually Means

Blood biomarkers can already give clinically useful information about Alzheimer's pathology, and in cognitively healthy research participants higher p-tau217 was associated with substantially greater later risk of impairment. But finding Alzheimer's-related proteins does not tell anyone with certainty whether or when their cognition will decline.

The headline percentages are not personal predictions. The research population was enriched for Alzheimer's pathology, and the absolute risks need validating in more representative groups. The immediate opportunity is diagnosis, where a blood test could give people who already have cognitive problems a faster, less invasive and potentially more accurate assessment.

Frequently Asked Questions

Can a blood test tell me whether I'll get Alzheimer's?

Not with certainty. Higher p-tau217 is associated with greater future risk of cognitive impairment, but it cannot tell an individual whether they will develop Alzheimer's or when symptoms will begin.

What does the 38% figure mean?

Participants with very high p-tau217 had an estimated 38% absolute risk of developing cognitive impairment within five years. Because the study used selected research cohorts with high levels of amyloid positivity, that figure should not be applied directly to the general population.

Should healthy people get an Alzheimer's blood test?

Routine testing of cognitively healthy adults is not established practice, and Alzheimer's Association guidelines recommend against it outside research studies or clinical trials. Researchers still need better evidence on how to turn results into individual risk and what should follow.

Are Alzheimer's blood tests available on the NHS?

They are not yet in routine use across NHS memory services. The ADAPT and READ-OUT trials are evaluating how blood biomarkers could become part of dementia diagnosis.

The Bottom Line

Alzheimer's blood testing has moved quickly, and p-tau217 can now separate cognitively healthy people with very different future risks of impairment. But an elevated result is neither a dementia diagnosis nor a forecast of when symptoms will start, the risk estimates come from selected research populations, and no treatment is routinely offered to healthy people because a blood biomarker is raised.

The nearest benefit is simpler: helping people who already have cognitive symptoms get a faster and more accurate diagnosis. The larger possibility comes later. We are getting much better at seeing Alzheimer's before symptoms appear, and the open question is whether seeing it earlier will help us prevent what follows.

For the everyday side of brain health — sleep, stress load and the habits that support cognition over decades — The Stress Reset offers a structured place to start, and pairs with the Reset Companion for day-to-day support.

Related reading: How to Improve Brain Health at Any Age · Can Statins Stop Dementia? What We Know · Can Caffeine Reduce Dementia Risk? · Glucosamine and Alzheimer's Progression

This is general information rather than medical advice. If you are concerned about changes in your memory or thinking, speak to your GP.

Popular Articles