Blood Test May Predict Dementia Risk In Women Up To 25 Years Early: But There’s A Racial Divide

A routine blood draw at the doctor’s office might one day flag whether cognitive decline is in store for women, not just soon, but potentially decades from now. In a large new study of nearly 2,800 older women, scientists found that a single protein measured in blood plasma was linked to who would go on to develop dementia or mild cognitive impairment up to 25 years later.

But the study also revealed that the test’s predictive power wasn’t equal across all groups, raising urgent questions about how race, age, genetics, and even hormone therapy might shape who benefits most from this emerging diagnostic tool.

The protein at the center of the research is called p-tau 217, a molecule that builds up in the brains of people developing Alzheimer’s disease. Unlike brain scans or spinal taps, measuring p-tau 217 requires nothing more than a blood sample, making it far cheaper and easier to obtain. Researchers have known for a few years that higher levels of this protein in the blood track with Alzheimer’s-related brain changes. What they hadn’t established was just how far into the future those elevated levels could forecast trouble, or whether the forecast looked different depending on who was being tested.

Published in JAMA Network Open, the study drew from the Women’s Health Initiative Memory Study, a landmark set of clinical trials that enrolled postmenopausal women from 39 sites across the United States between 1996 and 1999. Blood samples collected at the start of the study sat frozen for more than two decades before being thawed and analyzed in 2024. Researchers then linked those protein measurements to detailed cognitive assessments conducted annually for as long as a quarter century, making it one of the longest looks yet at how a single blood measurement connects to real-world cognitive decline.

How the Blood Test for Dementia Study Worked

Researchers examined 2,766 women who were cognitively healthy when they enrolled, with an average age of about 70. Roughly 74 percent identified as White, about 18 percent as Black, and 7 percent as Hispanic, with smaller numbers from other racial and ethnic groups. Because the women were originally part of a hormone therapy trial, about half had been randomly assigned to take some form of hormone replacement while the other half received a placebo.

Blood samples were processed using a high-sensitivity lab test designed to detect even tiny amounts of p-tau 217. Over a follow-up period stretching a median of about 14 years and as long as 25 years, 1,311 women developed mild cognitive impairment or dementia. In separate analyses, 849 developed mild cognitive impairment and 752 developed dementia.

After accounting for factors like age, education, smoking, diabetes, heart disease, physical activity, kidney function, and cholesterol, women with higher levels of p-tau 217 in their blood had about two-and-a-half times the risk of developing mild cognitive impairment or dementia. When dementia was examined on its own, the link was even more prominent: about a threefold increase in risk. Women whose p-tau 217 levels landed in the top quarter had a sevenfold higher risk of dementia compared with those in the bottom quarter.

Race, Age, and Genetics Changed the Blood Test for Dementia Picture

Perhaps the most consequential finding involved the differences across subgroups. Among White women, higher p-tau 217 was strongly linked to both mild cognitive impairment and dementia. Among Black women, the protein still predicted dementia at roughly a 1.9-fold higher risk, but the association was weaker. P-tau 217 showed no statistically meaningful link to mild cognitive impairment in Black women at all.

Researchers offered several possible explanations. Mild cognitive impairment in Black women might more often stem from non-Alzheimer’s causes, such as blood vessel disease in the brain. It’s also possible that the cognitive tests used to diagnose mild cognitive impairment are less accurate in Black populations, a concern that other researchers have raised. When the team combined p-tau 217 with age to predict dementia specifically, the test performed similarly well in both White and Black women.

Age and a well-known genetic risk factor also mattered. Women older than 70 at the start of the study showed stronger connections between p-tau 217 and cognitive outcomes than younger participants. Carriers of a gene variant called APOE ε4, the strongest known genetic risk factor for Alzheimer’s, also showed stronger connections than non-carriers. Both findings make intuitive sense: the blood marker appears to be most predictive in people who already carry the heaviest biological burden of risk.

Hormone Therapy Added a Twist

Because study participants had been randomly assigned to hormone therapy or placebo, removing the usual complications of self-selection, researchers had a rare opportunity to explore whether hormone use changed the relationship between p-tau 217 and dementia. It did, but only for one type of hormone regimen.

Women assigned to a combination of estrogen and progestin showed a stronger link between elevated p-tau 217 and dementia than women on placebo. Each standard-unit increase in the protein was associated with roughly a fourfold increase in dementia risk for the combination hormone group, compared with about a threefold increase in the placebo group. No such difference emerged for women assigned to estrogen alone versus placebo.

In the original trials, the combination of estrogen and progestin had already been linked to double the risk of dementia compared with placebo. The new findings suggest that p-tau 217 may interact with, or at least be amplified by, the effects of certain hormone therapies. Researchers noted that the specific formulations used in the trial may not reflect hormone therapies commonly prescribed today, which often involve lower doses or different delivery methods.

The Future of Dementia Screening

Adding p-tau 217 to basic demographic information like age improved the ability to distinguish women who would go on to develop dementia from those who would not. On its own, the blood marker outperformed demographics alone for predicting dementia. When combined with age, it achieved similar accuracy in both Black and White women for dementia prediction: a meaningful finding given persistent disparities in Alzheimer’s diagnosis and care across racial groups.

Still, the results make clear that no single blood test operates in a vacuum. A protein level that powerfully flags cognitive decline risk in a 75-year-old White woman carrying the APOE ε4 gene variant may carry a very different meaning in a 66-year-old Black woman without that genetic risk. Clinicians hoping to use p-tau 217 as a screening tool will need to weigh these contextual factors carefully, especially as blood-based Alzheimer’s tests begin moving from research laboratories into everyday medical practice.

A blood test for dementia risk that requires nothing more than a routine draw is no longer a distant prospect. Whether it serves every patient equally will depend on how honestly the field confronts its blind spots, particularly around race, sex, and the hormonal histories that millions of women carry into older age.

---

Disclaimer: This article summarizes findings from an observational study and does not establish a direct cause-and-effect relationship. While researchers identified associations between the factors studied and dementia risk, the results cannot rule out other explanations. Individuals concerned about their cognitive health should consult a qualified healthcare professional.

---