Study reveals gender, racial disparities in dementia risk for people over 90

For decades, researchers have assumed that cognitive decline after age 90 was simply inevitable—a natural consequence of extreme longevity. But until very recently, they had almost no hard data to prove it. The knowledge gap mattered less when ninety-year-olds were rare. By 2100, that will no longer be true. Demographers project roughly 230 million people worldwide will be 90 or older, and doctors, scientists, and policymakers are scrambling to understand how to care for them.

A team at UC Davis Health and Kaiser Permanente has begun filling that void. Since 2018, they have been running the LifeAfter90 study, tracking a large cohort of people 90 and older who enrolled before showing any signs of dementia. Every six months, researchers examined these patients to monitor their cognitive health. Because the participants were long-term Kaiser Permanente members, the team had access to decades of medical history—some records stretching back to the 1960s. When the researchers published their latest findings in The Lancet Healthy Longevity, they had examined medical records from over 800 participants with a median age of 92. This was the first major effort to investigate dementia patterns in a highly diverse group of people in their tenth decade.

What they found echoed patterns seen in younger populations, but with new urgency. Women 90 and older faced a dementia risk twice as high as men. Racial and ethnic disparities persisted too: Black participants showed a 75 percent greater risk than Asian participants. Hispanic participants also had significantly higher incidence rates than white and Asian participants. "It is striking that the racial and ethnic disparities in dementia risk observed in younger adults continue into the tenth decade of life," said Hilary Colbeth, a UC Davis postdoctoral scholar and first author on the paper. Rachel Whitmer, the study's senior author and a UC Davis professor of public health sciences and neurology, noted that researchers had long known women in their 60s and 70s faced higher dementia risk, but no one had confirmed whether that pattern held after 90.

The study also examined how a single gene shaped dementia risk in this age group. The APOE gene has long been linked to Alzheimer's disease. One variant, APOE2, acts as a shield, lowering the risk of developing Alzheimer's. Another variant, APOE4, does the opposite. The researchers found that APOE2's protective effects remained powerful past age 90, reducing dementia risk by 60 percent. APOE4 told a more complicated story. For the study group as a whole, it did not substantially increase dementia incidence. But when the researchers broke down the data by gender and ethnicity, a different picture emerged. APOE4 increased risk among men. More strikingly, it essentially doubled the risk among Black participants.

These findings raise as many questions as they answer. Some participants had high blood pressure, high cholesterol, and other known dementia risk factors in their 60s and 70s, yet remained cognitively sharp into their 90s. The same was true for some people carrying the APOE4 variant. The researchers want to understand what protected these individuals—what biological or behavioral factors allowed them to escape the expected trajectory. If those mechanisms could be identified and replicated, they might offer a path toward prevention.

For now, Whitmer emphasized a more immediate lesson for clinicians. "Doctors need to know that certain groups are at higher or lower risk," she said. "We can't just assume that someone from a high-risk group makes it to 90 without dementia and they're in the clear. We need to talk about risk reduction for everyone." The findings suggest that reaching 90 without dementia is not a guarantee of safety. It is a moment to intensify attention, not to relax it. As the population of the very old grows, that distinction could reshape how medicine approaches the final decades of life.