LifeAfter90 Finds Dementia Risk Patterns After Age 90

Key Takeaways

• Among U.S. Kaiser Permanente members aged 90 and older without prevalent dementia, women had about a two-fold higher dementia risk than men.
• Black participants had a 75% higher dementia risk than Asian participants.
• APOE ε2 remained associated with lower dementia risk after age 90, with investigators reporting an approximately 60% lower risk.
• APOE ε4 was not significantly associated with higher dementia risk overall, although investigators said its association with dementia might differ by sex.

Adults who live into their 90s account for a growing share of dementia care, yet incident dementia in this age range is less well characterized than risk patterns earlier in late life. Clinicians still have limited evidence on whether differences by sex, race and ethnicity, and genetic background persist after many competing health risks have already shaped survival into very late life. In that setting, investigators used the U.S. LifeAfter90 cohort to follow cognition prospectively before dementia developed.

The LifeAfter90 prospective cohort study in The Lancet Healthy Longevity enrolled Kaiser Permanente Northern California members aged 90 or older from the San Francisco Bay Area and Sacramento. This analysis excluded participants with prevalent dementia and followed participants with clinical evaluations every six months from July 2018 through November 2024. This analysis included 805 participants with a median age of 92 years. Participants underwent clinical evaluations every six months, either in person or remotely. Incident all-cause dementia was diagnosed using a combination of physician assessment, the Clinical Dementia Rating, and a Functional Activities Questionnaire. The cohort was racially and ethnically heterogeneous, with 26% Asian, 24% Black, 20% Hispanic or Latinx, 28% White, and 2% from other racial or ethnic groups.

Women had a nearly two-fold higher dementia risk than men (sHR 1.89, 95% CI 1.30–2.76). Black participants had a 75% higher dementia risk than Asian participants (sHR 1.75, 95% CI 1.07–2.88).

In the cohort's apolipoprotein E (APOE) genotype findings after age 90, APOE ε2 carriers had an approximately 60% lower dementia risk than non-carriers (sHR 0.39, 95% CI 0.17–0.88). APOE ε4 carriers did not have a statistically significant higher dementia risk than non-carriers overall (sHR 1.51, 95% CI 0.92–2.47), although investigators said the association between APOE ε4 and dementia might differ by sex.

During a mean follow-up of 2 years, 138 participants (17%) developed dementia and 295 (37%) died. The age-standardized dementia incidence rate was 116.82 cases per 1,000 person-years (95% CI 93.69–139.96). Investigators used age-adjusted Cox and Fine–Gray competing-risk models, with death treated as a competing risk in the Fine–Gray analyses.

According to the investigators, LifeAfter90 helps address a longstanding gap in incident dementia data after age 90 in a racially and ethnically heterogeneous U.S. oldest-old cohort. The reported pattern suggests that dementia risk in very late life was not evenly distributed across sex, race and ethnicity, and APOE-defined groups.

Clinician Questions

Why is the LifeAfter90 cohort unusual for dementia research after age 90?

LifeAfter90 prospectively studied dementia incidence and risk factors among adults aged 90 years and older in a racially and ethnically heterogeneous population, an age range and population for which the investigators noted that relatively little is known about dementia incidence and its risk factors.

What did investigators find about APOE after age 90?

APOE ε2 carriers had a significantly lower dementia risk than non-carriers (sHR 0.39, 95% CI 0.17–0.88). APOE ε4 carriers did not have a statistically significant higher risk overall (sHR 1.51, 95% CI 0.92–2.47), although investigators said the association between APOE ε4 and dementia might differ by sex.

Do the LifeAfter90 dementia findings apply to younger older adults or only to adults 90 and older?

The reported associations came from U.S. Kaiser Permanente Northern California members aged 90 years and older who were free of prevalent dementia in this analysis, so the findings directly describe this oldest-old population. The study did not establish that the same associations apply to younger older-adult groups.