Lifetime risk of incident dementia and incident mild cognitive impairment in older Black and White adults

Background and Objectives Estimating the lifetime risk of dementia and mild cognitive impairment (MCI) is critical for understanding population burden and informing prevention strategies. This study aimed to estimate the lifetime risk of incident dementia and MCI from age 55 to 105, accounting for competing risk of death and delayed study entry, and to examine differences by sex, race, and Latino ethnicity. Methods We analyzed data from five harmonized community-based cohort studies at the Rush Alzheimer's Disease Center, including 4,611 older adults (mean age 76.6 years) for dementia risk estimation and 3,915 for MCI. Diagnoses were based on annual clinical evaluations. Lifetime risk was estimated using nonparametric cumulative incidence function curves with age as the time scale, adjusting for left truncation and competing risk of death. Analyses were stratified by sex and race, with exploratory analyses by Latino ethnicity, socioeconomic status, and vascular morbidity. Cox models were used to evaluate baseline risk factor associations. Results The lifetime risk of incident dementia from age 55 was 43% (95% CI: 38-47%), with a median diagnosis age of 88 years (IQR: 83-92). For MCI, the risk was 62% (95% CI: 57-67%), with a median diagnosis age of 86 (IQR: 80-90). Females had higher lifetime risks than males for both dementia (45% vs. 39%) and MCI (63% vs. 60%). Racial differences were modest: dementia risk was 45% in Black participants vs. 44% in White participants; Black participants had higher MCI risk through age 95 (41% vs 39%). Latino participants showed higher lifetime risks for both outcomes, these results warrant further replications due to small sample size and younger baseline ages. Stroke history was the strongest baseline predictor of elevated lifetime risk. Discussion These findings provide detailed age-stratified lifetime risk estimates for dementia and MCI in a well-characterized, predominantly Black and White U.S. cohort, extending beyond age 90. By incorporating competing risks and delayed entry, this study offers robust estimates that can inform individualized risk communication and support population-level cognitive aging planning. Exploratory analyses highlight the need for continued enrollment and follow-up of Latino participants to improve equity in risk prediction.