Each 10-year increase in sleep EEG-based brain age index was associated with a 39% higher risk of incident dementia across five community-based cohorts, highlighting its potential as a noninvasive digital marker for early dementia detection.
Each 10-year increase in sleep EEG-based Brain Age Index (BAI) was associated with a 39% higher risk of incident dementia after adjustment for covariates.
The association between BAI and incident dementia remained significant after additional adjustment for comorbidities and apnea-hypopnea index scores.
The association between BAI and incident dementia remained significant after further adjustment for apolipoprotein E ε4 (APOE ε4) genotype.
The study pooled individual participant data from five community-based longitudinal cohorts spanning diverse populations with varying follow-up durations.
The Brain Age Index was computed using interpretable machine learning incorporating multidimensional sleep EEG features, measuring deviation between EEG-estimated brain age and chronological age.
The association between higher sleep EEG BAI and incident dementia was consistent across sex and age groups.
This research suggests that the way our brains age — as measured during sleep — can predict whether a person will develop dementia years or even decades later. Scientists used a machine learning tool called the Brain Age Index (BAI), which analyzes the electrical activity of the brain during sleep (recorded by an EEG) and estimates how 'old' the brain appears compared to a person's actual age. A higher BAI means the brain looks older than it should. Analyzing data from over 7,000 adults across five large U.S. community studies, researchers found that for every 10-year increase in BAI, the risk of developing dementia increased by about 39%. This association held up even after accounting for factors like other health conditions, sleep apnea severity, and a known genetic risk factor for Alzheimer's disease (APOE ε4). This research suggests that routine sleep studies, which are already used to diagnose conditions like sleep apnea, could potentially provide additional valuable information about brain health and future dementia risk — without any invasive procedures or specialized brain scans. The finding was consistent across men and women and across different age groups, strengthening confidence in the result. The cohorts ranged widely in age and follow-up time, with some participants developing dementia as many as 17 years after their sleep study. This research suggests that sleep EEG-based brain age could serve as an early, noninvasive marker for detecting dementia risk in the general population. This could be particularly important because dementia develops slowly over many years, and earlier detection might open windows for intervention. However, the authors note that further research is needed to confirm whether the BAI can be practically used as a screening or predictive tool in clinical or community settings.
Sun H, Milton S, Fang Y, Taha H, Shiju S, Thomas R, et al.. (2026). Machine Learning-Based Sleep Electroencephalographic Brain Age Index and Dementia Risk: An Individual Participant Data Meta-Analysis.. JAMA network open. https://doi.org/10.1001/jamanetworkopen.2026.1521