JAMA Neurol. 2026 Oct 19. doi: 10.1001/jamaneurol.2026.3708. Online ahead of print.
ABSTRACT
IMPORTANCE: Late-onset epilepsy (LOE; onset at age ≥65 years) is increasingly recognized in older adults and may be associated with vascular risk.
OBJECTIVE: To identify vascular risk factors (VRFs) associated with incident LOE in older adults.
DESIGN, SETTING, AND PARTICIPANTS: This study is an individual participant data meta-analysis of 6 US population-based prospective cohort studies (the Epilepsy-Cog Study). In 5 cohorts linked to Medicare for epilepsy case ascertainment, follow-up for LOE began when a participant had accrued 2 years of continuous fee-for-service enrollment and was 65 years of age or older and extended to the end of continuous enrollment. Claims linkage covered 1991-2018 in the Atherosclerosis Risk in Communities Study (ARIC), 1992-2019 in the Cardiovascular Health Study (CHS), 2000-2019 in the Multi-Ethnic Study of Atherosclerosis (MESA), 2000-2020 in the Reasons for Geographic and Racial Differences in Stroke Study (REGARDS), and 1999-2019 in the Washington Heights/Inwood Columbia Aging Project (WHICAP). In the Northern Manhattan Study (NOMAS), which was not Medicare linked and used cohort-based epilepsy case ascertainment, follow-up for LOE began at cohort entry in 1993-2008 and extended to 2024. Data analysis was completed in January 2026.
EXPOSURES: Sociodemographic characteristics and VRFs assessed by self-report, examinations, laboratory measures, and medication use.
MAIN OUTCOME AND MEASURES: The primary outcome was incident LOE. Associations between risk factors and incident LOE were evaluated using multivariable Cox proportional hazards models to estimate hazard ratios.
RESULTS: From 43 753 participants in Epilepsy-Cog, 551 participants with prevalent epilepsy and 945 with no follow-up after Epilepsy-Cog baseline were excluded. Among 42 257 participants (mean [SD] age, 71.4 [6.9] years; 23 383 [56.0%] female) followed up for a mean (SD) of 9.3 (6.2) years, 1500 participants (3.5%) developed LOE (3.8 per 1000 person-years). In fully adjusted models, LOE risk was higher with current smoking (HR = 1.39; 95% CI, 1.18-1.64), hypertension (HR = 1.17; 95% CI, 1.04-1.31), diabetes (HR = 1.42; 95% CI, 1.25-1.62), chronic kidney disease (HR = 1.29; 95% CI, 1.13-1.48), history of stroke (HR = 1.99; 95% CI, 1.63-2.43), and apolipoprotein E ε4 allele (1 ε4 allele: HR = 1.22; 95% CI, 1.07-1.39; 2 ε4 alleles: HR = 1.84; 95% CI, 1.36-2.50). Associations of VRFs with LOE were similar across age, sex, and race and ethnicity subgroups.
CONCLUSIONS AND RELEVANCE: In this pooled analysis of 6 US cohorts, multiple VRFs and a history of stroke were associated with incident LOE. These findings suggest that LOE may reflect the underlying burden of brain vascular disease.
PMID:42895730 | DOI:10.1001/jamaneurol.2026.3708