IPM Take
Late-onset epilepsy may be one sign of a wider burden of brain and vascular disease. The new cohort analysis links epilepsy incidence with smoking, hypertension, diabetes, kidney disease and prior stroke. The clinical implication is a fuller assessment, not a claim that controlling one factor has been proven to prevent seizures.
Executive Summary
An individual-participant analysis combined six prospective US cohorts comprising 42,257 older adults. Over an average of 9.3 years, 1,500 developed late-onset epilepsy. In fully adjusted models, incidence was higher with current smoking, hypertension, diabetes, chronic kidney disease and previous stroke; the association with prior stroke was strongest among these clinical factors, with a hazard ratio of 1.99. APOE ε4 carriage was also associated with higher incidence. These are population associations and cannot identify the cause of a particular person’s seizures.
Why it matters
- Patients: A new seizure in later life should prompt investigation of potentially relevant vascular conditions as well as epilepsy.
- Clinicians: Neurological work-up can be connected with review of stroke history, blood pressure, diabetes and kidney health.
- Public authorities: The overlap between chronic-disease and brain-health services supports more integrated care planning.
- Researchers: Intervention studies are needed to determine whether improving vascular risk changes epilepsy incidence.
The pooled analysis in JAMA Neurology followed older adults who did not have epilepsy at baseline, allowing researchers to examine factors associated with new diagnoses over time. Previous stroke was associated with nearly twice the incidence of late-onset epilepsy. Current smoking, hypertension, diabetes and chronic kidney disease also showed positive associations after adjustment.
The findings make biological and clinical sense without establishing a single explanation. Vascular injury can affect the brain in several ways, while older patients may also have neurodegenerative disease and other contributors to seizures. The APOE ε4 association adds another potential connection, but it does not make genetic testing a routine answer to a new seizure.
Cohort data and health-care records can reveal patterns at a scale unavailable to one clinic, yet observational analysis cannot show that changing these exposures prevents epilepsy. The useful near-term response is to connect seizure assessment with a review of brain imaging, vascular history and cardiometabolic health. Prevention claims require studies that directly test whether modifying these risks changes future seizure outcomes.

