Late-onset epilepsy (LOE) is emerging as an increasingly important public health concern as the U.S. population ages, with the highest rates of new cases typically occurring after age 60. Yet despite its incidence among older adults, it has historically received less epidemiological study than other more common neurological disorders of aging, such as dementia or stroke, and is often difficult to diagnose because symptoms can easily be mistaken for other disorders such as syncope, delirium, dementia, or memory changes.
To address longstanding research gaps in the field, Hyunmi Choi, M.D., a neurologist specializing in epilepsy at NewYork-Presbyterian and Columbia, is leading the Epilepsy-Cog study, which pools decades of data from six U.S. longitudinal cardiovascular studies to examine both the factors that contribute to late-onset epilepsy and its long-term health impact.
In a methods paper published in Epilepsia, the investigators describe Epilepsy-Cog as the first large-scale multicohort observational study of epilepsy in older adults. Funded by the National Institute on Aging, the five-year project is now in its final year, and the goal is for it to be used as a resource for yielding future insights into the disease.
“The epidemiological evidence has been sparse because there haven’t been enough population-based studies that have focused entirely on late-onset epilepsy,” says Dr. Choi. “We want to look at its vascular risk factors and the consequences of LOE in terms of stroke, dementia, and cognitive decline.”
The epidemiological evidence has been sparse because there haven’t been enough population-based studies that have focused entirely on late-onset epilepsy. We want to look at its vascular risk factors and the consequences of LOE in terms of stroke, dementia, and cognitive decline.
— Dr. Hyunmi Choi
Establishing the Multicohort Study
The six well-known population studies Epilepsy-Cog draws on were originally created to investigate cardiovascular disease, stroke, aging, and cognition. Rather than establishing a new cohort, investigators applied validated ascertainment methods to the participant data to determine prevalent and incident epilepsy cases. The six parent studies include:
- The Atherosclerosis Risk in Communities Study (ARIC)
- Cardiovascular Health Study (CHS)
- Multi-Ethnic Study of Atherosclerosis (MESA)
- Northern Manhattan Study (NOMAS)
- Reasons for Geographic and Racial Differences in Stroke (REGARDS) study
- Washington Heights/Inwood Columbia Aging Project (WHICAP).
NOMAS and WHICAP are Columbia-based studies of northern Manhattan residents.
Of 68,544 participants across the six cohorts, 43,753 met Epilepsy-Cog eligibility criteria; 551 (1.3%) had epilepsy at baseline and 1,500 (3.4%) developed it during follow-up. In five Medicare-linked cohorts, researchers used validated algorithms incorporating International Classification of Diseases (ICD) diagnosis codes and, in some cases, antiseizure medication records to identify epilepsy cases. NOMAS was not linked to Medicare but epilepsy cases were identified through participant follow-up data and medical records, then confirmed by a study neurologist.
A major focus of the project has been harmonizing the data collected across the six cohorts by standardized variables related to demographics, health behaviors, medication use, cardiovascular events, stroke, mortality, and risk factors such as hypertension and diabetes, establishing common definitions across the cohorts.
The resulting dataset allows investigators to examine long-term associations between late-onset epilepsy, cognitive decline, vascular factors, and other health outcomes that would be difficult to evaluate within a single cohort.
The Connection Between Vascular Disease and Late-Onset Epilepsy
An early finding by the researchers from one pooled Epilepsy-Cog analysis recently published in JAMA Neurology was the identification of chronic kidney disease (CKD) as a previously unrecognized risk factor for late-onset epilepsy.
“CKD is both a marker of and a risk factor for vascular disease, so it isn’t surprising to us that it is also a risk factor for late-onset epilepsy,” Dr. Choi says. “It’s almost like having hypertension or having diabetes, which are other known risk factors for LOE.”
I envision a time when neurologists and epileptologists would look at epilepsy not just as a seizure disorder that occurs in older people but as more of a potential risk enhancer or risk marker for underlying vascular disease.
— Dr. Hyunmi Choi
The team’s broader hypothesis is that late-onset epilepsy may emerge along a continuum of vascular disease. Underlying vascular changes could develop silently for years before epilepsy appears, with stroke or cognitive decline emerging later along that subclinical trajectory.
Epilepsy-Cog researchers are now examining whether vascular risk factors and MRI evidence of covert vascular brain injury can help explain cognitive decline after epilepsy develops. They also plan to test whether incorporating LOE into established risk models improves prediction of stroke and Alzheimer’s disease.
If the vascular continuum hypothesis is confirmed, a diagnosis of late-onset epilepsy could serve as a clinical signal to evaluate patients more broadly for vascular disease. In addition to seizure management, physicians may have an opportunity to identify vascular risk factors that are unrecognized or undertreated and work across specialties to intervene earlier. Dr. Choi hopes that approach could help delay subsequent complications like stroke and cognitive decline.
Says Dr. Choi, “I envision a time when neurologists and epileptologists would look at epilepsy not just as a seizure disorder that occurs in older people — where we focus primarily on picking seizure medications and managing their seizures — but as more of a potential risk enhancer or risk marker for underlying vascular disease.”
