loading
loading
NIH grant 1U01DP006929-01: CEPIE. Collaboration to Estimate the Pediatric Incidence of Epilepsy
Institute: {'code': 'DP', 'abbreviation': 'NCCDPHP', 'name': 'National Center for Chronic Disease Prev and Health Promo', 'admin_org_id': '12501086', 'admin_funding_url': ''}. PI: Zachary Michael Grinspan. Award: $499756.
Start: 2026-09-30T00:00:00.
PROJECT SUMMARY / ABSTRACT
The epilepsies are a group of neurologic disorders characterized by unprovoked seizures that occur roughly in
six per thousand children. Physical, neurological, and psychiatric comorbidities are common. One-third of
children with epilepsy have seizures that continue despite medication. The most recent incidence estimates in
the United States are from almost a decade ago. In the interim, evidence has accumulated that there are
disparities in incidence, including our novel preliminary analysis suggesting an increased incidence of early-life
epilepsy among children who are socially vulnerable (i.e., racial minority, low income, low family education). In
addition, increasing attention is now focused on the possibility of epilepsy prevention; thus, the identification of
high-risk groups would lay the foundation for clinical trials. We propose to conduct three linked sets of activities
to investigate the incidence of epilepsy in children, disparities in incidence, and risk factors associated with
incidence. We will use eight data sets to do this, seven of which are in hand. These include nationally
representative survey data, Medicaid claims, electronic health record data, and the text of physician notes. Within
each theme, we will conduct rapid, extended, and exploratory activities. INCIDENCE. First, we will rapidly
calculate the incidence of four epilepsies with well-established case ascertainment algorithms (any epilepsy,
refractory epilepsy, infantile spasms, and neonatal seizures). Second, we will estimate the incidence of 13
epilepsies with ICD10 codes. Third, we will use large language models to develop a novel approach to epilepsy
case ascertainment. DISPARITIES. First, we will confirm or refute our preliminary analyses suggesting a higher
incidence of early-life epilepsy in socially vulnerable children. Second, we will assess the effect of a series of
social determinants of health on epilepsy incidence. Third, we will examine outlier geographies (i.e., places with
very high or very low epilepsy incidence) and interview physicians to develop hypotheses about novel risk factors
or protective factors. RISK FACTORS. First, we will assess the association of pre-incident child factors with
epilepsy incidence. Second, we will assess maternal factors to predict epilepsies in offspring in datasets with
maternal-child dyads. We will also test the hypothesis that the effect of social vulnerability on epilepsy incidence
is mediated through healthcare delivery factors (i.e., prenatal care). Finally, we will characterize the predictability
of epilepsy from pre-existing risk factors, using machine learning algorithms as benchmarks and variable
selection (“shrinkage”) models to create interpretable models. We will conduct these analyses in collaboration
with academic medical centers (Weill Cornell Medicine and University of San Diego California) and a commercial
analytics team (Mathematica).document fetched 2026-10-24 02:47