Abstracts

Rationale: Epidemiologic data are essential for the evaluation of the community burden of epilepsy and for improving quality and access to epilepsy care. This study estimates prevalence and incidence of pediatric epilepsy, measures neurological comorbidities, and determines risk factors for newly-diagnosed epilepsy. Methods: We performed a retrospective cohort analysis using commercial claims and encounters data from Truven MarketScan?(R) database between January 2009 and December 2013, including >1.5 million subjects aged 0-19 years in Georgia. Prevalent epilepsy cases were defined using ICD-9-CM codes occurrences among subjects having: (a) ?-2 codes 345.xx, (b) ?-1 code 345.xx and ?-1 code 780.39, (c) 1 code 345.xx and seizure medication claims, or (d) ?-2 codes 780.39 and seizure medication claims. Incident cases were determined among prevalent cases with no epilepsy or convulsion codes, and no seizure medication claims for at least 2 years and up to 4 years before epilepsy diagnosis. Based on coded diagnoses, 11 neurological comorbid conditions and 9 putative risk factors for epilepsy were identified. Pre-existing risk factors were identified if the conditions of interest occurred either before or < 90 days after epilepsy incidence date. Multiple logistic regression analysis estimated adjusted odds ratio of comorbidities, controlling for age, gender, and metropolitan statistical areas. Logit models determined pre-existing risk factors of developing epilepsy, controlling for age, metropolitan statistical areas, and neurological comorbidities. Results: We identified 9,863 prevalent cases of epilepsy between 2009 and 2013, among which were 1,239 incident cases between 2011 and 2013. For 2013, the estimated prevalence and incidence were 8.02 per 1,000 enrollees (95% C.I. 7.69-8.35) and 137 per 100,000 (124-151), respectively. The estimated epilepsy prevalence increased by age. The estimated incidence of epilepsy was highest in ages 0-4 years. The epilepsy prevalence and incidence were higher in non-metropolitan/rural areas and metro areas with < 1 million of population, compared to metro areas with ?-1 million of population. The most common neurological comorbidities were headache, hyperkinetic syndrome of childhood, and specific delay in development (Figure 1). Children with epilepsy were more likely to have multiple comorbidities compared to those without epilepsy. Those with malignant brain tumor, childhood static encephalopathy, central nervous system infection, and traumatic brain injury had higher risk of developing epilepsy (Figure 2), after controlling for age, metropolitan statistical areas, and neurological comorbidities. Conclusions: The risk of developing epilepsy is highest in the youngest ages, and those with epilepsy have a larger burden of neurological comorbid conditions than the general population. The most significant risk factors for newly diagnosed epilepsy are malignant brain tumor, childhood static encephalopathy, central nervous system infection, and traumatic brain injury. Funding: Children's Healthcare of Atlanta research grant funded by the Goizueta Foundation.