Abstracts

Rationale: Clinical documentation of seizure burden in children with epilepsy is heterogeneous in form and content yielding inconsistent analysis of the effects of anti-seizure medications (ASMs) on seizure burden over time. Electronic medical records (EMRs) allow for standardized seizure burden documentation, enhancing the ability to extract and analyze longitudinal patient-centered outcome measures consistently and at a large scale, providing new insights into comparative effectiveness of ASMs on seizures.

Methods: This was a single-center observational study of 237 patients with epilepsy under the age of 20 who were prescribed ASM and had seizure frequency records over multiple clinical encounters. Seizure frequency was extracted from the EMRs and converted from a structured set of common data elements (CDEs) used by participating providers into a scale ranging from 1-7, with higher numbers representing greater seizure frequency. Analysis included cohort comparisons by ASMs, seizure frequency, epilepsy syndrome, and broad epilepsy type (generalized, focal or combined). For each subgroup, those prescribed a specific ASM were compared against those who were prescribed other ASMs.

Results: Seizure frequency reduction over time was more likely for individuals with generalized seizures who were prescribed valproate over other ASMs (OR 1.64, 95% CI 1.09-2.44). While we did not find significant effects of ASMs on most syndromes, seizures were less likely to decrease in those with Childhood Absence Epilepsy (OR 0.34, 95% CI 0.08-1.05) when prescribed lamotrigine. Another important criterion for the effectiveness of ASMs is whether they can sustain seizure freedom (at least 1-2 years since last seizure). Individuals with focal epilepsies were less likely to maintain seizure freedom when prescribed valproate (OR 0.16, 95% CI 0.00-0.96) or clobazam (OR 0.19, 95% CI 0.02-0.72). Conversely, individuals with focal epilepsies were more likely to maintain seizure freedom when prescribed oxcarbazepine (OR 4.68, 95% CI 2.71-8.11). Individuals with generalized and combined epilepsies were more likely to maintain seizure freedom from lamotrigine and levetiracetam, respectively (generalized: OR 2.30, 95% CI 0.90-5.34; combined: OR 4.68, 95% CI 1.75-11.80). When our analysis broadened to include those maintaining seizure freedom or seizure frequency reduction, results were similar for focal epilepsies; if oxcarbazepine was prescribed, then reduction in seizure frequencies or maintaining seizure freedom was more likely (OR 2.21, 95% CI 1.54-2.91).

Conclusions: Using information from common data elements embedded in the EMR, we were able to determine distinct patterns of seizure burden and medication response by epilepsy type. This scalable framework based on routine and real-world care provides for a novel learning healthcare system approach for pediatric epilepsy.

Funding: Please list any funding that was received in support of this abstract.: Children’s Hospital of Philadelphia (CHOP), National Institute of Neurological Disorders and Stroke (NINDS), National Institute of Child Health and Human Development (NICHD), The Hartwell Foundation.