P4A224: Variation in Resource Utilization and Outcomes of Complex Febrile Seizures from 2009-2021 at Children's Hospitals

Background: Resource utilization for simple febrile seizures has decreased after the publication of guidelines (1); however, less is known about complex febrile seizures, for which no guidelines exist. The objective was to describe resource utilization and outcomes for complex febrile seizure encounters at children’s hospitals.

Methods: A retrospective cohort study was performed using the Pediatric Health Information System (PHIS) database. Children aged 6 months to 6 years were included if they presented to the emergency department for a first encounter with the diagnosis of a complex febrile seizure between Jan 1, 2009 and Dec 31, 2021. Exclusion criteria included: serious diagnoses at time of presentation (neoplasm, trauma or post-traumatic seizure, meningitis, or stroke), codiagnosis of simple febrile seizure, history of epilepsy or developmental delay, presence of a complex chronic condition (2), and hospitals having < 10 patients meeting study criteria in two or more years.

Data were compared between three time periods: (1) Jan 1, 2009 to Jan 31, 2011; (2) Feb 1, 2011 to Jan 31, 2016; (3) Feb 1, 2016 to Dec 31, 2021. The primary outcome was use of medical resources, including admission, PICU admission, obtainment of electroencephalogram (EEG), brain imaging, laboratory studies, and administration of antiepileptics and antimicrobials. Demographic and outcomes data were collected including length of stay (LOS), diagnosis with serious condition, and representation within 7 days.

Chi-square tests and generalized linear mixed models with a random intercept for sites were used to compare demographics, utilization practices, and outcomes between the time periods. P ≤ 0.05 was considered statistically significant. A ≥ 10% difference was considered clinically meaningful.

Results: The analysis included 17,865 patients from 36 hospitals. Across time, there were differences in ethnicity, patients from rural areas, and PICU admissions. The percentages of children admitted, received neuroimaging, any labs, lumbar puncture, or EEG decreased across the time periods except for drug screens and HSV testing, which increased. Fewer children received antiepileptics, antibiotics, and acyclovir across the time periods. Median LOS decreased from 1.6 days in period 1 to 1.3 days in period 3 (Table 1). Representation within 7 days and diagnosis with a new serious medical condition after presentation remained low.

Conclusion: Resource use for complex febrile seizures decreased across most domains since 2009 while rates of diagnosis of an underlying serious medical condition remained low. These findings highlight a potential area for management stewardship.