151 - Variation and Establishment of Achievable Benchmarks of Care in Neurodiagnostic Testing Among Children with Simple Febrile Seizures

Poster Number: 151
Publication Number: 151.416

John R. Stephens, University of North Carolina at Chapel Hill School of Medicine, Chapel Hill, NC, United States; Matthew Hall, Children's Hospital Association, Lenexa, Kansas, KS, United States; Matthew J. Molloy, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States; Jessica L. Markham, Children's Mercy Hospitals and Clinics, Kansas City, MO, United States; Jillian M. Cotter, University of Colorado School of Medicine, Aurora, CO, United States; Michael Tchou, Children's Hospital Colorado, Aurora, CO, United States; Paul L. Aronson, Yale School of Medicine, New Haven, CT, United States; Michael J. Steiner, University of North Carolina, chapel hill, NC, United States; Elisha McCoy, Le Bonheur Children's Hospital, Memphis, TN, United States; Megan E. Collins, Children's Mercy Hospitals and Clinics, Mission, KS, United States; Samir S. Shah, Journal of Hospital Medicine, Cincinnati, OH, United States

Background: Current guidelines recommend against routine neurodiagnostic testing for evaluation of simple febrile seizures. The degree of variation in neurodiagnostic testing across children's hospitals is unclear and Achievable Benchmarks of Care (ABCs) in the evaluation of simple febrile seizures have not been established.

Objective: Our aims were: 1) assess variation in rates of neurodiagnostic testing among children evaluated for simple febrile seizures at children’s hospitals and 2) establish ABCs for the evaluation of simple febrile seizures.

Design/Methods: We conducted a cross-sectional study of children 6 months to 5 years with simple febrile seizures evaluated in the emergency departments (EDs) at 38 children’s hospitals in the Pediatric Health Information System database between 2016-2019. We excluded children with epilepsy, complex febrile seizures, complex chronic conditions, and those receiving intensive care. Outcomes included proportions of children who received neuroimaging, electroencephalogram (EEG), or lumbar puncture (LP), costs, and rates of hospitalization, for the study cohort and individual hospitals. Hospital-specific outcomes were adjusted for patient demographics and severity of illness. For each measure we then utilized hospital-specific values to calculate ABCs, defined as the best (lowest) rate of performance for each outcome, using the number of hospitals that would include 10% of the study cohort.

Results: We identified 51,015 encounters. Among the study cohort 821 (1.6%) children had neuroimaging, 554 (1.1%) EEG, 314 (0.6%) LP and 2,023 (4.0%) were hospitalized. Neurodiagnostic testing rates varied across hospitals (Figure 1): EEG use ranged from 0-8.2%, neuroimaging 0.4-6.7%, and LP 0-12.7% in patients < 1 year old and 0-3.1% in patients >1 year old. Hospitalization rate ranged from 0-14% (Figure 2). Measured outcomes were higher among hospitalized vs. ED-only patients: neuroimaging 15.3% vs. 1.0%, EEG% 24.7 vs. 0.1%, LP 9.6% vs. 0.2% and cost $3096.6 vs. $465.1 (all pConclusion(s): Rates of neurodiagnostic testing and hospitalization for simple febrile seizures were low but varied substantially across hospitals. Rate of hospitalization also varied across hospitals and rates of neurodiagnostic testing were significantly higher among hospitalized patients. Calculated ABCs were 0-1% for all measures, demonstrating that adherence to current guidelines is an attainable goal.
Figure 1Heat map of rates of neurodiagnostic test use for all encounters by individual hospital, arranged by highest (hospital 1) to lowest (hospital 38) adjusted admission rate. Colors correspond to levels of use, ranging from red (highest rate) to dark green (lowest rate).
Abbreviations: ED – emergency department; EEG – electroencephalogram; LP – lumbar puncture
Figure 2Risk adjusted admission rates for patients with simple febrile seizures, adjusted for age, race, gender, payor and HRISK score. Upper and lower control limits are displayed for each value.
Abbreviation: HRISK - Hospitalization Resource Intensity Scores for Kids