P3D321: Neonatal Fever Clinical Pathway in the Emergency Department: Quality Improvement Initiative to Optimize Early Identification of High-risk Infants and Patient-related Outcomes

Purpose/Objectives: Fever in infants 0-60 days old (neonatal fever) is often of viral etiology but may also be the presenting symptom for serious bacterial infections (SBI). This dichotomy can lead to variability in clinical practice, misidentification of SBI, and unnecessary interventions. The American Academy of Pediatrics (AAP) has recently published guidelines for the management of neonatal fever. Implementation of those guidelines may improve patient-related outcomes. The purpose of this quality improvement project is to increase appropriate identification of neonatal fever and infants at high risk of SBI and decrease unnecessary lumbar punctures among infants presenting to the emergency department through adherence to the AAP guidelines. Specifically, we aimed to have 90% or greater adherence to performing an appropriately indicated lumbar puncture and administer appropriate antibiotics for infants less than 60 days based on risk stratification.

Design/Methods: In coordination with the Pediatric Infectious Disease Department and Emergency Department (ED) leadership, two algorithms were created to identify, evaluate, and treat at-risk infants. Algorithm 1 addressed infants 0-28 days of age and Algorithm 2 addressed infants 29-60 days of age. An order set was created that provided recommended diagnostic evaluation and antibiotics. Algorithms were placed in the ED and the Pediatric Resident Workroom. A review of the algorithm and order set was discussed in-person at an interdepartmental meeting, and details were emailed to residents and staff.
Outcome measures are the percentages of correctly identified at-risk infants and appropriately deferred lumbar punctures. Process measures are proper antibiotic use, utilization rate of the order set, and rate of algorithm-congruent diagnostic evaluation. Balancing measures are hospitalizations, patient mortality, and patient morbidity.

Results: Infants (0-60 days) evaluated in the ED for fever, hypothermia, ill appearance, or abnormal assessment based on the Pediatric Assessment Triangle were reviewed. Based on preliminary data, proper identification of at-risk infants increased from 79% to 91%. Specifically, correct classification of low-risk infants increased from 88% to 100%. Unnecessary antimicrobial usage decreased from 20% to 12.5%. Appropriate lumbar puncture deferral increased from 95% to 100%. Algorithm utilization was greater than 90%, although order-set usage was at 18%.

Conclusion/Discussion: Preliminary data shows that the protocolized algorithm and order set have increased correct identification of at-risk infants while reducing unnecessary invasive procedures and antibiotic administration. This evidence demonstrates how interdepartmental collaboration to develop a standardized approach can improve patient-related outcomes.