225 - Multi-site Quality Improvement Collaborative for Reduction of Empiric Antibiotics Among Infants At-Risk of Early Onset Sepsis

Poster Number: 225
Publication Number: 225.127

Anne-Marie Rick, University of Pittsburgh School of Medicine, Pittsburgh, PA, United States; Hui Liu, university of Pittsburgh, pittsburgh, PA, United States; Neera Goyal, Nemours Children's Hospital, Philadelphia, PA, United States; Neha S. Joshi, Stanford University School of Medicine, Stanford, CA, United States; Virginia A. Moyer, University of North Carolina at Chapel Hill School of Medicine, Chapel Hill, NC, United States; Carl J. Seashore, University of North Carolina at Chapel Hill School of Medicine, Chapel Hill, NC, United States; Arun Gupta, Stanford University School of Medicine, Palo Alto, CA, United States; Gysella Muniz Pujalt, UPMC Childrens Hospital of Pittsburgh, Pittsburgh, PA, United States

Background: Risk stratification for early onset sepsis (EOS) can safely reduce antibiotic use in term infants. Guidelines from the American Academy of Pediatrics (AAP) in 2018 recommend 3 approaches: a multivariate risk calculator, serial exams, or a categorial threshold. Implementation of a new approach presents numerous logistical, clinical, and inter-disciplinary challenges that may impede adoption of recent guidelines.

Objective: We aimed to established a multi-site quality improvement (QI) collaborative to facilitate implementation of new AAP EOS evaluation guidelines and to reduce antibiotics and laboratory evaluations by 30% and 15%, respectively, among chorioamnionitis-exposed newborns ≥35 weeks gestational age (GA).

Design/Methods: We used the Breakthrough Series Collaborative (BTS) Model for Improvement to establish a learning hub of newborn nurseries to develop, implement, and analyze approaches for EOS evaluation. The BTS included a 5-month planning period (September 2020 to January 2021) and a 9-month action period (February 2021 to October 2021). To establish the BTS, newborn nurseries participating in the Better Outcomes for Research in Newborns (BORN) Network were recruited via email and screened for eligibility including willingness to change EOS evaluation and to meet collaborative milestones (monthly webinar participation, establish local multi-disciplinary QI team, implement local Plan, Do, Study, Act cycles, and obtain 4 months of retrospective baseline data and 9 months of prospective data (pregnancy, birth, physical exam, laboratory characteristics) on infants ≥ 35 weeks GA exposed to chorioamnionitis). A multi-site core team facilitated the BTS through webinars, a centralized data platform, data analysis and feedback, and QI tools. Chi-square was used to compare antibiotic and laboratory usage during the baseline and action periods.

Results: Four sites completed the collaborative (Table 1). Over the action period, sites shared and implemented tools obtained through the network including code for electronic health record screening, forms for documentation, and parent engagement tools. During the action period, there was a significant reduction in blood cultures, sepsis laboratory testing, and antibiotic use (Table 2). No true positive blood cultures were identified. No adverse events were noted. Three infants were readmitted within 48 hours of discharge for jaundice or temperature instability but were not treated for EOS.Conclusion(s): BTS collaboratives can be successfully used to facilitate implementation of new AAP EOS evaluation guidelines in a variety of newborn nursery settings.
Table 1. Characteristics of Participating Sites.
Table 2. Comparison of antibiotics and sepsis laboratory evaluations for infants ≥ 35 weeks gestational age exposed to chorioamnionitis during the baseline and action periods across all sites.