202 - Decreasing Time from IV Antibiotic Order to Administration for NICU Patients with Suspected Early Onset Sepsis

Poster Number: 202
Publication Number: 202.44

Michelle Gontasz, Johns Hopkins University School of Medicine, Cockeysville, MD, United States; Giancola Katharine, Johns Hopkins Bayview Medical Center, Baltimore, MD, United States; Caroline Liang, Johns Hopkins Bayview Medical Center, Baltimore, MD, United States

Background: Morbidity and mortality for newborns with early onset sepsis (EOS) in the first 3 days of life remain substantial and delayed antibiotic administration can further increase these risks. International guidelines for pediatric sepsis recommend antibiotic administration within 1 hour of diagnosis. Ampicillin and gentamicin in combination are the first line empiric therapy for newborns with suspected EOS.

Objective: The goal of this project is to decrease the lead time from time of order to administration of both ampicillin and gentamicin from 107 minutes to less than or equal to 60 minutes for newborns with suspected EOS in this level III Neonatal Intensive Care Unit (NICU) by April 2023. 

Design/Methods: This quality improvement project used lean methodology and the define, measure, analyze, improve, control framework. The outcome measure is the median number of minutes from time of order to administration of both ampicillin and gentamicin for NICU patients with suspected EOS. Monthly data were reviewed using a Tableau dashboard in the electronic health record. A statistical process control (SPC) XmR chart was generated to track the median time to antibiotic administration. The multidisciplinary NICU Comprehensive Unit Safety Program (CUSP) team identified this as a problem and initiated a project in July 2022. At the first meeting, it was identified that nurses were purposely delaying gentamicin administration to 1 hour after completion of ampicillin due to concern for possible gentamicin inactivation of ampicillin it was given too soon after ampicillin. The CUSP team pharmacist reviewed evidence and educated unit staff that there is no need to delay gentamicin administration. In addition, the team utilized the SIPOC (suppliers, inputs, process, outputs, and customers) tool, process maps, and fishbone diagram to generate a key driver diagram (Fig. 1) to identify additional root causes of failures and potential change ideas.  

Results: Data from July 2021 to November 2022 were reviewed. The baseline median number of minutes from order time to administration of both antibiotics was 107 minutes. After initiation of the CUSP project and initial education, there was a downward center line shift to 77 minutes (Fig. 2). 

Conclusion(s): This CUSP team used lean methodology to analyze the problem and identify root causes for delayed antibiotic administration for NICU patients with suspected EOS. Next steps include developing system-level interventions to reach the goal of antibiotic administration within 1 hour of diagnosis.