TOP 75 - Electronic health record-based clinical decision support tool to minimize excess oxygen exposure

Publication Number: 4780.TOP 75

Ruth A. Riedl, Boston Children's Hospital, Boston, MA, United States; Youyang Yang, Boston Children's Hospital, Boston, MA, United States; Alireza Akhondi-Asl, Boston Children's Hospital, Boston, MA, United States; Dennis Daniel, Boston Children's Hospital, Boston, MA, United States; Alon Geva, Harvard Medical School, Boston, MA, United States

Background: Oxygen therapy is a cornerstone in the management of critically ill pediatric patients. However, a growing body of literature indicates that potentially excess oxygen exposure can be harmful. Both hyperoxia and liberal SpO2 targets have been associated with adverse outcomes, including increased mortality, longer hospitalizations, and prolonged invasive ventilation. Given our group’s prior finding that patients in our ICU are frequently exposed to high FiO2 even in the absence of hypoxia and that this exposure is associated with excess mortality, our goal is to develop tools to reduce this potentially harmful exposure. This study aims to utilize an electronic health record (EHR)-based clinical decision support (CDS) tool to alert providers in real-time to cumulative excess oxygen exposure (CEOE) concerns with the goal of reducing the duration of exposure to excess oxygen.

Objective: The objective of this study is to develop, implement, and evaluate an EHR-based CDS tool to minimize CEOE in pediatric patients receiving supplemental oxygen in the pediatric intensive care unit.

Design/Methods: This was a prospective observational study comparing CEOE before and after implementation of an educational intervention and an EHR alert prompting providers to decrease the FiO2 when the FiO2 > 21% and the SpO2 < 95% without recent documented hypoxia (SpO2 < 90% within 2 hours). All patients aged 0-21 years admitted to the Boston Children’s Hospital (BCH) Medical-Surgical ICU from June 2025-November 2025 who required supplemental oxygen with accurately measurable FiO2 were included. Patients on ECMO or with pulmonary hypertension or congenital diaphragmatic hernia were excluded. The primary outcome metric was CEOE and secondary outcome metrics were time to 21% FiO2, ventilator-free days in 28 days, mortality, and hospital length of stay. Process measures include alert frequency, alert deferral frequency, number of changes to FiO2 after alert, and percentage of time with hypoxia. The balancing metric was frequency of other alert deferral. Stakeholder meetings were conducted with those involved in oxygen titration. An EHR-based CDS tool was developed in conjunction with a brief educational video to educate providers on the risks of CEOE and on the intervention. FiO2 and SpO2 documentation by bedside nurses and respiratory therapists was collected. We utilized an interrupted time-series analysis to determine change in CEOE after implementation of the CDS intervention. This study was reviewed by the BCH IRB and determined to be exempt from IRB approval.