PM2031: Reducing Over-utilization of High Flow Nasal Cannula in Children with Bronchiolitis

Purpose/Objectives: Bronchiolitis is a leading cause of hospitalizations amongst children younger than 2 years. High flow nasal cannula (HFNC) has been associated with increased costs and intensive care unit (ICU) utilization without significant improvement in clinical outcomes. Despite this, HFNC is increasingly utilized in children hospitalized with bronchiolitis. At our institution, nearly all children hospitalized with bronchiolitis received HFNC support.
The objective of this project was to reduce the proportion of children with bronchiolitis who receive HFNC from a median of 100% to 60% in 6 months at our local academic children’s hospital.

Design/Methods: This quality improvement study included hospitalized children aged 1 month to 2 years with bronchiolitis and excluded those with prematurity < 32 weeks or pre-existing cardiopulmonary, genetic, congenital or neuromuscular conditions. Baseline data was collected from November 2019 to March 2020 and post-intervention data from October 2021 to March 2022. Measures included proportion of children who receive HFNC (process), length of oxygen supplementation and length of stay (outcome), and rates of ICU transfer, 7-day emergency department (ED) revisits, and 7-day readmissions (balancing). Stakeholders were gathered and implemented a HFNC initiation protocol which incorporated a categorical respiratory score, criteria for HFNC initiation, a multidisciplinary care-team huddle, and an emphasis on supportive care (Figure 1). Implementation was supported by integration of the respiratory score into the electronic medical record with decision tools, staff education, and audit and feedback. The Model for Improvement was utilized, testing changes via PDSA cycles. Analysis was completed using run charts as well as Fisher Exact and Mann-Whitney U testing.

Results: We analyzed 284 patients (104 baseline and 180 post-intervention). The proportion of children who received HFNC decreased from a median of 100% to 60% within two months of implementation, which was sustained over the next three months (Figure 2A). Length of oxygen supplementation decreased from a median of 46 to 35 hours (IQR 50-38.5, p=.005) and the length of stay decreased from a median of 54 to 43 hours (IQR 51-42, p=.02) (Figure 2B). There were no significant differences in ICU transfers, 7-day ED revisits, or 7-day readmissions.

Conclusion/Discussion: This quality improvement project safely reduced over-utilization of HFNC, length of oxygen supplementation, and length of stay for children with bronchiolitis through implementation of a HFNC initiation protocol.