O3232: Combating Overutilization by Reducing Length of High-Flow Nasal Cannula Treatment via Power Weans

Purpose/Objectives: Reduce the average length of HFNC treatment by 20% from 76.8 to 61.4 hours in 3 months.

Design/Methods: The Model for Improvement was utilized testing interventions for meaningful change in desired outcomes. Measures included length of HFNC treatment (outcome), proportion of patients on HFNC who receive a power wean (process), and the power wean pass rate, escalations in care, and readmissions (balancing). We created a respiratory therapist driven power weaning protocol whereby the HFNC rate is decreased by 50%. The protocol included criteria qualifying patients for a power wean, expectations for reassessment, and indications and next steps for failing a power wean. We embedded the process into an existing HFNC titration guide and implementation was supported by just-in-time staff education, audit and feedback, and communication tools for the care team.

Results: There were 683 patient encounters during the baseline period (1/1-6/3/23) and 1,214 during the post-intervention period (6/4/23-1/6/24). The average time on HFNC decreased from a median of 76.8 hours to 53.6 hours and the standard deviation decreased from 95.7 hours to 41.5 hours. Power wean use increased such that by the end of the study period a median of 52% of patients on HFNC received at least one power wean. Patients with bronchiolitis were most likely and asthma least likely to receive power weans (293, 52.4% vs 63, 27.4%; p< 0.0001). Patients with English as their preferred language were also less likely to receive one (333, 38.0% vs 161, 47.8%; p=0.007); but this difference disappeared when accounting for diagnosis. There was no difference across payer, ethnicity, race, or gender. A majority of patients passed their power weans (333, 89%). In the post-intervention population, we saw a decrease in ICU escalations (18, 2.6% vs 3, 0.2%; p< 0.0001) and 30-day readmissions (29, 4.2% vs 8, 0.7%; p< 0.0001), and no difference in 3-day ED revisits (2, 0.3% vs 0, 0%, p=0.13).

Conclusion/Discussion: Using a power wean protocol, we successfully reduced the average length of HFNC treatment for hospitalized children across all diagnoses by 30% within 3 months of implementation without an increase in ICU escalations or 30-day readmissions.