Improving Spontaneous Breathing Trials With a Respiratory Therapist-Driven Protocol

Abstract

Background

Respiratory therapist (RT)-driven spontaneous breathing trial (SBT) protocols have been shown to improve patient outcomes.

Research Question

Can an RT-driven SBT protocol be implemented and sustained to improve outcomes?

Study Design and Methods

This quality improvement (QI) project aimed to standardize and re-establish RT-driven protocol for screening patients for SBT readiness and administering SBTs to appropriate patients. Endotracheally intubated and mechanically ventilated adult patients admitted to an academic medical center ICU were screened daily by RTs for SBT readiness. Eligible patients received an SBT with extubation decisions made by the physician team. Patient demographics, indications for intubation, SBT eligibility and exclusionary indications, SBT ventilator settings, start times, duration, and outcomes were collected from the electronic health record. QI interventions included staff re-education, documentation tips, creation of process maps, and interdisciplinary open forum discussions.

Results

One hundred twenty-eight patients representing 759 safety screen weaning assessment opportunities were included over a baseline sample and three plan-do-study-act (PDSA) cycles. Documentation of SBT eligibility increased from 25% at baseline to 86% in PDSA cycle 3 (P ≤ .001). Patients assessed to be eligible for and who received an SBT constituted 42% at baseline, 35% at PDSA cycle 1, 36% at PDSA cycle 2, and 51% at PDSA cycle 3 (P = .092). Use of the protocolized SBT ventilator settings improved significantly from 18% to 83% (P ≤ .001). Patients who started an SBT before 9 am increased from 41% to 67% (P = .097), and the median duration of SBT decreased from 211 to 64 min (P = .008).

Interpretation

This study shows that standardization of an RT-driven SBT protocol is feasible despite multiple obstacles, including staffing and communication challenges and poor shared understanding of terminology.