110 - Evaluating Risk Factors to Create a Risk Assessment Tool to Reduce Unplanned Extubations in a Level IV NICU

Publication Number: 3105.110

Alexandra J. Golden, Weill Cornell Medicine, New York, NY, United States; Caitlin Ehret, NewYork-Presbyterian Komansky Children’s Hospital, New York, NY, United States; Meril Rajan, New York Presbyterian Weill Cornell, Orangeburg, NY, United States; Priyanka Tiwari, Weill Cornell Medicine, Williston Park, NY, United States

Background: Unplanned extubations (UEs) are the most common preventable adverse event during mechanical ventilation in the neonatal intensive care unit (NICU), and have been shown to cause significant short and long-term patient harm. UEs are often multifactorial in nature due to patient-related factors and factors related to equipment and the environment. Quality improvement (QI) initiatives have been shown to be effective in reducing UEs. UE prevention requires early risk identification, however, there is a lack of risk assessment tools validated for use in neonates, highlighting the need for a standardized risk assessment.

Objective: The SMART aim was to decrease the rates of UEs/100 ventilator days by 10% by June 2026 in intubated patients in a level IV neonatal intensive care unit through the creation of a risk assessment tool.

Design/Methods: A multidisciplinary team conducted an 65-month observational study (May 2020 to Sept 2025), with a baseline from May 2020 to Sept 2024. Previous baseline interventions included UE bundle audits, endotracheal taping standardization, Kamishibai card creation, debrief guide and airway card creation and X-ray annotation. PDSA cycles in the intervention period included utilization of the Delphi method to identify risk factors (Cycle #1), provider, nursing and RT education (Cycle #2) and creation and implementation of a risk assessment tool, comprised of 10 factors (Cycle #3) (Fig 1). The outcome measure was UEs/100 ventilator days and the process measure was the presence of multiple providers during high risk (HR) events. The balancing measure was avoidance of HR activities. Statistical process control charts (p and U charts) were used to analyze our data. Associates for Process Improvement (API) rules for special causes were applied.

Results: Sequential interventions reduced the rate of UEs from a baseline of 1.7 to 0.7 UEs/100 ventilator days (Fig 2) during the baseline, but there was no center line change with recent PDSA cycles. The number of UEs during HR events with multiple providers present did not change over time (Fig 3). There was no change in frequency of HR activities throughout the initiative.

Conclusion(s): Implementation of the Delphi method, provider, nursing and RT education, and creation and implementation of a neonatal risk assessment tool for intubated patients resulted in a sustained decrease in the number of UEs/100 ventilator days but no further decline. Future efforts will focus on embedding the risk assessment tool in the EMR to streamline the workflow across the hospital enterprise which will help evaluate the use of the tool as a process measure.

Figure 1: NICU High Risk Assessment Tool for Intubated Patients


Figure 2
Outcome Measure: U chart for UEs/100 ventilator days. X-axis is time (month_year). Y-axis is the UEs/100 ventilator days. Arrows with PDSA cycle number indicating time of intervention. The solid line indicates the center line (CL). Dashed lines indicate upper control limit (UCL) and lower control limit (LCL).

Figure 3
Process Measure: P-chart for % of UEs During High Risk Events with 2 Providers Present. X-axis is time (month_year). Y-axis is the % of UEs during high risk events with 2 providers present. Arrows with PDSA cycle number indicating time of intervention. The solid line indicates the center line (CL). Dashed lines indicate upper control limit (UCL) and lower control limit (LCL).