(120) Evaluating Early CPAP Use for Term Newborns in Michigan from 2020-2024

Background: Respiratory morbidity is the leading indication for NICU admission in term infants. Maternal, infant, and system factors are associated with increased risk for newborn respiratory conditions. Continuous positive airway pressure (CPAP) is often used both in the delivery room and after initial resuscitation, however its routine use has not been systematically evaluated for term infants.

Since its addition to NRP in 2011 as a delivery room consideration regardless of gestational age, use of CPAP for term infants has increased significantly over time, with an associated increase in complications such as pneumothorax. In light of this, there has been a recent movement to safely de-implement perceived over-use of CPAP at certain institutions. The NRP guidelines reflect the existing limitations in the current evidence, allowing for variability in the providers' clinical practice. It is currently unknown how much variation exists in provider CPAP use, including threshold for use in or out of the delivery room, duration of use and approach to discontinuation.

The Obstetric Initiative (OBI) is a quality improvement collaborative aimed at enhancing care for low-risk term pregnancies across the state of Michigan. Despite stable cesarean section rates over recent years, OBI data has shown an increase in neonatal morbidity, specifically regarding the use of CPAP after birth. This dataset provides a lens into the use of CPAP across many different delivery settings.

Objectives: 1) To identify maternal, infant, and system factors associated with CPAP use
2) To evaluate hospital-level variation in CPAP use
3) To identify characteristics of high and low CPAP utilization hospitals

Design/Methods: Retrospective cohort study of liveborn nulliparous term singleton vertex deliveries in the OBI database from 1/1/2020 to 12/31/2024.
We will present the first phase of our approach which includes descriptive statistics of outcome and predictor variables in adjusted Odds Ratios (aOR).
We may present our assessment of the degree of hospital-level variation, through the intraclass correlation coefficient (ICC).
Ultimately, we will build a hierarchical logistic regression model which will be reported as the predicted probability of receipt of CPAP by hospital in a caterpillar plot. The model will not be completed in time for this presentation.

Predictor variables include maternal, infant, and system factors
Outcome variables include CPAP used within 2 hours of birth (primary), and assisted ventilation for >6 hours (secondary)

Results:

Conclusions: