Association of Center-Specific Patient Volumes and Early Respiratory Management Practices with Death and Bronchopulmonary Dysplasia in Preterm Infants

Abstract

Objectives: To describe variability in admission volumes and approach to early respiratory support between neonatal intensive care units in the Australian and New Zealand Neonatal Network and to evaluate whether these center-specific factors are associated with death and bronchopulmonary dysplasia.

Study design: This retrospective cohort study included 19 099 neonates born between 25 and 32 weeks' gestation and admitted to 1 of 25 NICUs from 2007 to 2013. Center-specific factors evaluated were annual admission volume and rate of using continuous positive airway pressure (CPAP) rather than intubation as the first mode of respiratory support. Logistic regression was used to examine any association of these center-specific factors with death, BPD, and death or survival with BPD (death/BPD). Analysis was performed separately for 2 gestation groups (25-28 weeks and 29-32 weeks inclusive).

Results: Admission volumes and rates of early CPAP use varied widely across centers. Higher admission volumes were associated with lower odds of death or survival with BPD in the 25-28 week group (aOR 0.93, 99% CI 0.88-0.99 per increase of 10 babies per center annually). Centers with higher early CPAP use did not have lower odds of death or BPD than centers that intubated more frequently.

Conclusions: Higher admission volumes are associated with more favorable outcomes for the more preterm infants in the Australian and New Zealand Neonatal Network. Further investigation is required to explore why the individual benefits of early CPAP do not translate to better outcomes for centers that use this approach most frequently.

Keywords: health services research; hospital mortality.