Effectiveness of a care bundle for primary prevention of intraventricular hemorrhage in high-risk neonates: a Bayesian analysis

Abstract

Objective: To determine whether an intraventricular hemorrhage (IVH) prevention bundle featuring midline-elevated positioning reduced IVH among high-risk infants.

Study design: In a retrospective study design, we compared outcomes of infants <1250 grams birth weight or <30 weeks gestation before (N = 205) and after (N = 360) implementation of an IVH prevention bundle, using Bayesian and frequentist logistic regression to determine whether the intervention decreased any grade IVH.

Results: In both the Bayesian and frequentist analyses, there was no difference in odds of any grade IVH before and after the implementation of the prevention bundle (OR 0.993; 95% Credible Interval 0.751-1.323 and OR 1.23; 95% Confidence Interval 0.818-1.864 respectively). Bias analyses suggested that these results were robust to bias from potential deaths attributable to IVH.

Conclusion: In this retrospective analysis, we found no evidence for a protective effect of an IVH prevention bundle on IVH incidence among high-risk neonates at a level IV NICU.