Related Experiment Video
Updated: Jul 9, 2026
Assessment and Evaluation of the High Risk Neonate: The NICU Network Neurobehavioral Scale
Published on: August 25, 2014
Perinatal prediction of mortality for neonates born at 22-25 weeks gestation
Xuxin Chen1,2, Tianyao Lu3, Davide De Francisco3
1Division of Neonatology, Department of Pediatrics, Columbia University Vagelos College of Physicians and Surgeons and New York-Presbyterian, New York, NY, USA. xc2663@cumc.columbia.edu.
Objective:
To predict mortality of 22-25 weeks' gestation neonates in a population-based cohort using characteristics available at birth.
Study Design:
This observational study included 8981 actively treated 22-25 weeks' gestation infants born between 2008 and 2018 at CPQCC hospitals. Proposed models were used to relate maternal, infant, and birth hospital factors to the primary outcome of death before hospital discharge. These models were compared based on predictive discrimination and calibration.
Results:
Death before hospital discharge occurred in 2824 (32%) infants. In addition to the five factors in the NICHD model, we found two additional variables, receipt of prenatal care and maternal age, were associated with death. The c-statistics ranged from 0.72 to 0.75 across all models. When compared to the NICHD 5-factor model, our models showed similar discrimination.
Conclusion:
We found additional factors to the NICHD 5-factor model associated with death before hospital discharge, although these factors did not improve prediction performance.