Development and Validation of a Mortality Prediction Model in Extremely Low Gestational Age Neonates

Area of Science:

• Neonatalogy
• Pediatric Critical Care
• Medical Informatics

Background:

• Neonatal mortality remains a significant concern, particularly in extremely low gestational age (ELGA) infants (<28 weeks).
• Predictive models can aid clinicians in identifying high-risk neonates for targeted interventions and family counseling.

Purpose of the Study:

• To develop and validate an early prediction model for in-hospital mortality in ELGA neonates.
• To utilize readily available clinical variables within the first hour of birth for model development.

Main Methods:

• A predictive cohort study utilized data from the Swedish Neonatal Quality Register (2011-2021).
• A model was developed using 80% of the data and validated on the remaining 20%.
• Model performance was assessed using the area under the receiver operating characteristic curve (AUC).

Main Results:

• The study included 3,752 ELGA neonates with an 18% in-hospital mortality rate.
• The developed BAG model (Birth weight, Apgar score, Gestational age) achieved an AUC of 76.9% in the training cohort.
• The BAG model demonstrated an AUC of 68.9% in the validation cohort, outperforming models using only birth weight or gestational age.

Conclusions:

• The BAG model offers a novel, accessible tool for predicting mortality in ELGA neonates.
• Early identification of high-risk infants can inform clinical management and family support strategies.