A Neonatal Mortality Risk Modelling Scoring System

Abstract

Infections are the commonest cause of death in infants less than four weeks old. Treatment, in a neonate with signs of sepsis needs to be initiated as soon as possible, before the causative organism is known. In the developed world neonatal severity scores have been created to estimate the risk of a neonate having a poor outcome. These scores rely on biochemical and haematological parameters which are often unavailable in the developing world. The objective of this study is to derive a mortality severity score for neonates who live in developing countries.

The neonate mortality risk score is based on clinical signs that predict the likelihood of death. Neonatal patient risk models are built applying two popular methodologies: Logistics Regression and Decision Tree. Input variables that were used in established models in literature was selected to build the neonatal mortality risk modelling scoring system. Important factors that contributed to the resulting neonatal mortality risk score was birth weight, temperature, heart rate and seizure. Model accuracy was at least 85% amongst all models built.