Modelling Child Survival at Birth Data Using Logistic Regression Model: A Case Study of Yobe State Specialist Hospital Damaturu, Nigeria

Abstract

This research examines factors influencing the survival of child at birth in Yobe State Specialist Hospital using a well-known logistics regression model. A total of 150 data points were collected through transcription from the maternity record of the Hospital to test significant factors that affect survival of child at birth. Analysis of logistic regression model was applied to the data using the Generalized Linear Model (GLM) package in R programming software. The results indicate that Type of delivery and weight at birth have the most significant influence on the probability of child survival at birth at 0.001 level of significance. Correlation analysis results show that all the five variables (age, parity, apgar, birth weight, and type of delivery) have a weak relationship, which implies that there is no multicollinearity in the data. Therefore, these results may help policy makers and health personnel to educate pregnant women on the effect of overweight baby at birth to reduce the incidence of deaths at birth. This study recommend that pregnant women should be educated about the effect of baby weight in a worm as that may increases the chance of caesarean section which in turn may affect the likelihood of child survival at birth. Further studies are suggested to consider factors like educational level, income of level of the family, antennal status, and blood pressure. A more advance community-based survey is recommended since not pregnant women attain formal health care facilities for antennal and postnatal services, which may expose more factors that influence child survival at birth.