PREDICTING THE MODE OF DELIVERY AND THE RISK FACTORS ASSOCIATED WITH CESAREAN DELIVERY USING DECISION TREE MODEL

Background: The rate of cesarean section has been increasing worldwide and particularly Tamil Nadu has more than 60% of cesarean section. Objective: The purpose of this study is to affirm and suggest that decision tree model can be used to predict the mode of delivery and the risk factors associated with cesarean delivery. Study design: This is a study of women delivered live-born neonates in 2015 through 2017 (4043). The frequency of cesarean delivery is 61.61%; 33 variables are used for analysis. Decision tree model is applied to the 50% of the sample dataset to develop the predictive training models and the same is applied to the remaining 50% dataset. This method is applied on the sample data and the outcomes are tested to predict the mode of delivery. Results: 4043 women who had attempted delivery are included in the study. The overall modes of delivery are cesarean delivery (61.61%) and vaginal delivery (38.39%). The risk factors that are associated with cesarean delivery are age, height, BMI, child weight, HBP, sugar, thyroid, toxemia, multiple pregnancies, breech presentation, sleep disturbance. The highest risk for cesarean delivery of the patients can still have a 50% possibility to vaginal delivery. Conclusion: Decision Tree model can be used to predict the delivery mode. Applying this method to both training and testing dataset, the accuracy of the outcomes produced is 100% and 99.50%. However, the patients at highest risk for cesarean delivery have 50% possibility of successful vaginal delivery and therefore should be allowed to attempt vaginal delivery.