A Supervised Model for Predicting the Risk of Mortality and Hospital Readmissions for Newly Admitted Patients

Abstract

Mortality and hospital readmission are serious problems in healthcare, due to their negative impact on patients in specific and hospitals in general. Expenditure on healthcare has been increasing over the last few decades. This increase can be attributed to hospital readmissions; which is defined as a re-hospitalization of a patient after being discharged from a hospital within a short period of time. In this paper, we propose a system that is capable of using a supervised learning model to predict the status of patients when they get discharged from the hospital. For example, given a medical dataset, we would like to build a model that is capable of predicting with a high accuracy the likelihood that a newly admitted patient could be at risk of death or hospital readmission. For the purpose of prediction, we select five machine learning algorithms, namely Naïve Bayes, Decision Tree, Logistic Regression, Neural Networks, and Support Vector Machine. The results show the ability to predict the risk of mortality and hospital readmission with an accuracy of 82.78% and 63.59% respectively. The objective of the paper extends the scope of prediction to further include certain methods, such as feature engineering and boosting, to enhance the prediction accuracy.