Predicting all-cause 90-day hospital readmission for dental patients using machine learning methods

Abstract

Introduction: Hospital readmission rates are an indicator of the health care quality provided by hospitals. Applying machine learning (ML) to a hospital readmission database offers the potential to identify patients at the highest risk for readmission. However, few studies applied ML methods to predict hospital readmission. This study sought to assess ML as a tool to develop prediction models for all-cause 90-day hospital readmission for dental patients.

Methods: Using the 2013 Nationwide Readmissions Database (NRD), the study identified 9260 cases for all-cause 90-day index admission for dental patients. Five ML classification algorithms including decision tree, logistic regression, support vector machine, k-nearest neighbors, and artificial neural network (ANN) were implemented to build predictive models. The model performance was estimated and compared by using area under the receiver operating characteristic curve (AUC), and accuracy, sensitivity, specificity, and precision.

Results: Hospital readmission within 90 days occurred in 1746 cases (18.9%). Total charges, number of diagnosis, age, number of chronic conditions, length of hospital stays, number of procedures, primary expected payer, and severity of illness emerged as the top eight important features in all-cause 90-day hospital readmission. All models had similar performance with ANN (AUC = 0.743) slightly outperforming the rest.

Conclusion: This study demonstrates a potential annual saving of over $500 million if all of the 90-day readmission cases could be prevented for 21 states represented in the NRD. Among the methods used, the prediction model built by ANN exhibited the best performance. Further testing using ANN and other methods can help to assess important readmission risk factors and to target interventions to those at the greatest risk.

Fig. 1. Cumulative readmission rate by days to readmission after hospital discharge.

Circle symbols indicate the cumulative readmission percentage (%) from 0 to 90 days after hospital discharge.

Fig. 2. ROC and AUC of logistic regression models by using balanced dataset and the original dataset.

The oversampling (Over), undersampling (Under), bothsampling (Both), random oversampling examples (ROSE), and synthetic minority over-sampling technique (SMOTE) were applied to balance the dataset.

Fig. 3. Important feature selection by using random forest.

Blue bars show the mean decreased Gini index for 46 predictor candidates.

Fig. 4. Decision Trees for 90-day readmission data sets.

A feature with higher entropy is located closer to the root (top), and a branch with zero entropy is converted to a leaf node (bottom). Each leaf node displays the probability of the “Yes” or “No” class at that node and the percentage of total observations used at that node.

Fig. 5. ANN for 90-day readmission data sets.

ANN structure: 16-3-1. The optimal weights of input variables and intercepts are shown as black and blue numbers above the lines separately.

Fig. 6. AUC of ML methods for the 90-day readmission.

Five supervised machine learning algorithms: Decision Tree (DT), Support Vector Machine (SVM), Artificial Neural Network (ANN), k-Nearest Neighbor (k-NN), and Logistic Regression (LR) are applied to build prediction models.