Machine Learning for Prediction of Wait Times in Outpatient Clinic

Abstract

Patient perceptions on wait times during outpatient clinic visits are important factors affecting the satisfaction of patients. Providing precise information about patient wait times could enhance patients experience by minimizing uncertainty. Machine learning algorithms are powerful tools for analyzing and predicting the wait times from complex datasets. The current study predicted the patient waiting time before consultation and throughput time in an outpatient clinic using machine learning algorithms. The predictions are made using two algorithms Random Forest and XGBoost. The input variables used in the study are gender, day of visit, month of visit, time of visit, consultation start time, vitals examination, visit laboratory, visit pharmacy, repeated arrival, consultation session and weather condition. The performance is analyzed using performance indicators such as Accuracy, Precision, Sensitivity, Specificity, F measure and Receiver Operating Characteristic Curve. The model has the potential to be coupled with real-time data to obtain precise real-time estimates of patient wait times at outpatient clinics.