Forecasting emergency department patient flow using Markov chain

Abstract

Emergency department (ED) crowding has developed into a risk that severely impacts on the emergency medical service quality and access to health care. With regard to the essential cause of crowding, subjectively, it results from low efficiency of resource collocation or incomprehension of the patient flow rule; objectively, arises from the low patient flow rate. Hence, the prediction of patient flow in emergency department is significant in providing early warning of ED crowding for the interest department or staff, monitoring and controlling the condition of ED crowding, providing the policy to alleviate Emergency department (ED) crowding. The proposed methods for forecasting ED patient flow and crowding merely focus on temporal near-future prediction. In this paper, the authors suggest a patient flow network for supporting Markov chain model that forecasts ED patient flow rule; namely, it is an approach intended to predict the space transfer trend of patient flow from an ED module to another based on the ED process. The authors mainly concentrate on expounding the principle of the Markov chain model and comparing subsequent real-time data with those predicted. Preliminary results suggest that the proposed model is of considerable ease of use and of great value for forecasting the ED crowding and patient flow.