Predictive Analysis of Outpatient Volumes of a First-class Grade A General Hospital through ARIMA Models

Abstract

Objectives. To explore the effect of ARIMA (Auto Regressive Integrated Moving Average) models in predicting the outpatient volume, the short-term prediction of the outpatient volume of a hospital, and to provide a basis for hospital management decisions related to outpatient volume. Methods. Extract the outpatient data for the period between January 2010 and March 2014 from the information system of a first-class grade A general hospital. The time series modeler in PASW (Predictive Analytics Software) was used in combination with ARIMA models, the model effect was evaluated, and the outpatient volumes for the next 2 years were predicted. Results. The number of outpatients during 2010–2013 amounted to 3.036 million, with an annual average growth rate of 24.07%. (Male/female ratio 0.81/1, mean age 40.36 ± 19.32, internal/external medicine ratio 1.35/1.) Based on the outpatient volume during 2010–2013, the predicted value of the outpatient volume in the first quarter of 2014 had a relative error of 4.11%, the model had a good predictive effect. Based on the outpatient volume between January 2010 and March 2014, the predicted value of the outpatient volume in 2014 was 1.132 million, and in 2015 was 1.295 million. Conclusions. With the help of PASW, ARIMA models applied to the prediction of the outpatient volumes of large general hospitals had were easy to operate, the models were a good fit with a satisfactory predictive effect, and the results were easy to explain. ARIMA models are tools for short-term prediction of seasonally fluctuating quantitative indexes (outpatient volumes, capacity, number of surgeries, etc.), and therefore worth popularization.

Key words::

• Hospital information management
• Outpatient management
• Outpatient volume
• ARIMA
• Prediction

Declaration of interest

Project funds: Training funds for research talent projects of the General Hospital of Chengdu Military Region, scientific research projects (120566) of the Health Department of Sichuan Province.