Daily hospital outpatient visits prediction based on Seasonal Autoregressive Integrated Moving Average model

ObjectivesThis study aimed to create a straightforward and rapid tool for predicting daily outpatient visits across the entire hospital, capturing the inherent fluctuation patterns of outpatient visits to offer a reference for the overall operation of hospital outpatient services. MethodThe daily outpatient visits recorded by the information system of a tertiary grade A general hospital from January 1, 2023 to December 31, 2024 was collected. Using data from January 1, 2023 to November 30, 2024, a Seasonal Autoregressive Integrated Moving Average (SARIMA) model was established to predict the daily outpatient visits in December 2024. ResultsThe number of outpatients showed obvious cyclical characteristics and weekly seasonal trends. On the whole, the number of outpatient visits on Monday and Saturday is significantly higher than that on the other five days of the week, the number of outpatient visits on Sunday is the lowest, while the number of outpatient visits on Wednesday is slightly higher than that on Thursday and Friday. The SARIMA(1,1,0)(0,1,1)7 was the optimal model with strong fitting capabilities and good predictive performance, the mean absolute percentage error (MAPE) of the predicted result was 0.073. ConclusionsThe SARIMA model can effectively simulate the trend of changes in hospital outpatient numbers over time series, providing a theoretical basis for short-term forecasting of hospital outpatient visits. This can not only help hospitals optimize internal resource allocation and improve operational efficiency, but also provide patients with a better and more efficient medical experience. WHAT IS ALREADY KNOWN ON THIS TOPIC{Rightarrow} Medium and long term forecasts cannot offer adequate support for the precise allocation of medical resources at the operational level. Accurate short-term forecasts can offer theoretical support for managers to schedule human resources and prepare for emergencies, which is often more practical. WHAT THIS STUDY ADDS{Rightarrow} Our study indicates that a rapid tool for predicting daily outpatient visits can be developed by incorporating the weekly seasonal effect on outpatient numbers, while also capturing the inherent fluctuation patterns. HOW THIS STUDY MIGHT AFFECT RESEARCH,PRACTICE OR POLICY{Rightarrow} This study can help hospital managers arrange the working hours of medical staff reasonably, and deploy medical supplies such as drugs, medical equipment, etc. to ensure patients medical needs. {Rightarrow}Guide chronic disease patients to seek medical attention during off peak hours, alleviate the pressure of high outpatient visits during peak periods, and shorten patients waiting time.