Title: Time Series Analysis And Forecasting Of Emergency Cases At Rivers State University Teaching Hospital Using Autoregressive Integrated Moving Average (Arima) Model
Authors: BEREBON, Raphael Nubari and NKUTURUM, Christiana
Volume: 10
Issue: 9
Pages: 129-143
Publication Date: 2026/09/28
Abstract:
This study investigated the pattern, trend, and forecasting of emergency department (ED) cases at the Rivers State University Teaching Hospital (RSUTH), Nigeria, using the Autoregressive Integrated Moving Average (ARIMA) technique. The study covered a ten-year period from January 2014 to December 2023 for both Monthly and yearly records. The data analysis explored descriptive statistics, trend analysis, correlation analysis, and ARIMA modeling procedures using Minitab 22 and SPSS software. The findings of this study revealed that Road Traffic Accident recorded the highest average number of emergency visits per yearly, followed by Congestive Cardiac Failure, Acute Appendicitis, and Stab Wound, while Anxiety-Disorder recorded the lowest average number of visits. The trend analysis showed that eight emergency variables exhibited upward trends, namely Bleeding per Vagina, Cardiogenic Shock, Ectopic Pregnancy, Head Injury, Gunshot Injury, Renal Failure, Anxiety Disorder, and Stab Wound, whereas Road Traffic Accident, Acute Appendicitis, Congestive Cardiac Failure, while Acute Intestinal Obstruction exhibited downward trends. The correlation analysis indicated the existence of both positive and negative moderate relationships among some emergency variables. Positive moderate relationships were observed between Renal Failure and Bleeding per Vagina, Anxiety Disorder and Cardiogenic Shock, Congestive Cardiac Failure and Acute Intestinal Obstruction, and Anxiety-Disorder and Renal Failure. Negative moderate relationships were observed between Congestive Cardiac Failure and Renal Failure, as well as between Anxiety-Disorder and Congestive Cardiac Failure. The study concluded that ARIMA techniques provided effective and reliable tools for modeling and forecasting emergency department visits.