Modelling Dynamic Behaviour of Out-Patient Department Visits at the University of Cape Coast Hospital Using Time Series Analysis

Abstract

Background: Accurate and reliable forecasting of outpatient department visits enhances decision-making and planning for future healthcare demands and is the foundation for greater and better utilization of healthcare resources and increased levels of outpatient care and satisfaction. Though the literature has proposed several candidate models for predicting outpatient visits in some hospitals in Ghana, the model regulating outpatient visits at the University of Cape Coast Hospital (UCC) is unknown. There is therefore a need to determine the best model applicable to the specific case of UCC Hospital. Aim: This study sought to determine and model the dynamics of outpatient visits in UCC Hospital and to project outpatient healthcare demands at the facility for the period July 2021 to July 2024. Methods: This paper employed a monthly periodicity of 114-time series data sourced from District Health Information Management Systems Two (DHIMS 2) on outpatient department visits at UCC Hospital from January 2012 to June 2021. The autoregressive integrated moving average (ARIMA) models which are a form of the classical Box-Jenkins approach of Time Series Analysis were used to analyse the data. Analysis was performed in EViews 12. Results: The study results showed twenty-five non-seasonal tentative models for the hospital and ARIMA (4, 1, 4) was selected as the best fit model with a fourth-order autoregressive and moving average terms each and one order of nonseasonal differencing. Residual analysis of the fitted model indicates that the model is adequate for forecasting. The findings revealed an overall rising trend in the incidence of outpatient department visits to the hospital over the study period with an average of 6000 visits per month. This is expected to increase to over 7000 visits per month over the next three-year period of July 2021 to June 2024 according to projections. Conclusion: The forecast of outpatient visits in this study serves as early signals to the management of the University hospital and is intended to enhance human and material resource planning and allocation for better-quality healthcare delivery.