This Data Descriptor focuses on evaluating community health centre systems in Uganda from to, with a specific emphasis on developing and testing a time-series forecasting model for risk reduction. A time-series analysis approach was employed, incorporating data from various sources including patient records and administrative databases. The model utilised an autoregressive integrated moving average (ARIMA) technique to forecast future trends in service utilization and health outcomes. The ARIMA model demonstrated a significant reduction in forecasting errors by over 20% compared to baseline models, indicating improved accuracy in predicting risk factors such as disease prevalence and resource demand. This study highlights the potential of time-series forecasting for enhancing community health centre performance and risk management in Uganda. The findings contribute to evidence-based policy development aimed at improving healthcare delivery systems. The validated model should be further tested across different geographical regions and with varied data sets to ensure its robustness and applicability. Future research could explore integrating additional variables for more comprehensive forecasting. Treatment effect was estimated with logit (pᵢ) =₀+^ Xᵢ, and uncertainty reported using confidence-interval based inference.
Onyango et al. (Fri,) studied this question.