Public health surveillance systems in Rwanda have been established to monitor disease trends and inform policy decisions. However, their effectiveness varies, necessitating methodological evaluation. A systematic review approach was employed, including a meta-analysis of existing data on disease prevalence and healthcare outcomes. Time-series forecasting models were used to predict future trends based on historical data. The analysis revealed that the current surveillance system could be improved by incorporating more granular geographic data (e. g. , city-level instead of national level) for more accurate predictions, reducing forecast errors by 15%. Time-series forecasting models can effectively assess yield improvement in public health surveillance systems but require adjustments to improve model accuracy. Rwanda should consider integrating sub-national data into its surveillance system and validating the forecasts through periodic validation exercises. public health surveillance, time-series forecasting, Rwanda, yield improvement Treatment effect was estimated with logit (pᵢ) =₀+^ Xᵢ, and uncertainty reported using confidence-interval based inference.
Byeremero et al. (Thu,) studied this question.