Forecasting Malaria Dynamics in Western Kenya
Journal of Global Health
Download
Abstract: This study applies empirical dynamic modeling (EDM) to forecast malaria incidence in western Kenya using climatic and intervention variables. Based on data from 2008?2022, the research identifies nonlinear causal relationships between malaria and factors including temperature, rainfall, humidity, wind speed, and bed net coverage. Lagged effects and threshold values for each variable are computed, with optimal conditions found to be temperature between 30?35°C, rainfall between 30?120 mm, and humidity between 67?80%. Bed net coverage above 90% significantly reduces transmission risk. The model improves forecasting accuracy compared to traditional approaches and offers a valuable tool for climate-sensitive early warning systems. The paper calls for integration of forecasting models into national health surveillance platforms and for cross-sectoral data harmonization to guide public health policy and resource allocation under changing climate conditions.