Spatial modelling of malaria risk factors in a rural sector in the east of Rwanda

Malaria is a vector borne disease posing a severe health risk to the population of Sub Saharan Africa and particularly in the East African Valley. Over the past decades, billions of dollars have been invested to combat malaria, but it is still killing hundreds of thousands of people annually. The aim of this study is to spatially model malaria risk factors in a primarily rural area situated in south eastern Rwanda. The underlying data for the research were obtained through a combination of high-resolution ortho-photography and primary data collection. In addition to this, detailed malaria occurrence and socio-economic data were made available. Spatial clusters of malaria occurrence were subsequently determined using Getis and Ord spatial statistics. This cluster analysis showed that malaria distribution is characterized by zones with high malaria risk, so called hot spots, zones with moderate malaria risk known as not significant spots and zones of low malaria risk known as cold spots. Preliminary outcomes demonstrate that malaria prevalence varies from one household to another and from one administrative unit (village) to another. The relationship between malaria prevalence and malaria risk factors was assessed using a logistic regression model. Results clearly indicate that malaria infection increases with the proximity to irrigated farmland. It also increases with household size. It was also proven that lower housing quality (mud houses; unburnt brick walls, earth floor) is associated with higher risk of malaria infection.