Abstract

The impact of climate change on the burden of malaria frequently evaluated using mechanistic and statistical models, remains inconclusive. Most analytic model do not directly incorporate interactions between geo-climatic changes, malaria interventions and socioeconomic effects, which are likely to influence the disease burden in the future. This is partly explained by the scarcity of quality, long-term malaria data.

There are now a few health and demographic surveillance systems (HDSS) in Africa that routinely collect data on vector densities, malaria incidence, interventions, household-related HDSS offer unique platforms and opportunity for modelling the spatio-temporal interactions between climatic and non-climatic factors on malaria burden. Strengthening model-based disease surveillance for forecasting outbreaks will enhance preparedness to the impacts of climate change. However, our knowledge about their forecasting potentials for early warning compared to purely statistical models is limited.

In this talk we present robust Bayesian spatiotemporal to estimate the contribution of climatic factors related to the changes in malaria incidence and mortality in the DRC and South Africa that can be used to identify outputs of climate models with the highest ability to predict the disease burden; and select the models for projecting the climate effects on the disease.