RiVerMad (Contribution of mechanistic and spatio-temporal modelling approaches to the understanding of vector-borne diseases epidemiology: application to Rift Valley Fever virus in Madagascar)

"Rift Valley Fever (RVF) is a zoonotic disease caused by an arbovirus (RVF virus, or RVFV), affecting both livestock and humans, with a substantial public health and economic burden in Africa and the Arabian Peninsula. Circulating at low enzootic levels in some geographical areas, it may also occasionally cause important epizootics in ruminants, associated with zoonotic transmissions. The pathogen is mainly spread through the bites of infected Aedes and Culex mosquitoes, but transmission can also occur via direct contact with infected animal tissues or fluids, or via aerosols. Human clinical presentations range from mild, influenza-like symptoms to severe pathologies, including encephalitis and hemorrhagic fever. In Madagascar, important RVFV epizootics-epidemics occurred in 1990-1991, 2008-2009 and 2021, although the virus was also shown to circulate enzootically in-between these episodes. Although environmental factors and cattle trade have been suggested to favor the emergence or re-emergence of RVFV, the virus' epidemiological dynamics are poorly understood. Climate change is expected to affect the occurrence of RVFV outbreaks. The objectives of the RiVerMad project are (i) to build updated predicted risk maps of RVFV infection in Madagascar accounting for recent epidemiological trends in order to optimize surveillance, (ii) to explain the historical trends of the virus' spread within and between the different Malagasy eco-regions (especially regarding the 2021 outbreak), and (iii) to project scenarios for RVFV's future dynamics in the animal, vector and human compartments in a context of changing climate. We will combine serological and entomological data collected in multiple regions of the country, as well as cattle movement data, with two types of modelling approaches. First, we will build a spatio-temporal statistical models to predict epizootic risk, accounting for environmental covariates and livestock movements. Second, a meta-populations mechanistic model, incorporating vector, human, and livestock compartments and accounting for environmental variables, will mimic RVFV transmission dynamics within and between Malagasy eco-regions. It will be fitted to the data, and simulations of the model according to multiple climate change scenarios will allow to project future animal and human health burdens. The project will involve four teams in France and Madagascar, and build a bidirectional collaboration with field and virological expertise brought from Malagasy teams and modelling expertise brought by teams in France. Although based in France, the candidate will visit collaborators in Madagascar at least once, to visit data collection sites, to co-build surveillance strategies and to assist and train local students or researchers with analysis methods. The expected outcomes include two main peer-reviewed publications, but also the development of evidence-based surveillance strategies tailored to the Malagasy context and capacity building on all sides. Finally, the modelling framework may be used a basis for a further utilization in other areas in Africa."