T H E S E presented and defended by Vanessa MACHAULT, before the FACULTY OF MEDICINE OF MARSEILLE

December 23, 2010

Abstract

Despite national an international efforts, malaria remains a major public health in many countries and sanitary systems are hindered by the lack of information on the actual burden of malaria, on the plasmodium transmission risk and on their geographical distribution. Nevertheless, spatial heterogeneity can be important and in this context, malaria control could be improved if could be focused in place and time. On one hand, the environment is a major determinant of malaria biodiversity, because of the vectorial transmission and the vectors bioecological preferences. On another hand, orbiting satellites can provide environmental, climatic and meteorological data that already have been used for the study of infectious diseases. “Tele-epidemiology” has been defined as an integrated approach aiming at associating medical, epidemiological or entomological ground data, with remotely-sensed environmental data, based on the in depth comprehension and measurement of the involved physical and biological mechanisms. In cities, it has already been possible to highlight associations between mapped urban settings and malariometric indices, using satellite data at appropriate scales. Among travellers, in the objective to evaluate malaria risk or efficacy of prophylactic devices, it would be useful to evaluate and predict transmission levels in the visited places. The objective of the present thesis was to identify environmental factors that could be remotely-sensed and that could be used in the evaluation of malaria risk among travellers on one hand and in urban settings on the other hand. First, remotely-sensed data have been used to evaluate levels of exposure to malaria risk of militaries, in the scope of a study on their risk factors for clinical malaria. Results have showed that, even when taking into account age and compliance to chemoprophylaxis as confusion factors, the environment was the factor the most strongly associated to clinical malaria risk. In parallel, an extensive entomological study has been conducted during five years in Dakar and allowed demonstrating a strong spatial and temporal heterogeneity of malaria transmission in the city. Collected information were centralized in a georeferenced database (GIS - Geographic Information System) containing all entomological, environmental, meteorological, biological and physical data collected on the field or by remote sensing. Finally, modelling of entomological risk in the capital city of Senegal was undertaken, based on data collected on the ground and environmental data issued from satellites. A first step showed the evolution of malaria transmission risk areas and allowed declaring that the fraction of human population that was at high risk for transmission decreased between 1996 and 2007. A second step led to the development of 1) a map of the breeding sites with a productivity indicator in Dakar city, 2) a map of aggressive adult Anopheles densities, and 3) a dynamic aspect was added to those maps, taking into account the variations of their meteorological determinants. The results of the work undertaken in this thesis demonstrated that remotely-sensed information, associated with a large amount of ground data, allow to adjust predictive models and to draw entomological risk maps, in urban settings or for moving populations.