Final Report Summary - IMCCA (Innovative Modelling to Optimise Control of Childhood Anaemia across Africa)
The goal of the IMCCA project was to develop rigorous data-driven Bayesian geostatistical models that link all available anaemia and mortality survey data in Africa with spatially explicit, model-based estimates of malaria, schistosomiasis, soil transmitted helminthiasis (STH), malnutrition and poverty in order to estimate anaemia burden, anaemia-related mortality in children under 5yrs old and assess the relative contribution of the main risk factors of anaemia across different regions in Africa with different levels of endemicity of the diseases related to anaemia.
The project has developed a comprehensive database of anaemia, malaria, malnutrition, mortality, childhood health interventions and socio-economic indicators across Africa by compiling, processing and geo-referencing all available household survey data in the continent. Furthermore, we compiled schistosomiasis and soil transmitted helminthiasis survey data across Africa through literature searches and made them publicly accessible via the Global Neglected Tropical Disease (GNTD) database at www.gntd.org. Data-driven Bayesian methodology was developed to enable the analyses of the above data. In particular we developed methodology to i) analyse very large non-stationary geostatistical data ii) align age-heterogeneous helminth surveys across locations to a common age by integrating geostatistical and mathematical transmission models iii) estimate the age-specific prevalence of helminth infections taking into account diagnostic error as a function of infection intensity iv) select the most important predictors that are related to the disease geographical distribution in a geostatistical model and to v) estimate the geographical variation of the effects of predictors on the disease outcome for large geostatistical data arising from different sources (that are not geographically aligned). The above methodology was used to obtain i) schistosomiasis and STH risk surfaces at high spatial resolution across sub-Saharan Africa (SSA) ii) high spatial resolution malaria risk estimates and assess the effects of malaria interventions in space iii) obtain countrywide estimates of the geographical distribution of anaemia and assess the contribution of factors related to it and (iv) estimate the contribution of moderate/severe anaemia and anaemia-malaria comorbidity to under-five year old child mortality.
The project has developed a comprehensive database of anaemia, malaria, malnutrition, mortality, childhood health interventions and socio-economic indicators across Africa by compiling, processing and geo-referencing all available household survey data in the continent. Furthermore, we compiled schistosomiasis and soil transmitted helminthiasis survey data across Africa through literature searches and made them publicly accessible via the Global Neglected Tropical Disease (GNTD) database at www.gntd.org. Data-driven Bayesian methodology was developed to enable the analyses of the above data. In particular we developed methodology to i) analyse very large non-stationary geostatistical data ii) align age-heterogeneous helminth surveys across locations to a common age by integrating geostatistical and mathematical transmission models iii) estimate the age-specific prevalence of helminth infections taking into account diagnostic error as a function of infection intensity iv) select the most important predictors that are related to the disease geographical distribution in a geostatistical model and to v) estimate the geographical variation of the effects of predictors on the disease outcome for large geostatistical data arising from different sources (that are not geographically aligned). The above methodology was used to obtain i) schistosomiasis and STH risk surfaces at high spatial resolution across sub-Saharan Africa (SSA) ii) high spatial resolution malaria risk estimates and assess the effects of malaria interventions in space iii) obtain countrywide estimates of the geographical distribution of anaemia and assess the contribution of factors related to it and (iv) estimate the contribution of moderate/severe anaemia and anaemia-malaria comorbidity to under-five year old child mortality.