Many health outcomes of international public health importance can be mapped at global or continental scale and high resolution using geostatistical methods combined with publicly available georeferenced data. Several household-level determinants of enteric infectious disease (EID) transmission are routinely ascertained in censuses and surveys but have yet to be comprehensively mapped at high resolution for EID-endemic regions: access to an improved drinking water source and sanitation facility, having a covered floor (as opposed to bare earth), caregiver education and crowded living conditions. Without high-resolution sub-national estimates of the distribution of these covariates within at-risk populations, it will not be possible to produce pathogen-specific estimates of disease burden to inform vaccine development and program priorities. The long-term goal of this proposed project is to provide reliable estimates of populations at elevated risk of EID transmission to inform the allocation of public health resources. The overall objective is to provide the research community with standardized geostatistical model-based maps of the current and near-future distribution of five important household-level risk factors for EID transmission in a format that can be easily used in a geographic information system (GIS). The central hypothesis is that prevalence of these factors varies spatiotemporally as a function of geography, socio-demography and environment and can be modelled using publicly available global datasets and geostatistical methods. The rationale underlying the proposed research is that high resolution global maps of the selected covariates will make it possible to identify hotspots of infectious disease transmission risk, further map the predicted incidence of selected pathogens over large areas and project trends several years into the future. We plan to objectively test the central hypothesis and thereby attain the objective of this project by pursuing the following specific aims:
Aim 1 : Compile a large database of georeferenced data relating to five selected household-level covariates of infectious disease transmission.
Aim 2 : Produce and make publicly available global modeled surfaces of each covariate using standardized geostatistical methods, environmental covariates. Data relating to these five household-level risk factors will first be compiled from publicly available census and household survey data. Then project will produce raster files using spatiotemporally explicit hierarchical generalized linear regression model will be fitted to the point-prevalence data within a Bayesian model-based geostatistical framework and using a suite of candidate environmental and socio-demographic predictors selected by generalized additive models. The gridded predictions will be made available to the public for download through an online repository, so that they can be imported into a GIS and used in further analyses by the end user. Researchers or program planners who wish to characterize risk factors for disease transmission in a particular community will be able to import these surfaces into a GIS as layers and extract standardized estimates for the suite of five variables at any location for which the coordinates are known.
Many important public health indicators can now be mapped at global or continental scale, however, several household-level determinants of infectious disease transmission that are ascertained in censuses and surveys have yet to be mapped at high resolution across endemic regions. We propose to produce geostatistically modeled surfaces of estimates of these indicators using georeferenced population-level data. The gridded predictions will be made available to the public for download through an online repository, so that they can be imported and analyzed in a GIS by the end user.
