Eliminating polio, and many other vaccine-preventable childhood diseases, requires ensuring that almost all children under 5 years of age receive the vaccine. To achieve this, accurate and recent data on the numbers of under 5 year old children and where they live are needed to calculate how much vaccine is needed, how many vaccinators are required and for how long, and for planning and monitoring vaccinator routes and vaccine delivery. In northern Nigeria, initial efforts relied on projections from the 2006 census data to undertake this, together with broad estimates of the proportions of the population under 5 years of age. These data were quickly found to be inaccurate in many places, with either vaccine running out where more children than expected were found, or too much vaccine left where there were too few, resulting in wastage.
In collaboration with the Bill and Melinda Gates Foundation and Oak Ridge National Laboratories, WorldPop worked on new approaches to estimate population numbers and distributions in the absence of census data. High resolution satellite imagery was processed to map all settlements in the country, and these were also classified into settlement and neighbourhood types. Field survey teams then visited a selection of these settlements and neighbourhoods covering a range of geographies to measure population densities in each type. Geostatistical methods were then used to construct spatial models of population distributions based on the survey training data and geospatial covariate datasets. These models were used to predict population counts across the country and were integrated with similar models for estimating population age structures to produce new estimates of where children under 5 years old were.
TEDx talk by WorldPop director, Professor Tatem, highlighting how the integration of geospatial datasets can support disease elimination estimates
A range of studies have since shown the value and accuracy of the new estimates over the projected 2006 census data, and the estimates now form the basis for planning in the Nigeria vaccination tracking system (http://vts.eocng.org). The work lead to the adoption of a similar approach in Afghanistan and the establishment of the GRID3 program, where geospatial support and analysis is provided in collaboration with multiple national statistics offices across Africa (grid3.org).
Screenshot of the Nigeria vaccination tracking system, showing estimated population numbers produced by WorldPop researchers.