Modelling coverage of Indoor Residual Spraying (IRS)

In this project, we generate high-resolution maps of the proportion of household sprayed with IRS annually across sub-Saharan Africa. Mapping the distribution of IRS coverage at fine-scale is important for understanding what effect IRS has on malaria burden across Africa and for targeting interventions in the future. IRS is more costly than insecticide treated nets and therefore countries adopt a targeted subnational approach. 

IRS campaign data and survey data are combined to generate our IRS coverage estimates. Firstly, coarse administrative level IRS coverage data is gathered from a variety of sources including National Malaria Control Programme reports, President’s Malaria Initiative spray reports and the World Health Organisation. Secondly, using Demographic and Health Surveys (DHS) data, spatiotemporal geostatistical models are developed to predict the fine-scale structure of IRS coverage. Environmental and anthropological covariates are used to inform the model. We find that accessibility, high probability of occurence of Anopheles gambiae mosquitoes, high temperature suitability index for Plasmodium falciparum, and low elevation correlate with high IRS coverage. Finally, we combine the first two steps with the administrative level data is used to calibrate the results of the geostatistical model. 

By utilizing multiple data sources we produce a complete picture of IRS coverage over the past 20 years that may be valuable for policy-makers at international, national, and subnational scales. 

2020Indoor residual spraying for malaria control in sub‑Saharan Africa 1997 to 2017: an adjusted retrospective analysisMal J
DataSpatial ResolutionSpatial CoverageTemporal ResolutionTemporal Coverage
IRS Coverage5 KmAfricaYearly2000-2020


Liverpool School of Tropical Medicine

Big Data Institute, University of Oxford

President’s Malaria Initiative