Fine scale spatial mapping of urban malaria prevalence for microstratification in an urban area of Ghana

BackgroundMalaria in urban areas is a growing concern in most sub-Saharan African countries. The growing threats of Anopheles stephensi and insecticide resistance magnify this concern and hamper elimination efforts. It is therefore imperative to identify areas, within urban settings, of high-risk of malaria to help better target interventions. MethodsIn this study, we combined a set of environmental, climatic, and urban covariates with observed data from a malaria prevalence study and used geospatial methods to predict malaria risk in the Greater Accra Region of Ghana. Georeferenced data from 12,371 surveyed children aged between 6 months and 10 years were included in the analysis. ResultsPredicted malaria prevalence in this age group ranged from 0 to 52%. Satellite-driven data on tasselled cap brightness, enhanced vegetation index and a combination of urban covariates were predictive of malaria prevalence in the study region. We produced a map that quantified the probability of malaria prevalence exceeding 10%. ConclusionsThis map revealed areas within the districts earmarked for malaria elimination that have high malaria risk. This work is providing evidence for use by the National Malaria Elimination Program and District Health Managers in planning and deploying appropriate malaria control strategies. Summary boxO_ST_ABSWhat is already known?C_ST_ABSReduction in malaria incidence globally has stalled in the past few years. Malaria endemic countries are being encouraged to use local data to inform appropriate malaria control strategies. Malaria prevalence studies seldomly provide estimates below regional administrative levels. The availability of environmental, climatic, and socioeconomic factors as well as computational methods has enhanced predictive methods that quantifies the disproportionate variation of malaria risk between and within urban areas. What are the new findings?Predictive maps of malaria at high spatial resolutions such as 100m allows for visualizing fine-scale heterogeneity of malaria in neighbourhoods. Inclusion of urban covariates in models predicting malaria risk in urbanized communities helps to account for socioeconomic disparities and their effect on malaria risk. What do the new findings imply?Malaria control efforts needs to be guided by highly granular data. Systems to generate granular data on a continuous basis needs to be strengthen in malaria endemic countries, especially, to better inform deployment of appropriate interventions in resource constraint settings. This type of analysis provide information on which intervention is appropriate in a specified geographical area.