Advances in mapping malaria for elimination: fine resolution modelling of Plasmodium falciparum incidence

Abstract

The long-term goal of the global effort to tackle malaria is national and regional elimination and eventually eradication. Fine scale multi-temporal mapping in low malaria transmission settings remains a challenge and the World Health Organisation propose use of surveillance in elimination settings. Here, we show how malaria incidence can be modelled at a fine spatial and temporal resolution from health facility data to help focus surveillance and control to population not attending health facilities. Using Namibia as a case study, we predicted the incidence of malaria, via a Bayesian spatio-temporal model, at a fine spatial resolution from parasitologically confirmed malaria cases and incorporated metrics on healthcare use as well as measures of uncertainty associated with incidence predictions. We then combined the incidence estimates with population maps to estimate clinical burdens and show the benefits of such mapping to identifying areas and seasons that can be targeted for improved surveillance and interventions. Fine spatial resolution maps produced using this approach were then used to target resources to specific local populations, and to specific months of the season. This remote targeting can be especially effective where the population distribution is sparse and further surveillance can be limited to specific local areas.

Figure 1. Generalised mapping of fine scale P. falciparum incidence in low transmission settings.

(A) Shows an example landscape. Certain areas such as swamps are breeding sites for anopheles mosquitoes that sustain malaria transmission resulting in focal transmission areas (hotspot with infected houses). Often hotspots are located in areas far from health facilities. (B) The probability of seeking treatment at health facilities by population located far from health facility catchment area reduces with distance in addition to other socio-demographic factors. With infected people generally having low immunity in a low-endemic malaria transmission settings, fever onset is likely to be rapid once infected, leading to presentation at health facilities. However some can be missed, such as the house in the lower right corner. (C) Assembled cases over time at facilities combined with remotely sensed data from satellites (vegetation) showing potential risk areas. The vegetation index map was obtained from Moderate Resolution Imaging Spectrometer (MODIS) satellite imagery (http://modis.gsfc.nasa.gov/) and map created in ESRI ArcGIS 10.2 software (http://www.esri.com/software/arcgis/). Satellite data cover large geographic areas and are useful in predicting risk in areas with no sampled data. (D) Mapping from health facility data coupled with environmental covariates leads to a gridded fine spatial resolution map of predicted incidence for each month, highlighting only populated areas (coloured squares) that can be targeted for interventions and active surveillance.

Figure 2. Generalised schematic representation of data flow and analysis for prediction of incidence.

Technical details of the modelling framework at prediction stage are provided in the methods section.

Figure 3. Spatio-temporal maps of predicted mean monthly incidence of P. falciparum per 1000 population in northern Namibia in areas with population density greater than 0.01 people per km².

The maps were created in ESRI ArcGIS 10.2 software (http://www.esri.com/software/arcgis/). Malaria incidence peaked in the December to April period and was lowest in May to October. The maps suggest less spatial variation between May and October. Accompanying uncertainty maps are included in the supplementary information. Data were from 29 months (from January 2012 to May 2014) based on confirmed malaria cases (n = 20,689 cases) from 322 health facilities in northern Namibia.

Figure 4

(A) The modelled seasonal pattern of incidence of P. falciparum per 1000 population by month in 22 northern districts in Namibia. Modelling results suggest that incidence varied by district and month, but was lowest between May and October. (B) Scatterplot showing the estimated mean population at risk of P. falciparum malaria in Namibia in 2014 and the Bayesian credible intervals based on predicted incidence by district and month. The credible intervals (Crl: Bayesian credible interval) were wider where the estimated mean population at risk was more than 10000 but closer to the mean where population at risk was <10000. (C) The estimated population at risk of P. falciparum malaria in Namibia in 2014 by district and month. Population estimates are based on the worldpop dataset (http://www.worldpop.org.uk/) for Namibia. P. falciparum incidence estimates were modelled in a Bayesian framework from assembled malaria cases between 2012 and 2014 and incorporating space-time covariates.

Figure 5. Graphical representation of malaria incidence modelling framework.

P is the output pixel level incidence at pixel level, η is the specification of linear predictor with spatial effects ν, covariate effects β, residual error σ_e, and effect of the month (time) τ_t.