Investigating the spatial micro-epidemiology of diseases within a point-prevalence sample: a field applicable method for rapid mapping of households using low-cost GPS-dataloggers

Abstract

Point-prevalence recording of the distribution of tropical parasitic diseases at village level is usually sufficient for general monitoring and surveillance. Whilst within-village spatial patterning of diseases exists, and can be important, mapping infected cases in a household-by-household setting is arduous and time consuming. With the development of low-cost GPS-data loggers (< £40) and available GoogleEarth(TM) satellite imagery, we present a field-applicable method based on crowdsourcing for rapid identification of infected cases (intestinal schistosomiasis, malaria and hookworm) by household. A total of 126 mothers with their 247 preschool children from Bukoba village (Mayuge District, Uganda) were examined with half of these mothers given a GPS-data logger to walk home with, returning the unit the same day for data off-loading, after which, households were assigned GPS coordinates. A satellite image of Bukoba was annotated with households denoting the infection status of each mother and child. General prevalence of intestinal schistosomiasis, malaria and hookworm in mothers and children was: 27.2 vs 7.7%, 28.6 vs 87.0% and 60.0 vs 22.3%, respectively. Different spatial patterns of disease could be identified likely representing the intrinsic differences in parasite biology and interplay with human behaviour(s) across this local landscape providing a better insight into reasons for disease micro-patterning.

Figure 1

Schematic map of Uganda with part of the Mayuge District magnified: Bukoba village is situated on the Mayuge peninsula on the Lake Victoria shoreline. Lwanika and Bugoto were the other two villages surveyed in the large-scale parasitological survey.

Figure 2

GoogleEarth^TM map of Bukoba village with households annotated as blue circles. Other annotations include the blue swimmer identifying the primary water-contact site for the village as well as the yellow house, identifying the church which served as team headquarters during the study. The images were captured and ‘geostamped’ using the Oregon 550t unit: (1) the closest heath care centre (HC IV), around 2 km outside the map; (2) a small child about to be bathed in a basin full of water collected from the lakeshore; (3) the primary water-contact site for the village, with villagers’ normal water-contact behaviour (washing bicycles or clothes, collecting water or getting on boat); (4) a thatched hut with mud walls and (5) a tin-roofed house with wall made from brick.

Figure 3

Distribution of schistosomiasis, malaria and hookworms (green for uninfected and red for infected) by household and demographic group (children and mothers). In the top three maps, up to two children (each child symbolised by a circle) may be present per household.