Validation of three geolocation strategies for health-facility attendees for research and public health surveillance in a rural setting in western Kenya

STRESMAN, G. H.; STEVENSON, J. C.; OWAGA, C.; MARUBE, E.; ANYANGO, C.; DRAKELEY, C.; BOUSEMA, T.; COX, J.

The goals of the geolocation exercise will influence the optimum strategy. First, the ideal scale will depend on the spatial pattern of the disease and the size of the area of interest [5] . For example, if the objective was to identify foci of infections of a highly heterogeneous disease such as malaria in a low endemic or epidemic setting [7, 9, 20] then achieving higher precision would be essential. Conversely, if the distribution of sexually transmitted infections was being studied, less precision may be acceptable or even necessary to guarantee anonymity [20] . Second, the ideal strategy will depend on the purpose of geolocating cases. If it is for programmatic use such as passive public health surveillance, or to establish disease distribution at a regional or national level, then using the nearest health facility, with a larger mean catchment area may be sufficient. However, if greater precision and accuracy were required, for identification of foci for disease elimination or identifying where to implement control, for example, then knowing the exact boundaries of the catchment area or having a comprehensive postal network that can be geocoded to a high precision would be essential.