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.

Overall, the nearest health facility and primary school were reported correctly 84·9% (95% CI 82·2-87·2) and 73·4% (95% CI 68·8-77·7) of the time, respectively, based on straight-line distance (median distance 1486 m, IQR 1008-2241 m). The use of the self-reported nearest primary school was able to locate 82·0% (95% CI 78·1-85·8) of participants' compounds to the correct Euclidian distance catchment area (mean area of 6·7 km 2 ) ( Table 2) with a median distance to the self-reported nearest school of 878 m (IQR 522-1234 m). The self-reported nearest health facility was able to locate 78·1% (95% CI 73·8-82·1) of compounds to an area of 12·3 km 2 . When the combination of responses was tested, the mean area reduced to 1·7 km 2 and 48·7% (95% CI 43·6-53·6) of participants' compounds were correctly located.