Table 2.
Geolocation of cases and hotspot identification
| Reference, year | Geolocation of cases | Hotspot identification |
| HIV (one study) | ||
| Goswami et al29 | Notified cases of either tuberculosis (TB) (N=150), HIV (N=665) or syphilis (155) between 1 January 2005 and 31 December 2007 were geolocated to households. The method for geolocation of the cases was not described in the paper. | A kernel density map of the cases was produced. Areas with the highest densities of three diseases of HIV, syphilis and TB (greater than 10 cases per square mile) were classified as hotspots. Two hotspot neighbourhoods were identified in the county. |
| TB (four studies) | ||
| Moonan et al24 | Notified TB cases (N=991) from 1 January 1993 to 31 December 2000 in Tarrant County, north central Texas, USA were geolocated to zip codes using residential addresses and zip codes that patients gave at the time of diagnosis with the aid of a GIS software. | Areas with the highest TB notification rates and high percentage of genotypically clustered TB isolates were identified as hotspots. Three neighbourhood hotspots were identified. |
| Goswami et al29 | Notified cases of TB (N=150), HIV (N=665) and syphilis (N=155) that were notified between 1 January 2005 and 31 December 2007 were geolocated to households. The method for geolocation of the households was not described in the paper | A kernel density map was developed. Areas with the highest densities of three diseases of HIV, syphilis and TB (greater than 10 cases per square mile) were classified as hotspots. Two hotspot neighbourhoods were identified in the county. |
| Cegielski et al31 | Notified TB cases between 1985 to 1995 (N=128) and all notified LTBI from 1993 to 1995 (N=311) were geocoded to their households using the addresses that patients gave at the time of diagnosis. In addition, field workers tracked addresses to households to get household coordinates of addresses that failed to geolocate. | The points of cases were plotted on a map and areas with the densest clusters of points of cases were identified as hotspots, two neighbourhoods were identified in the county. |
| Leprosy (three studies) | ||
| De Souza Dias et al32 | Notified leprosy cases that occurred between 1998 and 2002 (N=368) in the municipality of geocoded to households. The method for geolocation of cases was not described. | Density map with a radius of 100 m of the notified leprosy cases was produced. Four hotspot areas were identified |
| Jim et al33 | Notified leprosy cases from 2002 to 2006 (N=502) were geolocated to households. Field workers visited all notified cases to get household GPS coordinates using a GIS device. | A density map based on 1 mile radius of the notified leprosy cases. Areas with high concentration of cases classified as hotspots. |
| Barreto et al25 | Notified leprosy cases from January 2004 to February 2010 (n=633) were geocoded to households. Field workers visited households of registered cases to collect GPS coordinates. | Hotspots were identified using the Kulldorff’s spatial scan statistic and by stratification of the leprosy notified rates. Two hotspots were identified. |
| Malaria (four studies) | ||
| Srivastava et al34 | Notified malaria cases between 2000 and 2005 were obtained from the State Department of Health based on the cases notified in clinics in the blocks or districts. | Blocks or districts with a percentage of Plasmodium falciparum malaria cases of all notified cases that was either 100% or consistently greater than 30% from 2000 to 2005 or greater than 70% in 2005 |
| Herdiana et al35 | Notified malaria cases from 2007 to 2008 in addition to other self-reported malaria cases that were found during a survey (n=319) were geocoded to households. Field workers obtained the GPS coordinates of households using GIS devices. | Villages that had the majority of malaria cases were classified as hotspots. 14 out of 18 villages were identified |
| Bousema et al36 | June and July 2011, 17 503 individuals tested in a malaria prevalence survey for the prevalence of P. falciparum antibodies (AMA-1 or MSP-10). Field workers collected the GPS coordinates of the households using GPS devices. | Segments of the study area were scanned in the 2×4 km rolling windows and areas with higher (p<0.05) prevalence of antibodies and age-adjusted antibody density than the local average values were identified as hotspots. |
GIS, geographical information system; GPS, global positioning system; m, metres; N, number; STD, sexually transmitted disease.