Hot-spots of HIV infection in Cameroon: a spatial analysis based on Demographic and Health Surveys data

Abstract

Background: The Human Immunodeficiency Virus(HIV) infection prevalence in Cameroon has decreased from [Formula: see text] in 2004 to [Formula: see text] in 2018. However, this decrease in prevalence does not show disparities especially in terms of spatial or geographical pattern. Efficient control and fight against HIV infection may require targeting hotspot areas. This study aims at presenting a cartography of HIV infection situation in Cameroon using the 2004, 2011 and 2018 Demographic and Health Survey data, and investigating whether there exist spatial patterns of the disease, may help to detect hot-spots.

Methods: HIV biomarkers data and Global Positioning System (GPS) location data were obtained from the Cameroon 2004, 2011, and 2018 Demographic and Health Survey (DHS) after an approved request from the MEASURES Demographic and Health Survey Program. HIV prevalence was estimated for each sampled area. The Moran's I (MI) test was used to assess spatial autocorrelation. Spatial interpolation was further performed to estimate the prevalence in all surface points. Hot-spots were identified based on Getis-Ord (Gi*) spatial statistics. Data analyses were done in the R software(version 4.1.2), while Arcgis Pro software tools' were used for all spatial analyses.

Results: Generally, spatial autocorrelation of HIV infection in Cameroon was observed across the three time periods of 2004 ([Formula: see text], [Formula: see text]), 2011 ([Formula: see text], [Formula: see text]) and 2018 ([Formula: see text], [Formula: see text]). Subdivisions in which one could find persistent hot-spots for at least two periods including the last period 2018 included: Mbéré, Lom et Djerem, Kadey, Boumba et Ngoko, Haute Sanaga, Nyong et Mfoumou, Nyong et So'o Haut Nyong, Dja et Lobo, Mvila, Vallée du Ntem, Océan, Nyong et Kellé, Sanaga Maritime, Menchum, Dounga Mantung, Boyo, Mezam and Momo. However, Faro et Déo emerged only in 2018 as a subdivision with HIV infection hot-spots.

Conclusion: Despite the decrease in HIV epidemiology in Cameroon, this study has shown that there are spatial patterns for HIV infection in Cameroon and possible hot-spots have been identified. In its effort to eliminate HIV infection by 2030 in Cameroon, the public health policies may consider these detected HIV hot-spots, while maintaining effective control in other parts of the country.

Keywords: Cameroon; Clustering; Getis–Ord statistics; HIV; Hot-spots; Spatial.