Nov 13 - 17, 2016

LAND COVER DETERMINANTS OF PLASMODIUM FALCIPARUM PREVALENCE IN URBAN AND PERI-URBAN AREAS OF NORTHERN BIOKO ISLAND

Jordan M. Smith, Dianna E.B. Hergott, Christopher Schwabe, Wonder P. Phiri, Akum Aveika, Jose Luis Segura

MCDI, Silver Spring, MD, United States

Beginning in 2015, the Bioko Island Malaria Control Project (BIMCP) adopted a strategic and targeted approach of indoor residual spraying (IRS) for malaria prevention. Evidence suggests that hotspots of malaria transmission to be targeted are best evaluated by finding increased exposure to infectious mosquito bites. Despite routine vector monitoring throughout the island, mosquito collection in urban and peri-urban areas of northern Bioko has been logistically challenging. Remote sensing and geographic information systems (GIS) are frequently used to explore associations between land use/land cover (LULC) and mosquito-borne diseases by functioning as a proxy for mosquito abundance, while spatial scan statistics are used to detect spatial clustering of malaria prevalence. Data on prevalence of P. falciparum parasitemia were collected from a representative sample of 5,286 households throughout the island during the 2015 Malaria Indicator Survey. Seven LULC types were classified through supervised classification of remotely sensed high resolution satellite imagery. Areas at higher risk of transmission were evaluated using a Bernoulli purely spatial scan statistic. Spatial associations of LULC and P. falciparum parasitemia were performed using geographically weighted logistic regression analyses to determine case environmental risk factors. Complete data were available for 19,666 individuals. Although several statistically significant clusters were detected, two clusters of excess risk appeared to have been driven by one to two households with several cases. The most likely spatial clustering of malaria prevalence was detected in peri-urban areas, particularly in the northwestern region of the island, which appears to be temporally heterogeneous. Results of geographic weighted regression will be analyzed further. Plasmodium falciparum malaria prevalence is heterogeneous in space in this urban and peri-urban study area. Geographical and housing risk factors associated with prevalence will be explored further. This analysis improves the planning of IRS interventions targeting high-risk areas.

Source: http://www.astmh.org/ASTMH/media/Documents/ASTMH-2016-Annual-Meeting-Abstract-Book.pdf