In this ecological study, we investigated spatial patterns of human visceral leishmaniasis (VL) incidence, its correlation with socioeconomic aspects, environmental indices (obtained through remote sensing) and canine VL during 2011-2016 in the municipality of Rondonópolis, a relevant endemic area for VL in Central-Western Brazil. Human VL cases were georeferenced and point patterns were analyzed by univariate Ripley's K function and Kernel density estimation (KDE). Poisson-based scan statistics were used to investigate spatial and spatiotemporal clusters of human VL incidence at the neighborhood level. Socioeconomic and environmental characteristics were compared between neighborhoods within and outside spatial human VL clusters. Also, we assessed the correlation between smoothed human VL incidence and canine VL seropositivity rates within and between neighborhoods. Human VL cases were clustered up to 2000 m; four hotspots were identified by KDE in peripheral areas. Spatial and spatiotemporal low-risk clusters for human VL were identified in central and southern areas. Neighborhoods within spatial low-risk cluster presented higher mean income, literacy rate, sanitary sewage service coverage and lower altitude, compared to the rest of the municipality. A positive correlation was found between the occurrence of human and canine VL. On the northern outskirts, high human VL incidence was spatially correlated with high canine VL seropositivity in surrounding neighborhoods. In conclusion, human VL demonstrated a heterogeneous, aggregated and peripheral spatial pattern. This distribution was correlated with intra-urban socioeconomic differences and canine VL seropositivity at the neighborhood level.
Keywords: Geographic information systems; Remote sensing technology; Socioeconomic factors; Spatial analysis; Spatio-temporal analysis; Visceral leishmaniasis.
Copyright © 2021 Elsevier B.V. All rights reserved.