Landscape determinants of Saint Louis encephalitis human infections in Córdoba city, Argentina during 2010.

VERGARA-CID, C; ESTALLO, EL; ALMIRÓN, WR; CONTIGIANI, M; SPINSANTI, L

ACTA TROPICA

ELSEVIER SCIENCE BV

Lugar: Amsterdam; Año: 2013 vol. 125 p. 303 - 308

0001-706X

AbstractBackgroundSaint Louis Encephalitis virus (SLEV) is endemic in Argentina. During 2005 an outbreak occurred in Cordoba, being the first register of epidemic activity in South America. In the following years only sporadic cases were registered, but during summer-autumn of 2010 an outbreak occurred in Cordoba city with a lower magnitude than the one reported in 2005. Understanding the association of different landscape elements related to SLEV hosts and vectors in urban environments is important for identifying high risk areas for human infections, which was here evaluated. MethodsThe current study uses a case-control approach at a household geographical location, considering symptomatic and asymptomatic human infections produced by SLEV during summer-autumn 2010 in Cordoba city. Geographical information systems and logistic regression analysis were used to study the distribution of infected human cases and their proximity to water bodies, vegetation abundance with a threshold value of Normalized Difference Vegetation Index (NDVI) higher than 0.3, agricultural fields and housing density classified as high/low density urban constructions. Population density at a neighborhood level was also analyzed as a demographic variable. Results Logistic regression analysis revealed vegetation abundance (NDVI>0.3) was significantly and positively associated with the presence of human infections by SLEV. Areas with lower density of urban constructions were also important to predict the occurrence of SLEV infections. Other variables like different types of water bodies and agricultural fields were not associated to the occurrence of human infections by this virus. The logistic model was used to develop a map of probability of human infections in Cordoba city. The population density analysis shows that SLEV infections are more likely to occur when population density by neighborhood is lower. ConclusionsThe model highlights areas that are most likely to experience SLEV infections. Major landscape contributing to the outbreak of SLEV in 2010 were the proximity to places with NDVI>0.3 (parks, squares, riversides) and the presence of low density urban constructions, like residential areas. These findings and the predictive map developed could be important and useful for public health surveillance and to improve prevention of vector-borne diseases.