Spatial pattern of reinfestation by Triatoma infestans in Chancanif, Argentina

We examined the environmental correlates and the spatial pattern of infestation by Triatoma infestans, a vector of Chagas disease, in a rural area of Argentina five years following an insecticidal campaign. Patterns of infestation were identified in an entomological survey, as mapped with high-resolution satellite imagery and analyzed in a geographic information system. Logistic regression was used to relate infestation to observed household characteristics as well as the location and density of households. Location was the most significant predictor of infestation for domiciles. For peridomestic structures surrounding the domiciles, the combination of location and the presence/absence of goat pens was most significant. In considering any infestation, whether domiciliary or peridomestic, the combination of location, presence/absence of animal pens, and the type of household construction were found to be most significant. Using these statistical relationships to back-classify the field data resulted in accuracies between 85% and 87%. A map of infestation probability for the town of Chancaní was developed from the logistic regression.