Is autumn the key for dengue epidemics in non endemic regions? The case of Argentina

A. E. CARBAJO; M. V. CARDO; P. C. GUIMAREY; A. A. LIZUAIN; M. P. BUYAYISQUI; T. VARELA; M. E UTGÉS; C. M. GIOVACCHINI; M. S. SANTINI

PeerJ

Peer J

Año: 2018 vol. 6

Background. Dengue is a major and rapidly increasing public health problem.In Argentina, the southern extreme of its distribution in the Americas, epidemictransmission takes place during the warm season. Since its re-emergence in 1998 two major outbreaks have occurred, the biggest during 2016. To identify the environmental factors that trigger epidemic events, we analyzed the occurrence and magnitude of dengue outbreaks in time and space at different scales in association with climatic, geographic and demographic variables and number of cases in endemic neighboring countries.Methods. Information on dengue cases was obtained from dengue notificationsreported in the National Health Surveillance System. The resulting database wasanalyzed by Generalized Linear Mixed Models (GLMM) under three methodologicalapproaches to: identify in which years the most important outbreaks occurred inassociation with environmental variables and propose a risk estimation for future epidemics (temporal approach); characterize which variables explain the occurrence of local outbreaks through time (spatio-temporal approach); and select the environmental drivers of the geographical distribution of dengue positive districts during 2016 (spatial approach).Results. Within the temporal approach, the number of dengue cases country-widebetween 2009 and 2016 was positively associated with the number of dengue cases in bordering endemic countries and negatively with the days necessary for transmission(DNT) during the previous autumn in the central region of the country. Annual epidemic intensity in the period between 19992016 was associated with DNT during previous autumn and winter. Regarding the spatio-temporal approach, dengue cases within a district were also associated with mild conditions in the previous autumn along with the number of dengue cases in neighboring countries. As for the spatial approach, the best model for the occurrence of two or more dengue cases per district included autumn minimum temperature and human population as fixed factors, and the province as a grouping variable. Explanatory power of all models was high, in the range 57-95%.