PT - JOURNAL ARTICLE AU - Catherine A. Lippi AU - Anna M. Stewart-Ibarra AU - Ángel G. Muñoz AU - Mercy J. Borbor AU - Raúl Mejía AU - Keytia Rivero AU - Katty Castillo AU - Washington B. Cárdenas AU - Sadie J. Ryan TI - The social and spatial ecology of dengue presence and burden during an outbreak in Guayaquil, Ecuador, 2012 AID - 10.1101/112185 DP - 2017 Jan 01 TA - bioRxiv PG - 112185 4099 - http://biorxiv.org/content/early/2017/02/27/112185.short 4100 - http://biorxiv.org/content/early/2017/02/27/112185.full AB - Background Dengue fever, a mosquito-borne viral disease, is an ongoing public health problem in Ecuador and throughout the tropics, yet we have a limited understanding of the disease transmission dynamics in these regions. The objective of this study was to characterize the spatial dynamics and social-ecological risk factors associated with a recent dengue outbreak in Guayaquil, Ecuador.Methods We examined georeferenced dengue cases (n = 4,248) and block-level census data variables to identify potential social-ecological variables associated with the presence and burden of dengue fever in Guayaquil in 2012. We applied LISA and Moran’s I tests to analyze hotspots of dengue cases and used multimodel selection in R computing language to identify covariates associated with dengue incidence at the census zone level.Results Significant hotspots of dengue transmission were found near the North Central and Southern portions of Guayaquil. Significant risk factors for presence of dengue included poor housing conditions (e.g., poor condition of ceiling, floors, and walls), access to paved roads, and receipt of remittances. Counterintuitive positive correlations with dengue presence were observed with several municipal services such as garbage collection and access to piped water. Risk factors for increased burden of dengue included poor housing conditions, garbage collection, receipt of remittances, and sharing a property with more than one household. Social factors such as education and household demographics were nega>tively correlated with increased dengue burden.Conclusions Findings elucidate underlying differences with dengue presence and burden and indicate the potential to develop dengue vulnerability and risk maps to inform disease prevention and control, information that is also relevant for emerging epidemics of chikungunya and zika.AICAkaike’s Information CriterionAICcAkaike’s Information Criterion corrected for small sample sizeCDCCenters for Disease ControlCIconfidence intervalsDENVdengue virusDHFdengue hemorrhagic feverEWSearly warning systemsGAgenetic algorithmGISgeographic information systemGLMGeneralized Linear ModelINAMHINational Institute of Meteorology and HydrologyLISAlocal indicators of spatial associationNCEP-NCARNational Center for Environmental Protection–National Center for Atmospheric ResearchVIFvariance inflation factors