Dengue fever is an emerging infectious disease in the Galápagos Islands of Ecuador, with the first cases reported in 2002 and subsequent periodic outbreaks. We report results of a 2014 pilot study conducted in Puerto Ayora (PA) on Santa Cruz Island, and Puerto Baquerizo Moreno (PB) on San Cristobal Island. To assess the socio-ecological risk factors associated with dengue and mosquito vector presence at the household level, we conducted 100 household surveys (50 on each island) in neighborhoods with prior reported dengue cases. Adult mosquitoes were collected inside and outside the home, larval indices were determined through container surveys, and heads of households were interviewed to determine demographics, self-reported prior dengue infections, housing conditions, and knowledge, attitudes, and practices regarding dengue. Multi-model selection methods were used to derive best-fit generalized linear regression models of prior dengue infection, and Aedes aegypti presence. We found that 24% of PB and 14% of PA respondents self-reported a prior dengue infection, and more PB homes than PA homes had Ae. aegypti. The top-ranked model for prior dengue infection included several factors related to human movement, household demographics, access to water quality issues, and dengue awareness. The top-ranked model for Ae. aegypti presence included housing conditions, mosquito control practices, and dengue risk perception. This is the first study of dengue risk and Ae. aegypti presence in the Galápagos Islands.

Background Dengue, chikungunya, and Zika are arboviruses of major global health concern. Decisions regarding the clinical management of suspected arboviral infection are challenging in resource-limited settings, particularly when deciding on patient hospitalization. The objective of this study was to determine if hospitalization of individuals with suspected arboviral infections could be predicted using subject intake data. Methodology/Principal findings Two prediction models were developed using data from a surveillance study in Machala, a city in southern coastal Ecuador with a high burden of arboviral infections. Data were obtained from subjects who presented at sentinel medical centers with suspected arboviral infection (November 2013 to September 2017). The first prediction model-called the Severity Index for Suspected Arbovirus (SISA)-used only demographic and symptom data. The second prediction model-called the Severity Index for Suspected Arbovirus with Laboratory (SISAL)-incorporated laboratory data. These models were selected by comparing the prediction ability of seven machine learning algorithms; the area under the receiver operating characteristic curve from the prediction of a test dataset was used to

18Background: Dengue, chikungunya, and Zika are arboviruses of major global health concern. 19 Decisions regarding the clinical management of suspected arboviral infection are challenging in 20 resource-limited settings, particularly when deciding on patient hospitalization. The objective of 21 this study was to determine if hospitalization of individuals with suspected arboviral infections 22 could be predicted using subject intake data. Methodology/Principal Findings: Two prediction 23 models were developed using data from a surveillance study in Machala, a city in southern coastal 24Ecuador with a high burden of arboviral infections. Data was obtained from subjects who presented 25 at sentinel medical centers with suspected arboviral infection (November 2013 to September 26 2017. The first prediction model-called the Severity Index for Suspected Arbovirus (SISA)-27 used only demographic and symptom data. The second prediction model-called the Severity 28Index for Suspected Arbovirus with Laboratory (SISAL)-incorporated laboratory data. These 29 models were selected by comparing the prediction ability of seven machine learning algorithms; 30 the area under the receiver operating characteristic curve from the prediction of a test dataset was 31 used to select the final algorithm for each model. After eliminating those with missing data, the 32 SISA dataset had 534 subjects, and the SISAL dataset had 98 subjects. For SISA, the best 33 prediction algorithm was the generalized boosting model, with an AUC of 0.91. For SISAL, the 34 best prediction algorithm was the elastic net with an AUC of 0.94. A sensitivity analysis revealed 35 that SISA and SISAL are not directly comparable to one another. Conclusions/Significance: Both 36 SISA and SISAL were able to predict arbovirus hospitalization with a high degree of accuracy in 37 our dataset. These algorithms will need to be tested and validated on new data from future patients. 38 Machine learning is a powerful prediction tool and provides an excellent option for new 39 management tools and clinical assessment of arboviral infection. 40 41 Keywords: machine learning, arboviral disease, dengue fever, undifferentiated febrile illness 42 43 Author Summary 44 Patient triage is a critical decision for clinicians; patients with suspected arbovirus infection are 45difficult to diagnose as symptoms can be vague and molecular testing can be expensive or 46 unavailable. Determining whether these patients should be hospitalized or not can be challenging, 47 especially in resource-limited settings. Our study included data from 543 subjects with a diagnosis 48 of suspected dengue, chikungunya, or Zika infection. Using a machine learning approach, we 49 tested the ability of seven algorithms to predict hospitalization status based on the signs, 50 symptoms, and laboratory data that would be available to a clinician at patient intake. Using only 51 signs and symptoms, we were able to predict hospitalization with high accuracy (94%). Including 52 laboratory data also resulted in hig...

No abstract