Spatio-temporal forecasting of dengue in the Americas -A data set
Description
This dataset accompanies the systematic review and meta-analysis entitled “Spatio-temporal forecasting of dengue in the Americas through hybrid mechanistic and data-driven models: Systematic review and meta-analysis.” It provides a comprehensive synthesis of dengue modeling efforts across the Americas from 2016 to 2025, integrating mechanistic, statistical, machine learning, and phylogenetic approaches. The files included present structured evidence on dengue dynamics, model structures, and analytical frameworks: Detailed Overview of Sub-model Structures: Comparative representation of vector–host and within-host modeling frameworks. Table 1. Data Extraction Matrix: Summary of key variables, parameters, and outcomes derived from the 19 included studies. Table 2. Comparative Summary: Model calibration, validation techniques, and analytical frameworks applied across studies. Table 3. Model Characteristics: Overview of dengue transmission model types, associated datasets, and contextual applications. Figure 1. PRISMA Flow Diagram: Study identification and selection process for the review. Figure 2. Geographical and Temporal Distribution: Mapping of dengue modeling studies across the Americas (2016–2025). Figure 3. Heatmap of Risk of Bias: Horizontal representation of quality assessment across the 19 studies. Figure 4. Forest Plot of Model Complexity: Comparative evaluation of mechanistic, statistical, ML/niche, and phylogenetic approaches. Figure 5. Compartmental Variables and Interventions: Forest plot synthesis of transmission model compartments and intervention strategies. Figure 6. Temperature Effects: Quantitative impact of temperature on dengue transmission risk in the Americas. Figure 7. Model Performance and Predictive Accuracy: ROC curve analysis across different modeling frameworks. Figure 8. General Compartmental Model: Conceptual diagram of dengue transmission in host–vector populations. Together, these materials support transparency, reproducibility, and meta-analytical comparison of dengue modeling strategies, offering a valuable reference for researchers, public health practitioners, and policymakers addressing vector-borne disease forecasting in the Americas.
Files
Steps to reproduce
Steps to Reproduce Define Review Protocol The systematic review and meta-analysis were conducted following PRISMA guidelines. Protocol registered in PROSPERO (ID: CRD420251130769). Literature Search Search performed across six databases (PubMed, Scopus, Web of Science, Embase, LILACS, and Cochrane Library). Timeframe: January 2016 – May 2025. Keywords included combinations of: “dengue”, “modeling”, “forecasting”, “simulation”, “machine learning”, “statistical”, “mathematical model”, “Americas”. Screening and Selection Duplicate removal. Title and abstract screening by two independent reviewers. Full-text screening based on predefined inclusion/exclusion criteria (focus on dengue transmission models applied in the Americas). Data Extraction Structured extraction of study characteristics, model type, calibration/validation methods, parameters, and outcomes into a standardized data extraction matrix (Table 1). Independent verification by a second reviewer. Quality and Bias Assessment Risk of bias assessed using a standardized evaluation framework. Results summarized in a horizontal heatmap (Figure 3). Data Synthesis and Meta-Analysis Comparative summary of calibration, validation, and analytical frameworks (Table 2). Forest plots (Figures 4–6) generated to quantify model complexity, intervention strategies, and environmental drivers (temperature). ROC curve analysis performed to compare predictive accuracy across model frameworks (Figure 7). Visualization and Conceptual Model Development Geographic and temporal mapping of included studies (Figure 2). PRISMA flow diagram constructed (Figure 1). General compartmental host–vector transmission model synthesized from reviewed studies (Figure 8).