Dataset for Feature Importance Analysis in the Construction of the Composite Dengue Mortality Risk Index in Mexico

This dataset contains the integrated clinical, environmental, and territorial variables used in the feature importance analysis for the construction of the Composite Dengue Mortality Risk Index in Mexico. The epidemiological data were obtained from the Open Data Portal of the General Directorate of Epidemiology (DGE) of the Mexican Ministry of Health. Environmental variables were derived from satellite-based remote sensing products accessed through Google Earth Engine, and sociodemographic indicators were obtained from the National Institute of Statistics and Geography (INEGI). The dataset integrates clinical variables (demographics, comorbidities, and hospitalization indicators), environmental predictors (temperature, precipitation, humidity, vegetation index, terrain and hydrological indicators), and territorial variables (population size, density, territorial area, and human exposure index). These variables were used as predictors in a machine learning feature importance analysis to identify the main determinants associated with dengue-related mortality risk across the 32 federal entities of Mexico. The dataset does not contain personal identifiers and was prepared exclusively for research and reproducibility purposes.