Predictive models for hospitalization and mortality in dengue using SINAN data: study protocol for development, temporal validation, and performance evaluation

The study focuses on the development and validation of predictive models for dengue using the Brazilian SINAN system. Two independent models will be created: one to predict hospitalization among reported dengue cases and another to predict mortality among hospitalized dengue patients. The study will be retrospective and will use data from 2017 to 2025. The protocol will follow TRIPOD+AI guidelines and include pre-defined predictors, handling of missing data, comparison of regression and machine learning algorithms, and internal and temporal validation. The aim is to create a transparent and reproducible analytical protocol to support the development of risk models that could be applied in clinical practice and epidemiological surveillance for dengue.