Statistical Models Addressing Acute Respiratory Diseases: A Scoping Review

Area of Science:

• Public Health
• Epidemiology
• Biostatistics

Background:

• Respiratory diseases pose a significant global health burden.
• Effective health management requires robust tools for surveillance and prediction.
• Statistical models offer powerful approaches to address these challenges.

Purpose of the Study:

• To conduct a scoping review on the application of statistical models in respiratory disease management.
• To synthesize findings on how statistical models are used for demand forecasting, epidemiological analysis, and risk/prognosis prediction.
• To identify best practices for utilizing statistical models in healthcare settings.

Main Methods:

• A comprehensive scoping review following Joanna Briggs Institute methodology.
• Searches conducted across multiple databases (Medline, Web of Science, LILACS, BVS, CINAHL, EBSCO, EMBASE) from August 2024 to June 2025.
• Inclusion of articles, theses, dissertations, and technical reports (2014-2025) focusing on statistical models for acute respiratory diseases.

Main Results:

• Twenty-nine studies were analyzed, categorized into demand forecasting (e.g., SEIR models), epidemiological analysis (e.g., time series, Bayesian models), and risk/prognosis (e.g., machine learning).
• Models were used for pandemic evolution estimation, case/death prediction, resource allocation, transmission pattern inference, and clinical prediction.
• Key findings emphasize combining diverse methods and data sources, adapting models to local contexts.

Conclusions:

• Statistical models are vital for effective respiratory disease surveillance, prognosis, and decision-making.
• Integrating traditional and modern statistical approaches with local data is essential for dynamic health management.
• Future efforts should focus on model adaptation and multi-source data integration for improved public health outcomes.