Outbreak detection algorithms based on generalized linear model: a review with new practical examples.

Area of Science:

• Public Health
• Epidemiology
• Biostatistics

Background:

• Public health surveillance is vital for monitoring and detecting infectious diseases.
• Outbreak detection algorithms are increasingly important, especially after the COVID-19 pandemic.
• Generalized linear models (GLMs) have seen significant advancements in surveillance applications.

Purpose of the Study:

• To present and compare generalized linear model (GLM)-based outbreak detection methods.
• To provide a historical overview of GLM family algorithms in surveillance.
• To demonstrate GLM applications using real-world Measles and COVID-19 data.

Main Methods:

• Review of existing literature on GLM-based outbreak detection.
• Comparative analysis of different GLM techniques.
• Application of selected GLM methods to Measles and COVID-19 datasets.

Main Results:

• A comprehensive overview of commonly used GLM-based outbreak detection algorithms.
• Demonstration of the practical application and comparison of these methods.
• Highlighting the utility of GLMs for both theoretical and practical surveillance research.

Conclusions:

• Generalized linear models offer a robust framework for infectious disease outbreak detection.
• This study provides a valuable resource for researchers and health managers seeking to understand and implement GLM-based surveillance tools.
• The comparative analysis facilitates informed selection of appropriate methods for public health challenges.