Estimate of influenza cases using generalized linear, additive and mixed models

Area of Science:

• Epidemiology
• Biostatistics
• Public Health

Background:

• Influenza surveillance is crucial for public health planning.
• Understanding influenza transmission patterns informs intervention strategies.

Purpose of the Study:

• To investigate the relationship between reported influenza cases and various covariates in Catalonia, Spain.
• To evaluate different statistical models for estimating influenza transmission dynamics.

Main Methods:

• Descriptive analysis of influenza cases reported between 2010-2014 from the SISAP program.
• Application of Generalized Linear Models (GLM), Generalized Additive Models (GAM), and Generalized Additive Mixed Models (GAMM).
• Calculation of incidence rate per 100,000 people, with seasonal comparisons.

Main Results:

• Influenza incidence was significantly higher in winter months (mean rate 13.75/100,000) compared to other periods (mean rate 3.38/100,000).
• Generalized Additive Mixed Models demonstrated superior adaptation to the temporal evolution of influenza, showing a serial correlation of 0.59.
• Additive models effectively estimated non-linear covariate effects, while mixed models accounted for data dependence and variability.

Conclusions:

• Generalized Additive Mixed Models provide a more accurate approach for analyzing and modeling influenza transmission over time.
• Accurate epidemiological modeling is essential for effective public health management of seasonal diseases like influenza.