Delay-differential SEIR modeling for improved modelling of infection dynamics

Area of Science:

• Epidemiology
• Mathematical Biology
• Computational Science

Background:

• Classic Susceptible-Exposed-Infected-Recovered (SEIR) models often use mass-action laws, limiting their ability to reproduce observable infection dynamics.
• Limitations include difficulty in accurately representing incubation periods and disease symptom progression.

Purpose of the Study:

• To propose a new approach for simulating epidemic dynamics using differential equations with time delays and instant transitions.
• To more accurately approximate transition process durations and enhance model parameter clarity.
• To apply this novel approach to model the COVID-19 pandemic in Germany and France.

Main Methods:

• Developed a delay-based modeling approach utilizing a system of differential equations with time delays.
• Incorporated factors such as testing, symptom progression, vaccination, immunity duration, and virus strains.
• Utilized the stringency index to characterize non-pharmaceutical government interventions.
• Performed parameter identifiability analysis.

Main Results:

• The proposed approach allows for more accurate simulation of disease progression, including incubation periods and symptom severity.
• Parameter identifiability analysis showed a significant reduction in the number of parameters and improved their identifiability.
• Successfully developed and applied delay-based models for the COVID-19 pandemic in Germany and France.

Conclusions:

• The novel delay-based modeling approach offers a more accurate and interpretable method for studying infectious diseases compared to traditional SEIR models.
• This approach provides a flexible framework applicable to various infectious diseases beyond COVID-19.
• Publicly available models facilitate further research and application in public health.