A Framework for a Statistical Characterization of Epidemic Cycles: COVID-19 Case Study

Area of Science:

• Epidemiology
• Mathematical Modeling
• Public Health

Background:

• COVID-19 pandemic necessitates understanding local transmission dynamics for effective control.
• Existing modeling approaches aim to predict epidemic behavior but require refinement.

Purpose of the Study:

• To present a framework for characterizing variables driving COVID-19 epidemic cycles.
• To provide statistical tools for local health authorities to support decision-making.

Main Methods:

• Normalized comparison of epidemic cycles across different locations and population sizes.
• Derivation of a reproduction number model, accounting for underreporting (subnotification).
• Application of logistic and inventory models to determine actual infections, incubation periods, and cycle onset.

Main Results:

• Identified a consistent triangular pattern in epidemic cycles, enabling duration prediction.
• Estimated effective reproduction number (R_t) and subnotification effects (<5% error).
• Determined probable epidemic onset dates, average incubation period (approx. 5 days), and total infections for Germany, Italy, and Sweden.

Conclusions:

• Relatively simple mathematical tools can reliably understand COVID-19 local epidemic cycles.
• The integrated framework effectively identifies patterns and calculates key drivers: R_t, subnotification impact, onset dates, total infections, and incubation periods.