A modeling framework for the analysis of the SARS-CoV2 transmission dynamics

Area of Science:

• Epidemiology
• Mathematical Biology
• Biostatistics

Background:

• The true burden of SARS-CoV-2 infections is underestimated due to case under-ascertainment.
• Reported deaths offer a more reliable metric for inferring pandemic dynamics.
• Understanding infection burden is vital for effective public health strategies.

Purpose of the Study:

• To develop a flexible Bayesian framework to estimate the true cumulative number of SARS-CoV-2 infections.
• To provide a more accurate estimation of the time-varying reproduction number ($$).
• To assess the impact of mobility and testing on inferred epidemiological quantities.

Main Methods:

• Utilized reported deaths to infer total infections, accounting for age distribution and probability of death.
• Employed a continuous-time transmission model with a diffusion process for the transmission rate.
• Developed a Bayesian tool in Stan for parameter estimation and uncertainty quantification.
• Analyzed data from three pairs of European countries.

Main Results:

• Estimated the true cumulative number of SARS-CoV-2 infections for selected European countries.
• Provided more accurate estimates of the time-varying reproduction number ($$).
• Inferred the daily reporting ratio, offering insights into testing and surveillance effectiveness.

Conclusions:

• The Bayesian framework effectively estimates SARS-CoV-2 burden using reported deaths.
• Accurate estimation of $$ is achievable by accounting for true infection numbers.
• Changes in mobility and testing significantly influence inferred epidemiological parameters.