Bayesian mechanistic model of COVID-19 transmission

dynamics including the effect of vaccination

Javier Blecua, Juan Fernandez-Recio

Instituto de Ciencias de la Vid y del Vino (ICVV), CSIC-Universidad de La Rioja-Gobierno de La Rioja

Brief description:

A new mechanistic model that describes the transmission dynamics of COVID-19 was applied initially to estimate the effect of the non-pharmacological measures by means of Bayesian analysis methods. The model was extended to include also the impact of the vaccination and of the different virus variants.

The consistent results obtained for a total of 30 European countries, with data following very different patterns, confirmed that the model used is an appropriate method for describing the present and future evolution of the disease.

Keywords:

COVID-19 modelling, non-pharmacological measures, vaccination, virus variants, mechanistic model, Bayesian analysis.

Abstract:

COVID-19 (coronavirus disease 2019) is a pandemic disease caused by a new type of coronavirus called SARS-CoV-2, which has caused unprecedented medical, economic and social burden worldwide. A variety of models to describe the transmission dynamics of the virus and the impact of non-pharmacological measures have been reported. Among them, Bayesian mechanistic models using MCMC optimization have shown good description of the transmission dynamics and have potential for accurate predictions of future evolution of the pandemics.

In our group, we have been working on optimizing a previously reported COVID-19 transmission model, which has been extended here for the analysis of multiple periods of different transmission rates, enabling the inclusion of an arbitrary number of non-pharmacological measures. Additionally, the model has been extended to include the effect of vaccination and the impact of the different virus variants on the transmission dynamics.

The algorithm computes the evolution of the daily number of infections by fitting a SEIR model to the observed daily deaths, in a Bayesian framework, using MCMC optimization to obtain the a posteriori distribution for the parameters that best describe the impact on the transmission rate of each intervention measure. The model captures the positive impact of the vaccination on the evolution of the disease, takes into account the immunity of the recovered population and considers specific transmission parameters for the different virus variants.

The model was successfully applied to a total of 30 European countries, obtaining good fit results and conclusions related to the impact of the different interventions that were consistent with results from other studies.

Interestingly, the model also estimates the percentage of immune population required to reach the herd immunity in the different countries, which is a valuable tool to understand the evolution of the pandemics on the long term and help in future worldwide control strategies.

Methodology and results:

All details, including the description of the methodology and the results and conclusions obtained from the application of this model to data of a total of 30 European countries, will be described in a paper, currently under preparation.

The access to the results is already possible using the links in the navigation panel located on the left side of this web page.

The values that quantify the impact of different intervention measures, presented as numeric values in tables and graphics, represent the factor with which each intervention contributes to changes in the value of the reproduction number (R_t). Values less than 1 correspond to effective measures that have contributed to a reduction of the transmission rate, while values greater than 1 are associated with periods or events that have led to an increase in the reproduction number (R_t) and thus in the number of infections.

References:

Blecua, J., Fernandez-Recio, J. (2025) Bayesian mechanistic model of COVID-19 transmission dynamics including the effect of vaccination (currently under preparation)

Blecua, J., Fernandez-Recio, J., Gutierrez, J.M. (2024) A Probabilistic Description of the Impact of Vaccine-Induced Immunity in the Dynamics of COVID-19 Transmission. Open Journal of Modelling and Simulation, 12, 59-73. doi: 10.4236/ojmsi.2024.122004.

Blecua, J. (2021) Optimisation of a COVID-19 transmission model for its application to multiple intervention periods (Master Thesis, Universitat Oberta de Catalunya)

Fernandez-Recio, J. (2020) Modelling the Evolution of COVID-19 in High-Incidence European Countries and Regions: Estimated Number of Infections and Impact of Past and Future Intervention Measures. 2020 J. Clin. Med. 9, no. 6, 1825

Flaxman, S., Mishra, S., Gandy, A. et al. (2020) Estimating the effects of non-pharmaceutical interventions on COVID-19 in Europe. 2020 Nature 584, pp. 257-261