Mathematical models are used to understand and predict the dynamics of epidemics, and to evaluate the effectiveness of different disease management measures, such as the quarantine of infected individuals. Some track the progress of the disease through individuals, but most publications modeling the COVID-19 epidemic use what is known as a ‘compartment’ or ‘SEIR’ model. This model tracks the number of individuals who are susceptible to disease but show no symptoms (incubating), those who are infected (show symptoms), and those who are recovering.
As well; These statistical models used to forecast the COVID-19 epidemic may not be sufficiently filed , according to academics at the University of East Anglia (UEA). Attention: this research has not yet been peer reviewed.
Its concussion stems from comparing a new approach to a published model of COVID-19 before isolation and social distancing measures were imposed. The comparison shows that the standard model underestimates the maximum infection rate by a factor of three, that is, it could be three times higher; on the other hand, the model substantially overestimates how long the epidemic will continue after reaching the peak of infections.
Weak points
The main weakness of statistical models to measure the progression of the coronavirus, according to these researchers, is that these models group all individuals in a compartment, ignoring the real time that passed since they were infected. In an article, Professor Alastair Grant, from the UEA Faculty of Environmental Sciences, argues that while SEIR and other models can predict to what extent the disease transmission rate must be reduced to stop the growth of a epidemic, they do a poor job of predicting its growth. In the words of the researchers, “they do not accurately capture the transient dynamics of epidemics.” But the available solutions are much more difficult to use than the SEIR model.
The proposed new approach, which was compared to the study, tracks the time since people were infected, while retaining the simplicity of a model that can be described in a few lines of code.
“If the SEIR models use estimated parameter values independently of the data, they underestimate the proportion of the population that will be infected at the peak of the epidemic,” according to Professor Grant, one of the study’s authors. “The paper shows how the Matrix models can provide a direct route to accurately model stage durations and therefore correctly reproduce epidemic dynamics. Our explicit time model shows that the peak of infection can be later than the peak in the simple SEIR model ”, he concludes.
Referencia:
Alastair Grant. ‘Dynamics of COVID‐19 epidemics: SEIR models underestimate peak infection rates and overestimate epidemic duration’, (2020). DOI: 10.1101/2020.04.02.20050674
