St Petersburg University mathematician: ‘New waves of the pandemic are inevitable until the number of people immune to the virus has increased significantly’

‘To reduce epidemic peaks and the infection rate, vaccination is required. This strategy for disease containment is being successfully implemented, as we can see from the example of many European countries. Currently, our vaccination rates are much lower than, for instance, in Great Britain and Hungary. These countries crossed the 12% mark in percentage of the population who have been vaccinated as early as in mid-April,’ said Professor Zakharov.

Mathematicians from St Petersburg University predict that the next wave of COVID-19 in Russia is likely peak by the end of June. The increase in daily incidence may reach over 21,000 people per day, with the simultaneous number of COVID-19 patients soaring to about 420,000 people.

In February 2021, the experts published a forecast predicting a decrease in COVID-19 incidence rate in Russia in the second ten days of April, and in St Petersburg – after May 10. These predictions took into account the assumption that vaccination would cover a significant part of the population susceptible to the virus, and susceptible people would continue to comply with the face mask rules. The forecast, however, proved to be overly optimistic and the prediction failed. According to the experts, to date, the number of people who gained antibody immunity to SARS-CoV-2 has grown insignificantly. In fact, this value does not affect the forecast – as if vaccination has not been carried out at all.

A paper by Victor Zakharov and Yulia Balykina on the case-based rate reasoning (CBRR) approach and predicting the dynamics of the coronavirus (COVID-19) pandemic based on a precedent model was published in the scientific journal ‘Vestnik of St Petersburg University. Applied mathematics. Computer science. Control Processes’, volume 16, issue 3, 2020.

A new mathematical CBRR model for the short-term forecasting of the COVID-19 pandemic spread under uncertainty was built by experts from the Intelligent Logistics Centre at St Petersburg University. The model is based on data from countries where the disease began to spread earlier. The description of the model and the simulation results were published in the prestigious international journal Mathematics. St Petersburg University mathematicians analysed what was happening in these countries using estimates of the four principal parameters affecting the pandemic spread: the total number of confirmed cases; the number of recovered individuals; the number of deaths; and the number of currently active cases.

Forecasts are made for four to five weeks periods, employing constantly updated real-time data. Importantly, data about the dynamics of the COVID-19 pandemic for the previously analysed period allows for more accurate disease forecasting for a future time period. The intelligent algorithms embedded in the model help to determine future values for the number of new cases, the total number of cases, and the number of active cases.

At the moment, the mathematical model developed by St Petersburg University experts allows for making fairly accurate predictions. The deviation of the predicted total number of confirmed cases from the actual one over a short-term time horizon is within 1.5%. Note that the error inherent in the first version of the model was about 10%. Researchers at the Intelligent Logistics Centre are currently developing a new model with improved counting algorithms to provide even greater accuracy.