VU Researchers Investigate the Pandemic’s Socio-economic Impact on Lithuania

The second wave of COVID-19 infection is more dangerous than the first, however, now we know more about the virus. In the spring of 2020, to assess the coronavirus situation in Lithuania, researchers of the Faculty of Mathematics and Informatics of Vilnius University (VU) together with medical specialists started providing the Government with short-term and long-term prognoses of spread of the epidemic and carried out research.

One of the main objectives of the research was to develop mathematical models of the COVID-19 epidemic based on data analysis to identify trends in the epidemic and assess the socio-economic impact of the epidemic in the context of Lithuania and other countries. VU researchers emphasise that the consequences of the first lockdown should be seen as a lesson in dealing with the second period of the lockdown.

Income inequality up, the most crisis-affected sectors named

To determine the income inequality during the lockdown period, the researchers looked into the data on the distribution of insured income provided by the state social insurance fund Sodra in the groups of insured persons and the groups of women and men who worked for at least 30 days a month.

“We have noticed that during the lockdown period there has been an increase the income inequality of all insured persons who worked for at least 30 days a month. In the summer months, when the lockdown restrictions were lifted, income inequality decreased and was the lowest in most age groups during the research period,” Assoc Prof Jurgita Markevičiūtė says.

The researchers used the Gini coefficient (or the Gini index), which shows inequality in income (but not wealth). It helped to assess income distribution, reflecting the gap between a fully equal and real distribution of population income. The researchers examined trends in the change of the Gini coefficient separately for men and women. They obtained different results but noticed that both of these groups were affected by the measures of the first lockdown.

In Lithuania, the first general COVID-19 pandemic lockdown was declared in mid-March and lasted for three months, affecting the movement and behaviour of the population as well as the economy. Due to the lockdown, educational institutions were closed, foreign and domestic travel was restricted, planned operations in medical institutions were postponed, and the activities of spas, hairdressers, barbers, beauty salons, accommodation service and other leisure activities were banned.

According to the researchers, the sectors most affected by lockdown restrictions include the manufacturing industry, food and beverage as well as accommodation sectors. “To determine the impact of the second lockdown, we will need newer economic data, which will be published by Statistics Lithuania quite soon,” Assoc Prof Markevičiūtė points out.

COVID-19 prognoses – the situation is set to improve in the nearest future

The COVID-19 prevalence prognoses prepared by scientists and updated on a daily basis allow for the planning of possible lockdown measures and identifying disease rates, and help assess developments for the next five days, decide on short-term constraints and plan hospital work and other activities. Official data provided by state institutions are used for statistical analysis, and the Government of the Republic of Lithuania and other institutions may take relevant decisions based on these prognoses to ensure the maximum security of citizens.

According to the data of January 25, in the near future the number of cases of coronavirus disease and the total 14-day rate of newly reported COVID-19 cases per 100 000 population will decline and is likely to fall below 500 cases per 100 000 population. This is the result of strict lockdown measures, and the number of cases may be slightly lower due to lower testing levels, although the proportion of positive tests has also decreased compared to December. In a week or two, the decrease in the number of patients is expected to bring a calmer period and reduce tension in hospitals,” Assoc Prof Markevičiūtė suggests.

Time series models are used to study short-term prognoses, which are also prepared based on similarities to other countries, i.e. by using cluster analysis. It is an integral method of data registration and analysis. By using algorithms based on neural networks, it makes possible to identify countries similar to Lithuania, to assess the trends of virus spread in different countries and to group them according to similarity.

“The analysis of individual clusters shows the specifics of virus spread and allows revising measures applied in the countries of the same cluster or in Lithuanian municipalities to stop the spread of the virus and assess the impact and effectiveness of the applied measures.

We narrow our focus and identify countries where the spread of the virus is similar to the spread of the virus in Lithuania and make prognoses based on their trends. This allows us to create and predict virus spread scenarios using the experience of other countries,” senior researcher Jolita Bernatavičienė supposes.

The deterministic models chosen for long-term modelling of epidemic spread not only picture the spread of the epidemic over the period of several months, but also, by adapting the relevant parameters, reflect the potential impact of various non-pharmaceutical interventions measures in flattening the curve, ensuring the proper functioning the healthcare system, reducing morbidity and mortality rates. To apply this model in determining the further course of the epidemic, it is necessary to take into account the number of people who have acquired immunity, the number of people vaccinated, and other factors.

Mathematician-developed models helpful for epidemiologists

VU Prof Alfredas Račkauskas says that infectious diseases have always been an important part of human history, as they are associated with great losses. “The Spanish flu epidemic in 1918-1919 killed more than 50 million people worldwide; annual seasonal flu epidemics cause 35,000 deaths worldwide, while the global death toll of COVID-19 has already topped 2 million people,” Prof Račkauskas says, highlighting the devastating impact of the Covid-19 pandemic.

When it comes to infectious diseases, epidemiologists have to play an important role in understanding the causes of an epidemic disease, predicting its course, developing methods for managing the disease and comparing them. “Mathematical models using computer simulations are useful for developing and testing various theories about complex biological systems related to the disease, assessing quantitative prognoses, determining the sensitivity of systems to changes in parameter values and evaluating the key epidemiological parameters of data,” Prof Račkauskas points out.

The researchers also intended to propose a methodology for the analysis of functional data to epidemiologists and to demonstrate its potential in analysing global and regional trends in the spread of COVID-19. When presenting the results of the research, Prof Račkauskas names the outcomes of the analysis of the main functional components – more than 90% of the variability of epidemiological parameters can be explained by three-dimensional vectors, which remarkably facilitates the research of such parameters.

„Moreover, a panel model was developed to compare the development of the pandemic in different regions of the world. It showed by what extent and how the development of the pandemic in Europe, America, Asia, Oceania differs from the general global trend,” Prof Račkauskas presumes.

The researchers also proposed a one-parameter model for comparing the two pandemic waves. This model can also be used to forecast the further development of the second wave. The researchers hope that the methodology of functional data analysis demonstrated in this research will be of interest to mathematical epidemiology specialists and provide a better understanding of the spread of various epidemics.

The research carried out by researchers of the VU Faculty of Mathematics and Informatics was funded by the Research Council of Lithuania, project ”COVID-19 infection in Lithuania: modelling and analysis of socio-economic consequences”.