Forecasting COVID-19 pandemic in Poland according to government regulations and people behavior

The coronavirus disease 2019 (COVID-19) outbreak is a worldwide pandemic problem that started in China in December 2019 and within a few months spread to all continents. Very high infectivity of SARS-CoV-2 virus and substantial disease severity caused medical care capacity shortage in many countries. Therefore, real-time epidemic forecasting of the COVID-19 is useful to plan public health strategies like country lockdown and healthcare reorganization. We used extended susceptible-infected-removed (eSIR) model to predict the epidemic trend of COVID-19 in Poland under different scenarios of the lockdown and lockdown removal. We used time-series data of SARS-CoV-2 infection from March 4 to May 22 2020. Our forecast includes the impact of a timeline of preventive measures introduced in Poland. Using eSIR algorithm we estimated the basic reproductive number and a total number of infections under different epidemic trend scenarios. Using eSIR modeling we estimated that the basic reproductive number in Poland concerning different scenarios of the lockdown removal is in a range of 3.91-4.79. The lowest predicted number of infected cases would be 263 900 (0 - 1 734 200, 95%CI) if the strict protective measures were maintained until the end of September. However, under different scenarios of precautions removal, a total number of infected cases may exceed one million within the next year. Relatively early introduction of strong precautions in Poland significantly slowed down epidemic spread in Poland in comparison with other European countries like Italy or Spain. However, early removal of protective measures may result in a significant increase in infection. Data shows that the number of new COVID-19 cases in Poland beyond May 18 is linear what could be a prognosis of a duration of the epidemic exceeding 300 days.