Modeling Covid-19 dynamics for real-time estimates and projections: an application to Albanian data

Erida Gjini

A big unknown is the initial number of imported cases from Italy to Albania during the month of the epidemic outbreak in Italy (8th of February until 8th of March), just prior to the first case detected in Albania. Using the data available from INSTAT (11) , about the yearly travel fluxes between Albania and Italy via air (1.7 million passengers) and by sea (1.6 million passengers), we obtain a total of 1.65 million arriving passengers on average from Italy to Albania in one year. This translates to 0.13 million passengers in one month. Thus, during the month prior to the first case in Albania, about 130.000 thousand passengers are expected to have arrived to Albania from Italy via air and sea. To compute the prevalence of infection expected in such passengers, we must account for the exponential disease dynamics in Italy during the same period. The prevalence of infection in travelers arriving to Albania from Italy can be assumed to be roughly the same as the prevalence of infection in the Italian population itself during this period (12) . I used the publicly available data for the epidemics in Italy (1), starting with 3 cases in February 8 and counting 5883 cases on the 8th of March, following an exponential growth at rate r = 0.25 (13).