We present a novel methodology in order to perform the epidemic risk assessment in terms of different factors which are useful to understand the different impact of an epidemic in different areas of a country. In particular we discuss the case of COVID-19 outbreak in Italy.

We characterize each region of Italy by considering the available data on air pollution, mobility, winter temperature, housing concentration, health care density, total and aged population. We find that the epidemic risk is higher in some of the northern regions of Italy with respect to central and southern part.

Our epidemic risk index shows strong correlations with the available official data of the COVID-19 outbreak in Italy and explain in particular why regions like Lombardia, Emilia-Romagna, Piemonte and Veneto are suffering much more than the rest of the country in terms of infected cases, intensive care units and deceased patients. Although the COVID-19 outbreak started almost in the same period, at the beginning of 2020, in both north (Lombardia and Veneto) and central part of Italy (Lazio), when the first infected were officially certified, the outbreak has been more diffuse and lethal only in those regions with higher epidemic risk.

Due to the fact that a large part of infected people do not show evident symptoms of the disease, we claim that in the regions with a lower epidemic risk, i.e. the central and south part of Italy, the epidemic, although probably already very diffused, should not be as lethal as in the northern part of Italy due to a different a-priori risk exposure. We also discuss some policy implications directly connected with our methodology, which results to be very flexible and adaptable to different set of epidemic historical data and to different countries.

A. Pluchino, G. Inturri, A. Rapisarda, A. E. Biondo, R. Le Moli, C. Zappala’, N. Giuffrida, G. Russo, V. Latora, A Novel Methodology for Epidemic Risk Assessment: the case of COVID-19 outbreak in Italy [submitted]