Analysing and comparing the COVID-19 data: The closed cases of Hubei and South Korea, the dark March in Europe, the beginning of the outbreak in South America

Stefano De Leo; Gabriel Gulak Maia; Leonardo Solidoro

Specifically for Italy, the virus seems to be more widespread in the Lombardia region, which is also the highest populated region in the country, see Fig. 6 . Looking at Fig. 7(a) , we can see that the Italian regions have already reached a horizontal development stage, with the α-factors of Lombardia, Emilia-Romagna, and Veneto starting to drop, while Piemonte's curve is flattening out, indicating a linear growth of the spreading rate. In the same figure, we also find the curve for the rest of Italy where the maximum of α factor is below 50, a factor five times smaller than the Lombardia maximum. In Fig. 7(b) , we find a more detailed look at the α-factor, ranging from 0 to 500 TCCpM. We can see that, after the range from 0 to 200, Piemonte and Lombardia's α-factors start to increase compared to the other Italian regions. In this case, with the same measures adopted, the difference in the future spreading rate comes from the recovered over confirmed ratio. This ratio is plotted in Fig. 8 . We can see the recovering rate increasing, specially for Lombardia, with a more than 25% recovery rate at day 72 (2 April). The Piemonte region started with a very flat and close to zero recovery rate. This means that the measures for this region were applied when the outbreak was in its initial phase. This explains why, notwithstanding the high α factor near 500 TCCpM, Piemonte stabilises below Lombardia and Emilia-Rommagna, where the measures were taken when the outbreak was already ongoing.