Abstract
We report on the dynamic scaling of the diffusion growth phase of the COVID-19 epidemic in Europe. During this initial diffusion stage, the European countries implemented unprecedented mitigation polices to delay and suppress the disease contagion, although not in a uniform way or timing. Despite this diversity, we find that the reported fatality cases grow following a power law in all European countries we studied. The difference among countries is the value of the power-law exponent 3.5 < α < 8.0. This common attribute can prove a practical diagnostic tool, allowing reasonable predictions for the growth rate from very early data at a country level. We propose a model for the disease-causing interactions, based on a mechanism of human decisions and risk taking in interpersonal associations. The model describes the observed statistical distribution and contributes to the discussion on basic assumptions for homogeneous mixing or for a network perspective in epidemiological studies of COVID-19.