What influences COVID-19 infection rates: A statistical approach to identify promising factors applied to infection data from Germany

Moritz Mercker; Uwe Betzin; Dennis Wilken

Although decay has recently been observed in the country-wide infection rates, Fig. 2 (a)-(f) suggests that there is a strong local variability in that decrease. To analyse the relative differences between different states, we again used appropriate GAMMs; corresponding results are presented in 3 (a)-(c). In a second step, we rescaled each β-value across all states (i.e., subtracting the mean and dividing by the standard deviation) to make the relative differences between states comparable with respect to β 1 , β 2 and β 3 . Corresponding results are given in 3 (d); the state-specific mean values from the rescaled β 1 , β 2 and β 3 thus represent the relative differences of recent infection rate changes between the different states. In particular, low values represent a strong decrease in infection rates, whereas high values represent a less strong decrease (respectively increase). The absolute values of the Y-axis in 3 (d) cannot be interpreted directly. Furthermore, confidence intervals in 3 (d) refer to the certainty of the mean value from the rescaled β 1 , β 2 and β 3 for each state. All rights reserved. No reuse allowed without permission. author/funder, who has granted medRxiv a license to display the preprint in perpetuity.