Data
The statistical part of this manuscript is based on daily estimates of confirmed COVID 19 cases in Germany using the dataset available on https://npgeo-corona-npgeo-de.hub.arcgis.com. The data analyzed cover a timeframe from 02. March to 31 May 2020. Due to the daily updates from local health authorities the numbers are sometimes adjusted retrospectively. I have made my last update of the dataset on July 27 and the numbers for the period mentioned are stable in the meantime. For most calculations, the “disease onset date” variable from that dataset was used, which is thought to be less affected by administrative reporting delays than the “report date”. Only where no clinical information on date of symptom onset is available, the RKI substitutes by the date of report. Data on daily mobility tracking based on mobile phone data were manually transcribed from the “Mobility Monitor”, published by the working group on their homepage “http//covid-19-mobility.org”. The time series presented with the mobility monitor starts March 12, the period before that date was set to “100 percent” mobility. This should be acceptable, since the data are presented as percentage deviation from baseline and the official reports published by the working group show that mobility was at previous year levels before that date. Information on PCR test activities per calendar week was derived from the regular publications issued by the RKI on that topic within the institutes regular journal “Epidemiological Bulletin” . Every Wednesday cumulative data on test activities are presented within that bulletin, with a break-out by calendar week starting from week 11 and summarizing all reported tests up to week 10. The numbers used were first published with edition 23 on 04 June 2020. Data for CW 10 were included with half of the test recorded up to that week in absence of a more precise approximation.