Questions concerning Science-Blog-6

[Mtagxx]

Your Question

• Website URL Path: https://www.coronawarn.app/de/science/2022-11-23-science-blog-6/

1. Question
   It would be useful, if in Science-Blog-6 the number of CWAs, which show a red warning would be given. Nevertheless the fraction of CWAs which turn "red" on a specific day can be estimated roughly from the number of red warnings from "Datenspende" and assuming that 15.4 million CWAs participate in "Datenspende". For example on 10. July 2022 this would result in 3.4 % and on 1. September 2022 in 1.06 % CWAs turning "red". Assuming 30 million CWAs this would be 1 million CWAs with a fresh red warning on July 10th or 320000 on September 1st. Is this estimation approximately right?

2. Question
   On 10. July 2022 (in brackets: data from 1. September 2022) there were about 170000 (120000) RAT-tests. Assuming 35 % CWA usage and 3.4 % (1.06 %) of it becoming "red" (see first question) then there were about 2020 (445) CWA users with a red warning, who got RAT-test that day. According to "Abb. 1" in Science-Blog-6 the positive rate for red warnings is 22 % (16 %), otherwise the positive rate is 15 % (5.5 %). So there were about 6 % (10.5 %) more or 120 (47) additional positives discovered related to a red warning of CWA. Compared to 25000 (6600) positive RAT-tests that day, this amounts to approximately 0.5 % (0.7 %) more positives, which may have been discovered with the help of CWA. Is this the order of magnitude of the actual efficacy of the CWA?
   On the other hand there had been 1 million (320000) CWAs with a red warning that day, but estimated only 2020 (445) got tested. Thus only one from 500 (720) CWA users with a red warning got an official RAT. Is this more or less correct?

3. Question
   According to RKI the serial interval for the Covid-19 omicron variant is 2.2 days, but in Science-Blog-6 risk contacts up to 10 days back are considered. Wouldn't a much shorter interval, like for example 4 days, show a much better correlation between infection and red warning and thus a higher positive rate?

[dsarkar]

@Mtagxx Thanks for your question, we will forward this and hopefully come back with some feedback soon.

[Mtagxx]

Any new findings or insights already?
Or are there answers to some of the questions at least?

I forgot to mention that the numbers are the "7-day moving average" values taken from
CWA-dashboard (https://www.coronawarn.app/de/analysis/ ).