We've submitted a manuscript to F1000 Research for peer-review:

“Bayes Lines Tool (BLT)-A SQL script for analyzing diagnostic test results with an application to SARS-CoV-2-testing”

Graph: Relation of True Positives VS True Negatives, Germany, 10.01.2021

#UnbiasedScience

2/: The open source SQL Code & example Excel file is available at: https://bayeslines.org 

The manuscript is available on a preprint-server:
https://zenodo.org/record/4459271#.YAwjxhYxk2w

Note: The paper has not been accepted yet by the journal. We are allowed to discuss it via preprint-servers.

3/: Our Bayes Line Tool is able to back-solve disease-prevalence, test-sensitivity, test-specificity, thus: true positive (TP), false positive (FP), true negative (TN) & false negative (FN) numbers from official governmental test outcome reports.

GER
10.01.2021
TN/FN

4/ Confusion matrices with four variables are created: TP, FP, TN, FN. The matrices are calculated as follows: we need prevalence, specificity, and sensitivity as well as the number of people that got tested (within a given period) and the number of positives.

5/: The number of positives & the number of tests are official government numbers. Prevalence, specificity, and sensitivity are unknown. Thus, we assume any combination of them ranging from 0-99%. These three combinations can amount up to millions of combinations.

6/: The tool calculates 7 million combinations. Of these 7 million combinations, only 1-100 usually match the government numbers (e.g. TP + FP = amount of performed tests).

Germany
Aug-Dec 2020
Positives Reported VS Matching Confusion Matrices for each daily report

7/: For further in-depth overview of the methodology focusing on The Netherlands and the United Kingdom, head over to Dr. Simon’s ( @goddeketal) thread: https://twitter.com/goddeketal/status/1353297941499424769

8/: In layman’s terms: we show posterior probability, independent of the circumstances in which the diagnostic tests were performed. We seek highest possible accessibility of the tool for the scientific-, health professionals- & policy makers-community:
https://bayeslines.org/#download 

9/: The Bayes Line Tool, its SQL-code & query, allows for the first time to display a real estimation of the #SARSCoV2 situation VS the background of testing volume and the actual quality and will help in future to develop reasonable test strategies & measures.

10/: We ask the community for support with data collection. We know that there are enough data-nerds around on twitter-verse, who can provide relevant data-spreadsheets of their respective countries and regions including daily testing numbers, daily positive tests.

11/: This tool will help to obtain a better basis for planning concerning interventions, it also allows individuals who got tested, to use the confusion matrices as an aid for interpreting their test results in view of the population they were tested in.

12/: I want to specifically thank @waukema, his courage, scientific integrity and his genius. https://twitter.com/waukema/status/1353296044986159105

13/: I encourage you to join our #UnbiasedScience Telegram channel, where we will provide unbiased information on a regular basis. https://t.me/unbiasedscience 

14/: I want to thank all the other authors for their contributions. We thank Michiel Maandag ( @michielmaandag) for bringing down-to-earth counterweight and alignment to the team.

15/: https://threadreaderapp.com/thread/1353302956314124293?refresh=1611489747

Addendum a): There is a typo in the first tweet:

I meant:
Graph: Ratio / Relation of True Positives VS False Positives, Germany, 10.01.2021

Sorry for that. https://twitter.com/Bobby_Network/status/1353302956314124293

Addendum b): Typo: UK is meant, not Germany: https://twitter.com/Bobby_Network/status/1353304065728524288

Summary:
- high rates of False Positives
- prevalence estimations are overrated
- official test-specificity is exaggerrated
I assume: Problems occur in mass-testing scenarios, align w/ latest updates by @WHO.
Code: https://bayeslines.org 

@naomirwolf

https://twitter.com/Kevin_McKernan/status/1352129155194880002