1/6 In Sweden, they claim that 26% of Stockholm will have been infected with coronavirus by May 1st. This is based on their own modeling. Flagging a few reasons why I remain skeptical (spider sense). https://www.nature.com/articles/d41586-020-01098-x

2/6 Their model calibration relies heavily on a survey of 738 people in Stockholm sampled between March 27 and April 3, 2020. 2.5% of them were PCR positive.

https://www.folkhalsomyndigheten.se/contentassets/2da059f90b90458d8454a04955d1697f/skattning-peakdag-antal-infekterade-covid-19-utbrottet-stockholms-lan-februari-april-2020.pdf

3/6 @AdamJKucharski has a very smart thread on why being PCR positive is not the same as being infectious. People can be PCR positive for weeks but not infectious. He describes how this could affect estimation. https://twitter.com/AdamJKucharski/status/1254084771535376391?s=20

4/6 From their model, they estimate that 98.5% of infections are going undetected (detecting 1 out of 67). This is actually down from earlier versions that considered 99%, 99.5%, or even 99.9% of infections are missed. This feels implausibly high. https://www.forbes.com/sites/davidnikel/2020/04/22/sweden-health-agency-withdraws-controversial-coronavirus-report/#2323b97e4349

5/6 They estimate that the total number of infectious people in Stockholm peaked on April 8th. So this would mean we would expect the number of new cases to decline over time. Keep an eye on this (figure shown for all of Sweden). I expect counts to continue to rise.

6/6 Finally, their model predicts that 26% of Stockholm residents will have been infected by May 1. That is only a week away. Compare this to recent estimates of 21% seroprevalence in NYC. Do we think Stockholm has had a comparable outbreak to NYC?

Conclusion: I'm skeptical. END