Thread by @naval, Recent serology studies may be wrong. But if wrong, they‘re all wrong [...]

Recent serology studies may be wrong. But if wrong, they‘re all wrong *in the same direction.*

Finnish: 20-50x undercount
Scottish: 27-55x
Stanford: 50-85x
Italian: 30x
Mass: 17x
Germany: 0.37% CFR
Denmark: 0.21%

If true, COVID19 has much more spread and much lower fatality.

We are still very far from herd immunity.

And because it’s novel, it can still overwhelm the hospital system.

Sources:

https://www.helsinkitimes.fi/finland/finland-news/domestic/17561-thl-coronavirus-may-have-infected-dozens-of-times-more-in-finland.html

https://www.medrxiv.org/content/10.1101/2020.04.13.20060467v1.full.pdf

https://www.medrxiv.org/content/10.1101/2020.04.14.20062463v1

https://www.technologyreview.com/2020/04/09/999015/blood-tests-show-15-of-people-are-now-immune-to-covid-19-in-one-town-in-germany/

https://www.bostonglobe.com/2020/04/17/business/nearly-third-200-blood-samples-taken-chelsea-show-exposure-coronavirus/

https://covmodel.org/2020/04/17/sars-cov-2-preliminary-serology-test-reports-from-scotland-denmark-and-finland-give-contradictory-results-on-infection-fatality-ratio/

THL: Coronavirus may have infected dozens of times more than confirmed in Finland

The Finnish Institute for Health and Welfare (THL) on Wednesday published the results of its preliminary study of the prevalence of coronavirus antibodies in the population of Uusimaa.

https://www.helsinkitimes.fi/finland/finland-news/domestic/17561-thl-coronavirus-may-have-infected-dozens-of-times-more-in-finland.html

To be fair, all serology tests, if biased, would bias the same way. When the actual percent infected is low and population samples are small, false positives will have an outsized impact compared to false negatives. But none of these shows less than a 10x undercount.

https://twitter.com/tejlgaard/status/1252195899305979904?s=21 https://twitter.com/tejlgaard/status/1252195899305979904