04-18-2020, 03:16 PM
(04-18-2020, 02:35 PM)MidTowner Wrote: I found this interesting: Santa Clara County COVID-19 cases could be 50 to 85 times higher than reported, Stanford study finds
Santa Clara County officially had 956 confirmed cases as of April 1, and 1870 as of yesterday. Researchers, based on three thousand blood tests for antibodies, estimate that between 2.49% and 4.16% of the population had been infected, making the real count up to 85 times higher than the official one.
On the other hand, Austria did random antibody testing of 1544 people, and found that only 0.33% (5 of 1544) had the antibody markers present. Based on the likely margin of error, they estimate that less than 1% of the population had been infected.
https://www.theguardian.com/world/2020/a...tudy-shows
Why such a big difference? Are the infection rates really that far apart? Or is it a difference in the testing technique or methodology? Each antibody test has a different level of sensitivity (percentage of actual antibody cases that are detected) and specificity (percentage of non-cases that are rejected). Typically both of these numbers are somewhere in the 90% range for approved tests, meaning up to 10% of false negatives and 10% of false positives. See here for data on available tests:
https://www.centerforhealthsecurity.org/...ID-19.html
Abbot claims to have developed a new one with nearly 100% sensitivity and specificity. Haven't seen any external validation of it yet.
https://www.kiro7.com/news/local/breakth...VSEVRRPYQ/