3 ms·

No, it's not telling us that. The antibody tests are designed with high specificity for particular antibodies, this study went out of their way to look for any

No, it's not telling us that. The antibody tests are designed with high specificity for particular antibodies, this study went out of their way to look for any reactivity.

lazide 6y ago
Designed for, but except in a very small number of cases not actually validated in the field to perform that way - and the specific antibodies may not be a useful indication of a useful immune response in the way we think it is. We do not have the actual evidence to demonstrate it.
- maxerickson 6y ago
  Which widely used antibody tests where not checked and validated against uninfected blood/individuals?
  - lazide 6y ago
    Pretty much any of those actually available to most people early on? [https://www.bloomberg.com/news/articles/2020-06-16/coronavirus-antibody-tests-are-suddenly-everywhere-and-confusing https://www.bloomberg.com/news/articles/2020-06-16/coronavir...] or even as of November 2020? [https://www.cdc.gov/coronavirus/2019-ncov/lab/resources/antibody-tests.html https://www.cdc.gov/coronavirus/2019-ncov/lab/resources/anti...] The emergency use authorizations allowed anyone who plausibly had a test to market it, even if it had not been independently validated in any way by anyone. Many folks just bought what they could find, even if it didn’t have an emergency use authorization. Combined with porous borders, e-commerce and the internet, lack of consequences for someone in China to ship junk here? We got buried in (at best) untested junk, and often outright scams. This has resulted in a rather predictable disaster when it comes to data and awareness - very few of the tests are being cross referenced to other data sources or tested against an independently verified data source, and even fewer (if any) are being evaluated by a independent party. The vast majority of tests in April have since been removed from authorization because they don’t work, don’t work well enough, etc. You can read a very nice paper from the FDA here summarizing it. [https://www.nejm.org/doi/full/10.1056/NEJMp2033687 https://www.nejm.org/doi/full/10.1056/NEJMp2033687] As far as I am aware, there is still no clear picture on which tests have actually been validated, by whom, and to what extent against what data set. Will they pick up variants, and if so which ones? With what accuracy? What are common field errors (collection problems, etc) that can lead to problems? What about manufacturing variance and supply chain issues?
    - maxerickson 6y ago
      So I used the sophisticated strategy of searching for the first test on https://www.fda.gov/medical-devices/coronavirus-disease-2019-covid-19-emergency-use-authorizations-medical-devices/eua-authorized-serology-test-performance https://www.fda.gov/medical-devices/coronavirus-disease-2019... and up pops https://jcm.asm.org/content/59/1/e02104-20 https://jcm.asm.org/content/59/1/e02104-20 (it's a study validating the sensitivity and specificity of the test). You've raised the bar a long ways from except in a very small number of cases not actually validated in the field to perform that way to a comprehensive analysis of all tests that have even maybe been available. Of course, the 'very small number of cases' makes it impossible to argue without doing a comprehensive study, but it's clear enough that the widely used commercial tests were developed with the idea that they should minimize false positives and tested against pre-Covid serum.
      - lazide 6y ago
        Thank you for digging that up! When I searched, I got a bunch of ads and junk from well before that study was run. Do you have any data on which tests are widely used? (As in by number of tests done?) I’m hoping it’s shifted the way you’re saying, but as of November there were more BS ones than legit ones, and it was hard to find a real validated test. None of the folks I knew getting antibody tests were able to find any of the validated ones. That validated tests are good is great if they are being used, or used widely. Because in November the ones being used widely were still junk, which was my point. If you have, per your study, 3 tests that are validated - are they used in the field for these studies? If so, objection withdrawn for current studies using them. If they are studies based on the older, unvalidated, and often junk tests - then that point still stands.
        maxerickson 6y ago
        I checked one test, using a non biased sampling method, and it contradicted the whole structure of your argument. Why would you expect anyone to continue the conversation on your terms?

No, it's not telling us that. The antibody tests are designed with high specificity for particular antibodies, this study went out of their way to look for any reactivity.

lazide 6y ago

Designed for, but except in a very small number of cases not actually validated in the field to perform that way - and the specific antibodies may not be a useful indication of a useful immune response in the way we think it is. We do not have the actual evidence to demonstrate it.

maxerickson 6y ago
Which widely used antibody tests where not checked and validated against uninfected blood/individuals?
- lazide 6y ago
  Pretty much any of those actually available to most people early on? [https://www.bloomberg.com/news/articles/2020-06-16/coronavirus-antibody-tests-are-suddenly-everywhere-and-confusing https://www.bloomberg.com/news/articles/2020-06-16/coronavir...] or even as of November 2020? [https://www.cdc.gov/coronavirus/2019-ncov/lab/resources/antibody-tests.html https://www.cdc.gov/coronavirus/2019-ncov/lab/resources/anti...] The emergency use authorizations allowed anyone who plausibly had a test to market it, even if it had not been independently validated in any way by anyone. Many folks just bought what they could find, even if it didn’t have an emergency use authorization. Combined with porous borders, e-commerce and the internet, lack of consequences for someone in China to ship junk here? We got buried in (at best) untested junk, and often outright scams. This has resulted in a rather predictable disaster when it comes to data and awareness - very few of the tests are being cross referenced to other data sources or tested against an independently verified data source, and even fewer (if any) are being evaluated by a independent party. The vast majority of tests in April have since been removed from authorization because they don’t work, don’t work well enough, etc. You can read a very nice paper from the FDA here summarizing it. [https://www.nejm.org/doi/full/10.1056/NEJMp2033687 https://www.nejm.org/doi/full/10.1056/NEJMp2033687] As far as I am aware, there is still no clear picture on which tests have actually been validated, by whom, and to what extent against what data set. Will they pick up variants, and if so which ones? With what accuracy? What are common field errors (collection problems, etc) that can lead to problems? What about manufacturing variance and supply chain issues?
  - maxerickson 6y ago
    So I used the sophisticated strategy of searching for the first test on https://www.fda.gov/medical-devices/coronavirus-disease-2019-covid-19-emergency-use-authorizations-medical-devices/eua-authorized-serology-test-performance https://www.fda.gov/medical-devices/coronavirus-disease-2019... and up pops https://jcm.asm.org/content/59/1/e02104-20 https://jcm.asm.org/content/59/1/e02104-20 (it's a study validating the sensitivity and specificity of the test). You've raised the bar a long ways from except in a very small number of cases not actually validated in the field to perform that way to a comprehensive analysis of all tests that have even maybe been available. Of course, the 'very small number of cases' makes it impossible to argue without doing a comprehensive study, but it's clear enough that the widely used commercial tests were developed with the idea that they should minimize false positives and tested against pre-Covid serum.
    - lazide 6y ago
      Thank you for digging that up! When I searched, I got a bunch of ads and junk from well before that study was run. Do you have any data on which tests are widely used? (As in by number of tests done?) I’m hoping it’s shifted the way you’re saying, but as of November there were more BS ones than legit ones, and it was hard to find a real validated test. None of the folks I knew getting antibody tests were able to find any of the validated ones. That validated tests are good is great if they are being used, or used widely. Because in November the ones being used widely were still junk, which was my point. If you have, per your study, 3 tests that are validated - are they used in the field for these studies? If so, objection withdrawn for current studies using them. If they are studies based on the older, unvalidated, and often junk tests - then that point still stands.
      - maxerickson 6y ago
        I checked one test, using a non biased sampling method, and it contradicted the whole structure of your argument. Why would you expect anyone to continue the conversation on your terms?