Antibody test?

[antipodean]

The WHO is a joke. It's not their fault. They are severely underfunded and understaffed. No major government actually pays attention to what comes of their offices in Geneva. None. I repeat, it's not their fault. They are trying to remain relevant and issue a statement once in awhile but their credibility is pretty much non-existent. Hopefully, in the future we can set up a beefed up WHO that really matters. But that will cost us $$$.

I agree with this - their failing is not being able to exert influence effectively. They are like a dog yapping outside your window, easier to ignore or throw it a bone to shut it up.

I have worked for an organization like this before and know first hand what it’s like trying to get parties to listen, let alone agree when they all have different political agendas on the home front. It is especially harder if you have a leader who lacks the deep relationships behind the scenes or ability to speak in plain terms and get people’s attention.

 

[easyliving]

It is very important when discussing antibody tests to understand there are false positives. This is very dependent on the prevalence of the disease in the population. So if we assume that the number of people infected in Argentina is 200x more than reported or say 3% of the population and the antibody test is 99% accurate, the probability of a false positive is 25%. If the prevalence is only 1%, then it's 50%.
So if you have an antibody test and think you have immunity, be aware, you might not just because of testing errors.

 

[TomAtAlki]

Is there any testing, antibody or disease, going on in Buenos Aires ?

 

[antipodean]

View this content on Instagram

View this content on Instagram

This guy describes his experience with two different tests and two different results.

Last night the President and Mayor of CABA said testing was primarily being focused on vulnerable communities and rest homes - no wider programs exist yet. Testing in the healthcare system is recommended / required for anyone presenting with fever and at least one other COVID symptoms.

 

[easyliving]

This is exactly the situation I was alluding to. Even great tests (and many are not even close) have some degree of error. Because of both false positives and false negatives, the test results need to be looked at within the context of the overall prevalence of the disease. In his case he needs another test for sure to get any degree of confidence but it still won't be 100% sure (very little in life is).

If anyone is curious about what the hell I am talking about, it all relates to Bayes theorem. I found this handy calculator where you can plug in the sensitivity of a test (true positive rate where .05 is 5%) and the specificity (true negative rate) as well as the prevalence in the population you are testing. It pops out the result for you of false positives and negatives. http://vassarstats.net/clin2.html

Someone did a nice graphic of what is going on here:

https://twitter.com/x/status/1249024017668780032

 

[another]

This is exactly the situation I was alluding to. Even great tests (and many are not even close) have some degree of error. Because of both false positives and false negatives, the test results need to be looked at within the context of the overall prevalence of the disease. In his case he needs another test for sure to get any degree of confidence but it still won't be 100% sure (very little in life is).

If anyone is curious about what the hell I am talking about, it all relates to Bayes theorem. I found this handy calculator where you can plug in the sensitivity of a test (true positive rate where .05 is 5%) and the specificity (true negative rate) as well as the prevalence in the population you are testing. It pops out the result for you of false positives and negatives. http://vassarstats.net/clin2.html

Just for the fun of it, want to calculate the percentage of people here who'd understand Bayes' theorem? Based on how many math graduates you think are on this forum and, given the average age of participants, how many of those still remember what we were taught at Statistics course?

 

[easyliving]

Are you suggesting we are not all expert virologists, epidemiologists, statisticians, economists and epistomologists?

I was only giving the reason behind my comments in case they sounded unhinged and then I gave a link for a plug and chug answer...

Back to swapping recipes...

 

[another]

Are you suggesting we are not all expert virologists, epidemiologists, statisticians, economists and epistomologists?

I was only giving the reason behind my comments in case they sounded unhinged and then I gave a link for a plug and chug answer...

Back to swapping recipes...

ooops, i thought i was subtle enough at that. made me nostalgic, it was by far my favorite class...the 3 actual experts in epidemiology and and mathematical statistics interviewed in the videos I posted in the coronavirus junkies thread are firmly behind you on the subject and it's corollaries. But such is the human nature that the more solid arguments are made against one's belief, the stronger one clings to that belief. so we should probably stick to swapping recipes lol...

 

[gracielle]

Coronavirus: son defectuosos los tests rápidos que usó el Gobierno

Así lo determinó un estudio del Ministerio de Ciencia y el Conicet; las pruebas no miden bien la circulación del virus

www.lanacion.com.ar

10 May 2020
Coronavirus: the rapid tests used by the Government are defective

Coronavirus: son defectuosos los tests rápidos que usó el Gobierno

Así lo determinó un estudio del Ministerio de Ciencia y el Conicet; las pruebas no miden bien la circulación del virus

translate.google.com

The rapid tests to monitor the circulation of the new coronavirus which were advised to buy by the national health authorities are defective: detect only "with a band of dim light" the presence of only one of the two antibodies against Covid-19 for which they were designed. A failure that could epidemiologically underestimate the circulation of the virus in the population, which is the main objective of these tests. If they were reliable, they would allow the intensity of the quarantine to be regulated.

In the Casa Rosada they recognize that the tests of the Chinese firm Zhuhai Livzon Diagnostics, which were already used in the train stations of Constitución, Retiro and Once, were not what was expected. They reported that an investigation by the Covid-19 Unit, which reports to the Ministry of Science and Technology and Conicet, "did not recommend its use" for one of the two antibodies. Regarding the second antibody, the one with the later appearance, he warns that the test presents problems for its reading....
So far 1,200 of the 170,000 rapid tests have been used in this population monitoring, with only eight positive cases, that is, just 0.66%, according to official information. The specialists consulted by LA NACION agree that it is an extremely low proportion for the metropolitan region with the highest viral circulation in the country.
...."The Covid-19 Unit carried out trials on the tests, with the following conclusions: weak detection of IgM. USE IS NOT RECOMMENDED," the official evaluation determined. For IgG, 80% of the positive samples were detected with the strips. by (the technique) Elisa. But it is important to highlight that 50% of these detections gave very faint bands. You must wait ten minutes to make the reading, that it be done in good light and consider it dim as positive. "...

 

[Redpossum]

Coronavirus: son defectuosos los tests rápidos que usó el Gobierno

Así lo determinó un estudio del Ministerio de Ciencia y el Conicet; las pruebas no miden bien la circulación del virus

www.lanacion.com.ar

10 May 2020
Coronavirus: the rapid tests used by the Government are defective

Coronavirus: son defectuosos los tests rápidos que usó el Gobierno

Así lo determinó un estudio del Ministerio de Ciencia y el Conicet; las pruebas no miden bien la circulación del virus

translate.google.com

The rapid tests to monitor the circulation of the new coronavirus which were advised to buy by the national health authorities are defective: detect only "with a band of dim light" the presence of only one of the two antibodies against Covid-19 for which they were designed. A failure that could epidemiologically underestimate the circulation of the virus in the population, which is the main objective of these tests. If they were reliable, they would allow the intensity of the quarantine to be regulated.

Yes, OK, but let's remember that governments all around the world are encountering similar problems.