The latest coronavirus antibody test is a lot more accurate (100% accurate at detecting antibodies)

[Celerity]

99.8% accurate at ruling out the presence of those antibodies

https://qz.com/1850518/roches-new-coronavirus-antibody-test-is-much-more-accurate/

Several coronavirus antibody tests have been authorized for public use, but so far their accuracy has been iffy. A new test created by Roche and cleared by the US Food and Drug Administration (FDA) for emergency use today, May 3rd, significantly ups the standard.

Roche announced that its test is 100% accurate at detecting coronavirus antibodies and 99.8% accurate at ruling out the presence of those antibodies, meaning only one in 500 tests will get a false positive. Antibody tests use blood samples to assess whether a person had been previously infected, so they’re useful to determine the true spread of coronavirus.

In comparison, the first test the FDA approved for emergency use, created by Cellex, is 93.8% accurate at detecting coronavirus antibodies (this is known as sensitivity), and 95.6% accurate at ruling out the presence of antibodies (known as specificity.) Meanwhile, Premier Biotech’s test has sensitivity of 80.3% and specificity of 99.5%.

Though the differences in accuracy may seem small, a Quartz simulation shows they have huge implications when only a small proportion of people are infected, as the number of false positives can easily equal the number of true positives. For example, if 5% of 1,000 people have had coronavirus then, using the Cellex test, a positive result stands only 49.5% chance of being correct. Using the Roche test on that same population means that a positive result is 96% likely to be correct.

snip


Roche’s COVID-19 antibody test receives FDA Emergency Use Authorization and is available in markets accepting the CE mark

https://www.roche.com/media/releases/med-cor-2020-05-03.htm

The serology test has a specificity greater than 99.8% and sensitivity of 100% (14 Days post-PCR confirmation)

The high specificity of the test is crucial to determine reliably if a person has been exposed to the virus and if the patient has developed antibodies

Roche will provide high double-digit millions of tests already in May for countries accepting the CE mark and in the U.S. under Emergency Use Authorization, further ramping up capacities thereafter

The test is available on Roche’s cobas e analysers which are widely available around the world


snip

Roche CEO says it is ‘very likely’ people develop immunity after recovering from coronavirus

https://www.cnbc.com/2020/05/04/roche-ceo-very-likely-coronavirus-patients-develop-immunity.html


Coronavirus patients can’t relapse, South Korean scientists believe

https://www.thetimes.co.uk/article/coronavirus-patients-cant-relapse-south-korean-scientists-believe-rkm8zm7d9

Thursday April 30 2020, 12.00pm BST, The Times





My wife and I both tested antibody positive Friday (we got the news this morning) with this blood serological test (we already had tested positive with a less accurate test but our doctor asked us if we wanted to be in a study with the Roche test) and we are both breaking quarantine here in Sweden and going back to work next Monday. Close to 3 months is enough, especially as there have been almost zero people under 40 die here of COVID-19 (revised figures fro our national heath service now are 10 deaths in the 30-39 age cohort, and zero deaths under 30 years of age,per our doctor at Karolinska universitetssjukhuset in Solna, and that is with most all of the schools under high school level open the whole time) and no one we know is still self-isolating, most never did. It obviously did not stop us from being infected anyway.

We do not at all regret erring on the side of caution, but the time has come to end this (for us, but each person needs to weigh their own risk/reward options.)

[onecaliberal]

1. Fresno county practically has zero tests

I hear about surrounding counties getting tests and antibody tests. I know people here who most likely had it and could not get tested.

[ismnotwasm]

2. I wish we could mass test here.

I’m very excited about the antibodies and the basic lack of reinfection

[Celerity]

3. I would think that Trump would want that info out, as it bolsters his case for re-opening

but Trump NEVER does what seems to make sense or what is right.

He is a fucking numpty berk.

[jberryhill]

4. That's 2000 people out of every million

So, in a large city, we’re talking about 10,000 false negatives.

[muriel_volestrangler]

6. 10,000 false positives

It's claimed it never shows a false negative.

[jberryhill]

10. Stated an odd way

"99.8% accurate at ruling out the presence of antibodies"

A "positive" is "the presence of antibodies".

A "negative" is "the absence of antibodies".

This is something of a conundrum, no?

What we want to know is "has this person been exposed to the virus". If it is 99.8% accurate at saying "this person has not been exposed to the virus" (i.e. negative), then I am confused by the definition of "positive".

Of those who have antibodies, the "100%" claim suggests that anyone with antibodies will correctly test positive with zero false positives.

Of those who do not have antibodies then the "99.8%" claim suggests that 99.8% will correctly test negative, and that .2% will falsely test negative.

Which, in Philadelphia, would mean that 10,000 people would be told they are at risk of infection, when they are not at substantial risk. (assuming we've nailed down, per other news, that the risk of reinfection is very low)

That is an err on the safe side outcome, but would suck to be one of those 10,000 people.

[muriel_volestrangler]

19. A "false positive" is someone the test claims is positive, but is in reality negative

In medical testing, and more generally in binary classification, a false positive is an error in data reporting in which a test result improperly indicates presence of a condition, such as a disease (the result is positive), when in reality it is not present, while a false negative is an error in which a test result improperly indicates no presence of a condition (the result is negative), when in reality it is present.

https://en.wikipedia.org/wiki/False_positives_and_false_negatives

Also, it looks like the Philadelphia population is just over 1.5 million*, so if we say 1.5 million have not yet had the virus, then 1 in 500 of that is 3,000, not 10,000, so:

Of those who have antibodies, the "100%" claim suggests that anyone with antibodies will correctly test positive with zero false negatives.

Of those who do not have antibodies then the "99.8%" claim suggests that 99.8% will correctly test negative, and that .2% will falsely test positive.

Which, in Philadelphia, would mean that 3,000 people would be told they are not at risk of infection, when they actually are. (assuming we've nailed down, per other news, that the risk of reinfection is very low)

That is not an error on the safe side outcome. It would suck to be one of those 3,000 people if they changed their behaviour and subsequently caught it.

*Unless you mean the larger Metropolitan Statistical Area of Philadelphia-Camden-Wilmington, which has about 6 million: https://en.wikipedia.org/wiki/List_of_metropolitan_statistical_areas

[Igel]

13. Yippee.

This falls into the "not much of a problem" category.

If you get the test done and it comes back negative when you've been infected, it's inconvenient.

If it comes back positive when you haven't been infected it means you could easily become infected.

10k false negatives would affect worker productivity and lifestyle, maybe just mean wear a mask, maybe mean stay at home.

The COVID PCR tests for active infection are potentially more important. If you get a false negative, you're releasing somebody infectious back into the population. This isn't a big deal when we're not down to virtually no new cases and transmission is widespread.

PCR tests that come back false positive are potentially more important but only in specific situations. If it means you self-isolate at home for a couple of weeks, again, it's inconvenient. If you're in a place where you'd immediately quarantined with possibly infectious sufferers, it all but guarantees you'd become infected.

[frazzled]

5. Are Antibodies the Way Back? The Research is Mixed

Italians Find Promise of Antibodies Remains Elusive, for Now

Talk of licensing people with the right antibodies, always ahead of the science, has faded as experts warn that they are still studying what level offers protection and how long it lasts. ...

Researchers and politicians in China, the United States, Germany, Britain and beyond have latched onto antibodies as a potential solution to the virus and an outlet from containment measures.
But that talk, always ahead of the science, has grown more muted in recent weeks. With the research refusing to cooperate, experts in Italy say the promise of antibodies may not be what people have imagined. At least for now.

... Infected people develop different quantities of antibodies, and researchers are still studying the level that offers protection, and for how long. “We don’t know how long they last,” Dr. Venturi said. ‘‘This is the central point.”

... Many regions, including Lombardy and Veneto, are still doing such screenings. But whereas last month the presidents of those regions promoted the idea of issuing licenses to members of an immune work force, now they have downgraded the tests from panacea to a research tool.

https://www.nytimes.com/2020/05/03/world/europe/coronavirus-antibodies-italy.html?searchResultPosition=1

Let's be clear: no one, from scientists to doctors, really understands much of anything about this virus yet. I have had that confirmed by two physicians (one in Belgium, one here in the US) in the past few days. Hooray for new antibody test, but it's hardly indicative of anything right now. It's great for research, but it won't yet indicate when things are safe for individuals or society, or whether there will be a second wave.

[CountAllVotes]

8. +1,000+++++

As was stated at a WHO meeting the other day re: COVID-19 --

"There is no treatment, a cure or a vaccine" .... *reality check* !!!

They DO NOT KNOW WTF they are dealing with!

If you think they do, you are putting your life and your faith in drug companies with an obvious agenda, that is to push an assortment of bogus drugs, treatments, etc. that DO NOT WORK!

[Igel]

16. You see that word "is"?

There is no agreement that human space flight is possible.

There is no agreement on what causes influenza.

There is no agreement that washing hands can reduce infection.

There is no agreement that gravity follows the same rules in Europe as in the New World.

All those statements are false. All those statements are true if I add a timestamp to them. The difference is when the word "is" points to. If to the present, they're utterly false. If the first is dated 1920, the second 1885, the third is 1850, and the fourth is 1600, then they're utterly true.

Last week there was no knowledge that antibodies confer immunity. Now there is.

Last week there was no knowledge that remdesivir had any benefit. Now there is--limited benefit, but it's something.

Still no word on vaccine. Or a good treatment. But all knowledge comes with a timestamp.

[Igel]

14. Some understand limited things well.

As the press release slightly revised into "breaking news" said, how long immunity lasts is a question. But it's clear that at least for most people antibodies confer immunity.

Why? Because at least in S. Korea, in hundreds of cases where reinfection was suspected there were no (count them, "no&quot live virions. It's still possible that just maybe it was a giant fluke--that all those people just happened to be retested for some reason, test positive, and have no actual recent exposure. It's looking very much like that's not the case, but weirder things have been known to happen. "Highly improbable" does not mean "impossible." We just round at 5 or 6 decimal places and figure that things that highly improbable happen so seldom that the distinction isn't one to be drawn.

[still_one]

7. The problem is most people don't get to choose which antibody test to use. They use the one which

the clinic they go to uses

[Celerity]

9. I think the Roche test will soon become the gold standard and the most widely used

[Igel]

17. Perhaps.

Roche cites two tests. 11 had emergency authorization for use. It's unlikely they picked the best two for comparison.

The OP is a slightly edited press release.

[CountAllVotes]

20. +1

The old professor I worked for ran Roche Pharmaceuticals out of his office one day and he was a very controlled man.

However, that particular day, he was screaming at them, "Five years! FIVE YEARS! Five years is nothing to study a drug!".

Roche did not venture back for a number of months which was kind of nice as I was the keeper of the gate to his office so to speak!

I learned more than one would ever care to know abt. the pharmaceutical industry in that job needless to say!

[Greybnk48]

11. According to John Oliver and the scientist on his show, ZERO tests

have gotten FDA approval. Four of the myriad tests out there have gotten a lower ranking but not approval as yet. Not a joke, sadly.

https://www.hollywoodreporter.com/live-feed/john-oliver-breaks-down-lack-coronavirus-testing-us-1293045

[Celerity]

12. He is now wrong, Roche Coronavirus Antibody Test Wins FDA Approval for Emergency Use

https://www.wsj.com/articles/roche-coronavirus-antibody-test-wins-fda-approval-for-emergency-use-11588505019


Pharma giant Roche gets US go-ahead for Covid-19 antibody test

https://www.theguardian.com/world/2020/may/03/pharma-giant-roche-gets-us-go-ahead-for-covid-19-antibody-test

Swiss drug maker Roche Holding AG says it has received emergency use approval from the US Food and Drug Administration (FDA) for an antibody test to help determine if people have ever been infected with the coronavirus.

Governments, businesses and individuals are seeking such blood tests to help them learn more about who may have had the disease, who may have some immunity and to potentially craft strategies to end lockdowns that have battered global economies.

Roche had previously pledged to make its antibody test available by early May and to boost production by June to “high double-digit millions” per month.

[Wounded Bear]

15. "Approval for emergency use" is not the same as "approval"...

Obviously, there is no large scale statistical basis for the efficacy of this test.

For now we're relying on the company data and press releases. I hope it's true, but we won't really know until it is put in mass use, IOW they're beta testing it on the public.

[CountAllVotes]

18. +1

I do not trust any of these drug companies.

I have a good reason for it.

I worked for years in the field of pharmaceutical research and these drug companies do all they can to push a drug through, properly tested or not.

If one cares to be a guinea pig, well go for it -- that is where it is at.

[Chainfire]

21. A good test is great

But where is the science on whether having the antibodies means you are protected from reinfection?

The last I read was it was in the "maybe" category.

[Chainfire]

22. A great drug (or test) for a drug company

is one that they can sell at fantastic profit margins.