University of Washington Lab Will Begin Testing of Thousands For COVID-19 Antibodies (geekwire.com) 64
GeekWire reports:
The University of Washington School of Medicine's Virology Lab is reporting encouraging results from trial runs of a new test from Abbott Laboratories that detects the antibodies created by people who have had COVID-19, whether they knew they had it or not. "This is a really fantastic test," Keith Jerome, who leads UW Medicine's virology program, told reporters today. He said UW's lab could process 4,000 samples per day starting next week, and conceivably ramp up to 14,000 samples per day within a couple of weeks.
The test will be made available through health care providers, in medical clinics or perhaps through workplaces... Epidemiologists say knowing who has had the virus will be key to tracking the true spread of COVID-19, and giving assurances to people who are returning to school and work — particularly in front-line jobs ranging from first responders and health care workers to grocery store clerks.
"It's possible this could be part of a back-to-work process," Jerome said. It's not yet clear how much immunity people develop to COVID-19 in the course of fighting off the virus, but if SARS-CoV-2 behaves like other coronaviruses, people with antibodies should have at least some protection from re-infection...
There's also a chance that being able to detect antibodies for COVID-19 will speed the development of vaccines in the months ahead.
1 million of the tests will be shipped this week in the U.S., and up to 4 million in the month of April, GeekWire reports.
And by June the company plans to be shipping 20 million tests per month.
The test will be made available through health care providers, in medical clinics or perhaps through workplaces... Epidemiologists say knowing who has had the virus will be key to tracking the true spread of COVID-19, and giving assurances to people who are returning to school and work — particularly in front-line jobs ranging from first responders and health care workers to grocery store clerks.
"It's possible this could be part of a back-to-work process," Jerome said. It's not yet clear how much immunity people develop to COVID-19 in the course of fighting off the virus, but if SARS-CoV-2 behaves like other coronaviruses, people with antibodies should have at least some protection from re-infection...
There's also a chance that being able to detect antibodies for COVID-19 will speed the development of vaccines in the months ahead.
1 million of the tests will be shipped this week in the U.S., and up to 4 million in the month of April, GeekWire reports.
And by June the company plans to be shipping 20 million tests per month.
maths (Score:2)
Re: (Score:2)
Re: (Score:1)
As long as I get my results before climate change is irreversible.
Re: (Score:3)
So "conceivably", those 5 MEELEEON tests could be processed within 360 weeks or so? About 10 years?
If it were *only* the UW lab running the tests, yes. However I suspect that Abbott (who manufactures the test) is shipping tests to other labs as well.
His math was wrong, too. (Score:3)
He got his arithmetic wrong, too. At 14,000 samples per day and running 7 days per week, UW's school could do more than five million tests per year.
Give them weekends off and you'd only need about 55 labs of the capacity of this one university med school lab to
Re: (Score:3)
Some nations can test a lot and get results out quickly.
South Korea, Germany, Taiwan, Australia.. .
Lots of advanced nations can do more testing. Their citizens at risk, citizens who have to keep working and in then for more and more citizens.
Re: (Score:1)
Presumably they'll get better at it before then. Or someone will. Or several will be equally slow, but they can divvy up the work.
One can hope, anyway.
Finally... (Score:4, Insightful)
I've suspected since early that this wasn't an end of the world plague but not a hoax either. I've also suspected that isolation for the low risk members of the world has done more harm than good.
Maybe a year or two from now we'll have the results of large scale studies and actually -know- something. I am thankful so many smart people are working hard to develop new tests, vaccines, cures, and protective medical gear.
My overall life view is highly positive. I've never believed this was the first step into global depression but I am concerned about how some folks turned super authoritarian and received support in that from so many regular folks. I'd rather live free and increase my risk of an early death than die a slave.
Be well, everyone.
Re: (Score:3)
I've had very little to say on cv19 because there's no solid information. It's all guessing and bad models and partisan politics.
The least solid part of the data has been total number of cases, whether or not they were 'symptomatic.' A good antigen test will clear that up.
Re: Finally... (Score:3)
Re: (Score:3, Interesting)
If you read the article, you will see that they did look at the test specificity with rigor and it does seem to be great, unlike many prior tests. The whole point of the article is that we finally have a test we can trust.
If Abbott can hit their numbers and deliver tens of millions of tests per month by July then by August we can start to reopen the economy for real. At the moment, we will go through a few cycles of attempted reopenings and then deaths will go up and it will be back to stay at home. But by
Re: (Score:3)
I've had very little to say on cv19 because there's no solid information. It's all guessing and bad models and partisan politics.
Smart play. Certainly the Slashdot groupthink does not want to hear that the continuous doomsaying from the media might be wrong.
But there's definitely something odd going on with CV19. There's no solid reason for the infection curve to have bent already. Even if you assume infection rates are 5x-10x the "confirmed" numbers that people use, [arcgis.com] which is the best current assumption from Iceland's "test everyone" effort, there's nowhere near enough people to affect the growth rate of the virus. Even if we say
Re: (Score:2)
There is a far simpler explanation. The infection rate stays relatively constant because there are only so many tests per day.
Re: (Score:2)
Old news - test manufacturing has caught up. 3.5 million have been tested in the US, five times as many as have been confirmed with the virus. Effectively, everyone who seeks care is now tested in the US.
If the virus is everywhere, there's more than one strain, and one is asymptomatic, as I said above. But there's no evidence for that, it's just a guess made up to fit the evidence we have.
Re: (Score:2)
3.5 million tests for a country of that size is not much. Spain, Italy and Germany each tested over a million, with a far smaller population - twice as many in relation to the population than the USA - yet they still consider it not good enough.
Re: (Score:2)
If everyone who is symptomatic is tested, then you can project the total number as 5x to 10x the confirmed number, based on what we now know. Since we have more than enough tests for everyone who seeks care, we can make reasonable projections.
Re: (Score:2)
This has been addressed by scientists rigorously. We need 500K-700K tests per day in the USA:
https://www.nytimes.com/intera... [nytimes.com]
We are actually not that far but not there yet. The number of tests needs to just about triple to be minimally satisfactory.
I believe though that this analysis assumed tests to be roughly reliable. In reality something like triple redundancy would be far better so I personally would love to see an order of magnitude increase in testing before anyone thinks about making any projection
Re: (Score:2)
Let me ask you this: do you believe it is possible to reason about a population as a whole by statistical sampling of that population, or do we know nothing at all about a population until we've applied the test to every element? Is "the field of statistics" some sort of conspiracy theory?
Re: (Score:2)
Statistical sampling is a great tool but you need what is known as a representative sample, meaning a sample from a cohort which has the same character as the population being studied and form which it is easy to extrapolate. Testing people who seek it is too biased for obvious reasons. Randomized testing is needed but has not been done almost anywhere because we just do not have enough tests. We know that the disease parameters look different in NY vs say WY so it we would need randomized testing from area
Re: (Score:2)
So, then, since Iceland tested almost everyone, and characterized the ratio of people with symptoms that cause them to seek care, to the total infected, we can reason about that, right?
I mean, it might not be exactly the same in the US, but let's use it within a factor of 2? So, based on that well-done sample, we can say that those actually infected in the US are 5x-10x the number of people with symptoms that cause them to seek care (as almost all of those are tested). Maybe not the exact ratio as in Icel
Re: (Score:2)
Oh, overall my oil patch stuff is up about 1/3rd. It's not like I bought at random. Holding on to most of it through the coming downturn, probably add more if things are a lot lower in a month.
Re: (Score:2)
I would tend to agree. There is Stanford study which says 50X but it has issues. That's likely a high outlier. There is the early estimate based on very little data which says 25% or so. But most recent studies seem to be converging to 5X (may be 10X but that seems like a stretch right now - could change). Anyways, that says this virus is about 7X more lethal than flu, and if 2/3 of US gets it (as seems to follow from R_0) then we should expect about 1.4 Million dead. If the number is 10X then we would get
Re: (Score:2)
Not even Germany has enough tests for everyone despite testing twice the percentage of the population and despite actually inventing the WHO test protocol. Only health care workers, people who land in a hospital or people who had a contact with a positive tested person get tested.
Re: Finally... (Score:1)
Re: (Score:2)
I keep hearing this stuff about two strains and this strange idea that we are not detecting one. That's not possible. We are testing for viral RNA with the most current tests and we are starting to test for antibodies. The RNA tests will detect any viral RNA, not just some specific sequence, and we can actually see from available data how virus evolves over time. Go to nextstrain.org and see for yourself. There are tons of mutations, there are many strains or at least many branches on that tree. There is no
Re: (Score:2)
the thought process that brought you from this:
We should still be seeing exponential growth, just at a slower rate.
to this:
More and more CV19 seems to be going away "on its own"
is not just stupid but very dangerously stupid. see why: https://coronawiki.org/compare... [coronawiki.org]
Re: (Score:1)
If the country hadn't been locked down, the spike in cases would definitely have overwhelmed the hospital systems (more than it has) - bodies rotting in the street, like you see in the third world. You can argue that it'd still only amount to a few percent of the population, and most everyone else (barring the psychological impact) would continue going to work.
That would surely have a smaller impact on the economy. But what about the people up for re-election this year? Imagine the hell they'd catch for tha
Re: (Score:2)
Re: (Score:2)
I've never believed this was the first step into global depression but I am concerned about how some folks turned super authoritarian and received support in that from so many regular folks. I'd rather live free and increase my risk of an early death than die a slave.
then i believe you should hurry up, because both the authoritarian drift and the depression were already underway, this will only accelerate the process. it's perfect, actually. shock doctrine at its finest.
the elite in each community will manufacture their own custom enemy. that seems to work when people is pissed off and has totally lost confidence in political systems that have been rotting for decades. this time around us seems to have picked china already.
Re: (Score:2)
Who actually thought it was an end-of-the-world plague? As of now it's caused about 160,000 deaths worldwide, with confirmed deaths rising at 5000-6000 per day. That's not world-ending, but that's pretty much what we were warned about back in January.
Re: (Score:2)
"It's all guessing and bad models and partisan politics."
Not quite. It is not about party affiliation or quality of the models. There is no guessing. 38932 people are already dead as of this writing so all our politicians have blood on their hands. This includes democrats like Pelosi who e.g. initially did not want to block foreign travel and it includes Trump (by default since he presided over this massacre).
Show me one party which wanted to impose martial law and nationwide shelter in place in late Januar
Re: (Score:2)
Smart nations and parts of the USA knew to have that ready after SARS...
The really, really smart part was some nations could read an expiry date and buy more PPE in. They kept their stockpiles of tested PPE ready over the years...
Other nations knew to test the masks they had and new test kits before use...
"Working hard" was done by nations like Taiwan, South Korea, Australia, Germany
Re: (Score:2)
"I'd rather live free and increase my risk of an early death than die a slave."
So... do you think that a suicide bomber should be free to get on board a plane? Or on a bus full of people? Because, you see, if you carry a virus then you are the weapon of mass destruction. Literally. Where does your freedom end and civilized life begin? When is a society at right to limit your freedom just a tiny bit so others may live? I do not know the answers. I hope you do. You seem very confident in your attitude. Do tel
Re: (Score:2)
" I'd rather live free and increase my risk of an early death than die a slave."
I'd rather live free than let you increase my risk of an early death.
some Orange buffoon will not be happy (Score:1, Offtopic)
It will ruin his numbers /s
Orange Man will be VERY happy. (Score:2)
some Orange buffoon will not be happy ... /s
It will ruin his numbers
Orange Man will be VERY happy. The entire point of his "cut the red tape and get several big companies going on each project - of their own choice" was to produce tests quickly and in volume.
This company's new test might not have shown up as quickly as you'd have liked. But this one alone, in the numbers they plan to be producing by July, can test the whole US population in under seven months.
And that's just ONE test from ONE comp
Re: (Score:1)
This company's new test might not have shown up as quickly as you'd have liked. But this one alone, in the numbers they plan to be producing by July, can test the whole US population in under seven months.
You understand they're talking about an antibody test right, not an infection test.
As in if you try to use this test to diagnose infection, you will end up with a TON of false positives.
Re: (Score:2)
Yes, this is the kind for finding people who HAVE had it (or are far enough along that they have developed some antibodies). Just what you need for real stats on how far it has been. For finding who is just coming down with it, nope.
If you tried to use it for a screening test you'd also get negatives for most of the people who are currently in the early stages and shedding the virus, too. (But it should go off for asymptomatic carriers who have been at it for a while and big time for sick people getting
Maybe, maybe not (Score:2)
if SARS-CoV-2 behaves like other coronaviruses, people with antibodies should have at least some protection from re-infection
Reports out of China [reuters.com] (yeah yeah), South Korea [usnews.com] (oh?) and Japan [forbes.com] are leading some to believe you can get reinfected. Inconclusive at the moment as no one is sure what the tests are picking up. Some say it might be the "dead" shell of the virus, leftovers of covid-19 DNA floating around or possibly bad testing.
At this point, as we hit 40K dead in the U.S in a month and a half, and how
Re: (Score:2)
Virologists all agree that reinfection may be possible but only over long time frames (eg. 10 years without a vaccine) similar to tetanus.
What you see in Korea and China is re-infection because their testing was so bad as to make it virtually ineffective. Both in Korea and China, testing comprised primarily out of body temperature readings.
Re: (Score:3)
"Virologists all agree that reinfection may be possible but only over long time frames (eg. 10 years without a vaccine) similar to tetanus."
No, my virologists seem to tend for 3-6 months, like with all the other coronaviruses.
Re: Maybe, maybe not (Score:2)
Virologists all agree that reinfection may be possible but only over long time frames (eg. 10 years without a vaccine) similar to tetanus.
This is all completely wrong. There is NO natural resistance from getting Tetanus at all, it's regular booster vaccines or nothing. Common seasonal Covids see antibodies start dropping after about 2 weeks and you rarely will find someone with any left after a year... anywhere from 1 to 6 months is a more realistic timeframe.
Re: (Score:2)
"Virologists all agree that reinfection may be possible but only over long time frames (eg. 10 years without a vaccine) similar to tetanus."
Hardly. That might be bacteriologists. Virologists have no clues about bacteria.
And that 10 year vaccine has to be done additionally to the 5 vaccinations as a kid and a 6th as a teen.
Re: (Score:1)
Virologists all agree that reinfection may be possible but only over long time frames (eg. 10 years without a vaccine) similar to tetanus.
No that isn't true at all guruevi. Care to show where you got these lies from?
What you see in Korea and China is re-infection because their testing was so bad as to make it virtually ineffective. Both in Korea and China, testing comprised primarily out of body temperature readings.
Complete nonsense again guruevi. Why is the US buying 750k tests from South Korea [cnn.com] if they are virtually ineffective? Even stupider to buy 'temperature checks' You have no idea at all. Or are purposely making up fake news.
Re: (Score:3)
My God, that's nearly *half* the number of people who died of coronaviruses and rhinoviruses last year!
Re: (Score:2)
That's not a good comparison considering it's a full year versus one and a half months.
Particularly given the exponential nature of infectious diseases.
It will easily best the normal flu despite dramatic reaction to mitigate.
The the real problem with exponential growth is that is is very difficult to have what feels like thte "just right" response. Either we won't take it seriously and have massive die off and think "we should have acted" or we take it seriously and keep it at roughly flu level making peopl
Re: (Score:2)
True, however, it also hardly appears to be an existential crisis, as was being hysterically claimed a month and a half ago. The extraordinary cost in human lives and well-being is completely out of proportion to the nature of the threat.
When this is over, everyone will be patting themselves on the back for "saving the country", when in fact we may be very rapidly approaching a civil war.
Re: (Score:2)
At this point, as we hit 40K dead in the U.S in a month and a half, and how fast this can and has spread, nothing should be assumed about this pandemic until more data is shown.
We'll probably pass that number today. https://www.worldometers.info/... [worldometers.info]
Still useless (Score:4, Insightful)
Tests like this may be useful to monitor small groups of people in high risk populations (first responders, populations at risk for health complications, etc.). But they do little to provide useful data for epidemiological studies. For that you need random sampling across the population at large. Not 100% coverage, but a sample of sufficient size to detect statistically significant changes in the rate of spread in the population at large. This is what we need to make policy decisions related to social distancing, closing businesses and services.
As long as tests are reserved for special classes of people, they do little other than to decide what restrictions should be placed upon individuals in that group. We can tell individual ER nurses when to stay home for example. They do nothing to manage the disease spread in the population at large.
But good luck redirecting any testing resources away from these special classes. We have the testing capacity to generate useful data today. But the political fodder that the cry of "Testing! Testing!" is just too useful to change the status quo.
Re: (Score:3)
Something like this:
https://www.theguardian.com/wo... [theguardian.com]
The criticism of the above is that their sample population was not randomized enough given they recruited through FB. I didn't rtfp to judge for myself. Even if you take that part with a big grain of salt and presume their antibody test had some proportion of false positives, it would reveal a very large multiplier of the 'official' numbers of covid positive folks to get the true population.
Re: (Score:2)
The criticism of the above is that their sample population was not randomized enough given they recruited through FB. I didn't rtfp to judge for myself. Even if you take that part with a big grain of salt and presume their antibody test had some proportion of false positives, it would reveal a very large multiplier of the 'official' numbers of covid positive folks to get the true population.
They other major issue is that 30 people to establish the specificity may be too small if there is cross reactivity with other coronaviruses. They are claiming a specificity of 99.5-100%. The only FDA approved COVID-19 test has a specificity of 95.4% or so. I suspect that their specificity is way off and so almost all of their result is false positives.
Starting to become useful. (Score:2)
Tests like this may be useful to monitor small groups of people in high risk populations (first responders, populations at risk for health complications, etc.). But they do little to provide useful data for epidemiological studies. For that you need random sampling across the population at large.
Make it "Tests like we've had and how we've used them so far" and I'll agree. ... good luck redirecting any testing resources away from these special classes
But that's what this story is about. This company has jus
Sure it's useful. (Score:2)
I agree it's important to do random population testing, but I suspect some kind of immunoassay is what we want there. We need to get a better figure for R0 so we can figure out how close we are to the herd immunity threshold. A PCR test won't tell us about people who've been infected and are now immune.
It *is* useful in finding asymptomatic people who are infectious, particularly ones who are in contact with a lot of people.
Re: (Score:3)
Get the USA back to work and exporting to the world again.
Re: (Score:2)
Test most of the nation.
Right. And if Trump says that the federal government is going to take charge, Cuomo throws a fit. But then he whines when we don't ship him enough ventilators and PPE. Actually, we aren't doing too bad, compared to other countries [ourworldindata.org]. But that's because of the federalist system of government we have. You want good national healthcare? Great. All of your state bureaucrats are out of jobs.
Get the USA back to work
It's not necessary to test 100% of the population to do this. All you need to do is to randomly select and test a statistical
visit my website (Score:1)