Sounds like he's being too honest. Expect him to be replaced imminently.
Seriously, how are we so bad at this? We knew it was coming and South Korea has been testing 10k/day since Feb, so it's clearly doable. We've done <10k total! What in the world is going on with the tests?
It appears the main driver has been lack of chemicals and in the beginning, the cdc required tests to be sent to them. The states can now test on their own.
There was literally a policy not to allow states to use testing from other countries. And also to restrict testing a lot, early on. Even India started aggressively testing many travelers long before.
Even so, I will echo the OPs main point of the USA had a head start and did nothing with it. These shortages and policies could have been reviewed and resolved back in January or even February.
It’s so hard to trust the media to report on anything without trying to inflame the “orange man bad” narrative.
In Australia almost no one with some or most of the symptoms is being tested. General Practitioners are basically telling people to fuck off, self isolate and call the hospital if symptoms worsen.
“Currently we are not testing everyone, and whilst this may change in the future, at the moment this isn’t feasible. The pressure on labs would be mammoth and the yield is likely low,”
At what point does it become OK to point out that the "orange man" is, in fact, bad? Fox news and republican leaders spent weeks (weeks that we can't get back) trying to tell people that this wasn't a problem, that they shouldn't prepare, and that this was all a partisan argument against the president. And that was all wrong. It's real, it's happening, and we've lost critical time.
That's bad, right? And it's bad in a distinctly partisan way, because the "other side" didn't say or do that stuff, and was broadly correct in its pronouncements. Can we say that? When can we say that? Or is any amount of wrongdoing or incompetence allowed as long as you get a whole political party to do it at once?
Especially since the article isn't about him. It mentions him twice, once citing the travel ban, another time quoting him as saying that we have good testing infrastructure (but noting that he did not elaborate). It does not describe him as "bad" or say anything directly negative. The article is about a high-level bureaucrat who says only "we".
The OP's comment is, unfortunately, indicative. The OP reads it as negative, and the reality is negative, but his primary concern is about the assignment of blame. In particular, assuring that it does not fall on him, and trying to shift it anywhere else. Meantime, the person actually cited in the article is trying to get something done -- which starts by admitting the there is a thing that needs to be done.
The idea is that anything that any article that does not go out of its way to praise him must be negative by definition, because that's how his mind works. The US's response to this current situation is a direct response by Trump to how he is being perceived.
Orange man is bad, but a stopped clock is right twice a day, and falling over themselves to deny whenever that happens has led the media to say all manner of braindead things.
Or maybe... (to strain the metaphor a bit) the clock has been working the whole time, but you didn't care because you didn't need to know the time?
I mean... the topic at hand in this subthread is the executive branch's ability to coordinate a single message, to tell the truth, and to ensure that public policy matches official pronouncements. And this has been a serious problem for the Trump administration since inauguration. The president will say things that are wrong, he'll mistate positions of others, he'll promise things that aren't borne out by policy, he'll contradict existing white house messaging, and the rest of the executive branch is constantly running around writing up explainers for how all of this is supposed to fit together.
And the media reported on that. But to you, maybe it just seemed like the president was speaking extemporaneously and that the media coverage was being excessively "nitpicky". What other people would call "lies" or "mistruths" seemed to you like just mistakes or irrelevancies. So to you, the media wasn't being fair.
Well, now the world is falling apart and the president's words matter. And the lies seem more like lies to most people. That's not because the media is treating them any different.
Guess what? They're all bad! Incompetence, perverse incentives, misinformation, exploitation: we're surrounded by it.
On Hacker News I expect a technical discussion and shared files how to build PCR kits at home to test for the virus. I want to see people sharing biohacks. Not this meaningless useless byllshit.
>In Australia almost no one with some or most of the symptoms is being tested. General Practitioners are basically telling people to fuck off, self isolate and call the hospital if symptoms worsen.
This is simply false:
>Here, testing is free and widely available, thanks to early and coordinated planning for a pandemic. On Thursday, Mr. Hanks said he and his wife had seen the efforts firsthand, as they tested positive for the virus.
It’s not false because I went through this myself yesterday. Was told over the phone “don’t come in, its probably a common cold, stay at home and if symptoms worsen called the health hotline”
Community forums are full others reporting similar stories
“Currently we are not testing everyone, and whilst this may change in the future, at the moment this isn’t feasible. The pressure on labs would be mammoth and the yield is likely low,”
Sure, but many people have colds. Did you report a runny nose? Did you have a fever? If yes and no, then it's probably not COVID-19 and there's not much point testing you.
Regardless of how reasonable things seem at the moment, when it comes down to it I trust the Australian authorities and healthcare system in a way that I just don't for the US.
You don't go in to get tested if you are suspected of having coronavirus. If the management is still in the contact tracing phase, they will send someone to you to take a swab.
If you have symptomatic CV and you go to a hospital waiting room to be tested, you have done a bad thing.
There is no clear-cut distinction between the symptoms of the two - it's more an "if you hear hooves, think horse not zebra" thing. The New York Times article in the grandparent post literally points to the fact that Tom Hanks could get tested for coronavirus based on symtoms that were just like a cold as proof it's much easier to get tested in Australia than the US.
Except that if you read the ABC article, Australia only tests people who both have symptoms consistent with the disease AND either have travelled to an affected country or had contact with a confirmed case. Also, the US press has managed to convince a large (or at least vocal) chunk of the American public that if they can't get a Covid-19 test based on their cold-like symptoms it's somehow proof their healthcare system is falling behind the rest of the world in testing as part of a Trump cover-up, and it's leaking to the rest of the English speaking world. It's all over social media.
> points to the fact that Tom Hanks could get tested for coronavirus based on symtoms that were just like a cold as proof it's much easier to get tested in Australia than the US.
I think this just points to the ease of being tested if you are Tom Hanks.
Realistically there is no "US" any more. Fauci is speaking for the NIAID, and arguably for the scientific community within the CDC/NIH/HHS. The white house certainly wouldn't be on board with this message, but as we saw last night even the white house is inconsistent with its own policies in important ways.
But yes: we messed up. We weren't prepared to begin with, we rolled back a bunch of existing preparedness work, when the disease arrived we turtled instead of testing, and then we refused to scale the testing capability for critical weeks. At this point it's too late. Better testing is still needed to inform the hard public health decisions that will have to be made, but it's no longer able to contain outbreaks that have grown orders of magnitude too large.
Absolutely. Was talking with a friend who was confused about why schools in remote areas are being shut down - testing needs to be done just so people realize that it's not just Washington or New York, this is pretty much everywhere or imminently about to be everywhere in the United States.
They did start working on it 4 months ago. The CDC has been on the case since then. Mind you, China was not cooperating with us and wouldn't allow our experts into the country while it was getting bad there. They are ramping up rapidly and we'll see that over the next few days.
I'm sorry, that's just... completely wrong. Everything you said is wrong.
People in the media and elsewhere have been clamoring for months for details from the CDC about how they'll be rolling out testing, and it never arrived. It still hasn't arrived. We're still in the low dozens of tests per day. You're telling me that it's somehow physically impossible for the CDC to have ramped testing beyond that when other nations are in the thousands or tens of thousands per day?
And as far as China not cooperating: they published the full sequence of this thing back in January. We had the genes for this thing even before we had a name for it, or for the disease. And that's not enough help for the CDC?
Where are you getting this information from? Please tell me it's not partisan media.
The best response was from a somewhere China thinks should not exist - the Taiwanese government. They started working on this when China finally let the rest of the world know. There is a lot of people traffic between those two countries so one might have expected Taiwan to have been hit hard, but the Taiwan health authorities were aggressive in being preventative.
Italy is geographically half the size of Texas and South Korea is 1/7 the size of Texas. It's considerably easier to do widespread testing when everyone practically lives next to everyone else.
This is the weirdest "US is special" argument. You can pack a hundred samples (blood/swab/whatever) in a box and send it to whatever lab in a major US city, in a matter of hours.
The chemicals needed to do a test are in short supply. I would assume Italy and South Korea used a lot of the supply, meaning less for anyone else. Italy of course clearly needed all they used (and more), while Korea could be argued used to much. Note that latter is a no-win situation - nobody knows exactly how much testing is needed, either you test what is probably too much or you don't test enough.
Or you can not test, and treat everybody as infected. For hospitals and high risks people you should be doing that anyway.
It's not just one group that's doing it, though. This sort of testing (as addressed in other comments) can be done through university and hospital labs throughout the country, along with state and local health agencies and various private testing firms (mostly consolidated at this point).
Instead of going it alone, the focus could've (perhaps should've) been on early collaboration with those partners on establishing consistent testing procedures (whether they used precisely the same measures, validation that they were getting the same results) and a process to distribute the workload.
This wasn't done, that's the reason the scaling failed.
Not when you run out of raw materials to make the tests. I could leave my job and create them, but that would take months to train me (and maybe years if I have to create my own equipment - it takes time to become a certified welder as just one task that doesn't scale quickly)
The white house said that testing is going "really smooth" and all travelers are being tested two hours ago on live TV, both are lies, obviously the white house is not on board with Fauci's message.
Interestingly enough from another current HN article posting [1]: "I have read accounts in the press that Dr. [Anthony] Fauci isn't really allowed to speak until Mike Pence has approved his messaging."
Be interesting whether this is official or behind the WH's back, this BBC article doesn't have too many details.
There is nothing secret about RNA testing. Any decent lab can synthetise its own markers from a file downloaded from internet. It is like genetic research is illegal in US or what.
Offering a diagnostic test without FDA approval is illegal, for good reason (there would be companies far worse than Theranos) but clearly in an emergency regulators need to adapt.
In Arizona, Mayo Clinic, Sonora Quest Laboratories, and the Translational Genomics Research Institute in Flagstaff are rolling out their own tests. Official test kits from the CDC are in woefully short supply, under 200 right now, and only being used for people showing significant symptoms. According to a news article the new tests will use a technique similar to the CDC test known as reverse transcription polymerase chain reaction, otherwise shortened to RT-PCR
2) The reagents needed for the testing are also in extremely short supply, and labs are having a hard time getting more. So, you can develop a test, but your testing capacity will remain limited.
I mean, kudos to CCF, but take this for what it's worth (from another tier 1 University Hospital):
Most university hospitals' in house test results are / are going to be available in similar time frame, but they do not have the media budget of the Cleveland Clinic. Vox article links to a CCF press release fluff piece.
Other labs across the US are ramping up similar tests. This is exactly what we need and we shouldn’t mind all of them taking credit for their contributions.
A: About 2 to 3 hours to do the testing, but because the process requires us to batch samples, you can't really just test the sample as they come in. In order to conserve reagents and labor, you' wait until you have a certain bolus of samples and then test them together. So you do testing once or twice a day.
So the process used in this testing is so common that research labs and any biochemistry lab in any college or high-school can run the test. But they probably shouldn't because they have no practice nor experience providing a diagnosis to a human - much less the strict statistical controls to interpret their results reliably.
A clinical lab that is used to running these kinds of (very common) protocols will see no difference in this protocol from others they commonly run (from a chain of custody, sample integrity, etc. perspective), just with a slight change to their reagents to specifically test for COVID. Their handling should be no different.
It is, but there have been major bureaucratic delays, and no one was willing to step up and say "forget it, we are doing it". In addition, there have been shortages and/or problems with RNA extraction reagents/kits (this is a ssRNA virus) as well as viral nucleic acid sequence for positive control (positive control is critical for this type of testing).
I have a hard time faulting the CDC for the response overall, but I just don't understand why they decided to make their own test at the beginning. Was it bravado? Did they think the 70% accuracy of the WHO test wasn't good enough?
If they didn't make their own test, they wouldn't have fubarred the manufacturing, which would have given us more tests.
One of the CDC's main goals, if not the main goal in this, is to spread out the panic. Given this goal, the fact that we're in flu season, and the lack of test kits, I can understand the stringent testing criteria. They choose the best option in a bad situation, and I think they've done a pretty good job. However, it seems like a lot of problems could have been avoided if they just used the WHO tests.
The test accuracy of the WHO test as far as I understood it was good if someone was already experiencing symptoms once viral load was high enough. The bigger issue was false negatives, people would get released from quarantine measures, then spread it (like the cruise ship). They recommended anyone within contact to self quarantine until symptoms had reached a point they could get a definitive test. At this point testing early and coming up negative doesn't mean anything, you should still stay quarantined and continuously get retested, since a lot of people who tested negative were found to have had it the whole time when symptoms present and were given the all clear before.
In summary, the testing early doesn't change the decision tree, which is stay quarantined for 14 days, until symptoms present or not, so little point was seen to testing until then.
Problem is now, we can't track it, so everyone ideally should self quarantine or limit as much as possible social interaction and get tested if they have good reason to believe they've been exposed. Early testing will catch a lot of people but not all, so it will slow the spread, so at this point the decision tree changes and we need to start testing more, if nothing else to get a sample to extrapolate the true spread of this thing.
> In summary, the testing early doesn't change the decision tree, which is stay quarantined for 14 days, until symptoms present or not, so little point was seen to testing until then.
As far as I can tell, the answer is quite simple: there was no WHO test at the beginning. The CDC's test was developed and rolled out pretty much in parallel with the German one the WHO has been using. They deployed it to state testing labs at pretty much exactly the same time as WHO Europe was rolling theirs out to one lab per European state. They also had a very aggressive plan to start testing flu surveillance samples in mid-February, much sooner than other countries. I'm not convinced they could pull off such an aggressive rollout without manufacturing their own tests.
What is the true/false negative/positive rate for these tests? I can't find that online and I have tried.
If FPR is 10% and we test 100,000,000 people in the US but only 100,000 have the disease right now, then assuming a TPR of 100% (and a FNR of 0%) we will have 10,100,000 positive results = 101 x the number of people with COVID? How is that useful?
84 comments
[ 3.3 ms ] story [ 163 ms ] threadSeriously, how are we so bad at this? We knew it was coming and South Korea has been testing 10k/day since Feb, so it's clearly doable. We've done <10k total! What in the world is going on with the tests?
It appears the main driver has been lack of chemicals and in the beginning, the cdc required tests to be sent to them. The states can now test on their own.
https://www.vox.com/science-and-health/2020/3/12/21175034/co...
In Australia almost no one with some or most of the symptoms is being tested. General Practitioners are basically telling people to fuck off, self isolate and call the hospital if symptoms worsen.
https://apple.news/ASvoRONOWS9ijWX4RrGmiTQ
“Currently we are not testing everyone, and whilst this may change in the future, at the moment this isn’t feasible. The pressure on labs would be mammoth and the yield is likely low,”
Similar stories from colleagues in Europe.
That's bad, right? And it's bad in a distinctly partisan way, because the "other side" didn't say or do that stuff, and was broadly correct in its pronouncements. Can we say that? When can we say that? Or is any amount of wrongdoing or incompetence allowed as long as you get a whole political party to do it at once?
The OP's comment is, unfortunately, indicative. The OP reads it as negative, and the reality is negative, but his primary concern is about the assignment of blame. In particular, assuring that it does not fall on him, and trying to shift it anywhere else. Meantime, the person actually cited in the article is trying to get something done -- which starts by admitting the there is a thing that needs to be done.
Without another working clock, that's still not super useful, as you can't figure out when it's right.
I mean... the topic at hand in this subthread is the executive branch's ability to coordinate a single message, to tell the truth, and to ensure that public policy matches official pronouncements. And this has been a serious problem for the Trump administration since inauguration. The president will say things that are wrong, he'll mistate positions of others, he'll promise things that aren't borne out by policy, he'll contradict existing white house messaging, and the rest of the executive branch is constantly running around writing up explainers for how all of this is supposed to fit together.
And the media reported on that. But to you, maybe it just seemed like the president was speaking extemporaneously and that the media coverage was being excessively "nitpicky". What other people would call "lies" or "mistruths" seemed to you like just mistakes or irrelevancies. So to you, the media wasn't being fair.
Well, now the world is falling apart and the president's words matter. And the lies seem more like lies to most people. That's not because the media is treating them any different.
On Hacker News I expect a technical discussion and shared files how to build PCR kits at home to test for the virus. I want to see people sharing biohacks. Not this meaningless useless byllshit.
This is simply false:
>Here, testing is free and widely available, thanks to early and coordinated planning for a pandemic. On Thursday, Mr. Hanks said he and his wife had seen the efforts firsthand, as they tested positive for the virus.
https://www.nytimes.com/2020/03/12/world/australia/tom-hanks...
Community forums are full others reporting similar stories
https://apple.news/ASvoRONOWS9ijWX4RrGmiTQ
“Currently we are not testing everyone, and whilst this may change in the future, at the moment this isn’t feasible. The pressure on labs would be mammoth and the yield is likely low,”
Regardless of how reasonable things seem at the moment, when it comes down to it I trust the Australian authorities and healthcare system in a way that I just don't for the US.
If you have symptomatic CV and you go to a hospital waiting room to be tested, you have done a bad thing.
Except that if you read the ABC article, Australia only tests people who both have symptoms consistent with the disease AND either have travelled to an affected country or had contact with a confirmed case. Also, the US press has managed to convince a large (or at least vocal) chunk of the American public that if they can't get a Covid-19 test based on their cold-like symptoms it's somehow proof their healthcare system is falling behind the rest of the world in testing as part of a Trump cover-up, and it's leaking to the rest of the English speaking world. It's all over social media.
I think this just points to the ease of being tested if you are Tom Hanks.
But yes: we messed up. We weren't prepared to begin with, we rolled back a bunch of existing preparedness work, when the disease arrived we turtled instead of testing, and then we refused to scale the testing capability for critical weeks. At this point it's too late. Better testing is still needed to inform the hard public health decisions that will have to be made, but it's no longer able to contain outbreaks that have grown orders of magnitude too large.
People in the media and elsewhere have been clamoring for months for details from the CDC about how they'll be rolling out testing, and it never arrived. It still hasn't arrived. We're still in the low dozens of tests per day. You're telling me that it's somehow physically impossible for the CDC to have ramped testing beyond that when other nations are in the thousands or tens of thousands per day?
And as far as China not cooperating: they published the full sequence of this thing back in January. We had the genes for this thing even before we had a name for it, or for the disease. And that's not enough help for the CDC?
Where are you getting this information from? Please tell me it's not partisan media.
https://jamanetwork.com/journals/jama/fullarticle/2762689
Or you can not test, and treat everybody as infected. For hospitals and high risks people you should be doing that anyway.
Instead of going it alone, the focus could've (perhaps should've) been on early collaboration with those partners on establishing consistent testing procedures (whether they used precisely the same measures, validation that they were getting the same results) and a process to distribute the workload.
This wasn't done, that's the reason the scaling failed.
It's just RNA sequencing, it can be done at any university using standard equipment and materials
b) If it's that scarce, we should've been stockpiling more.
c) If it's that scarce, we should've been manufacturing more.
Be interesting whether this is official or behind the WH's back, this BBC article doesn't have too many details.
[1] https://news.ycombinator.com/item?id=22556600
The Cleveland Clinic just came out with an 8 hour test and hopefully this will help the situation if they share the process.
https://www.vox.com/science-and-health/2020/3/12/21175034/co...
2) The reagents needed for the testing are also in extremely short supply, and labs are having a hard time getting more. So, you can develop a test, but your testing capacity will remain limited.
*unscientific guess for literary effect
Most university hospitals' in house test results are / are going to be available in similar time frame, but they do not have the media budget of the Cleveland Clinic. Vox article links to a CCF press release fluff piece.
Other labs across the US are ramping up similar tests. This is exactly what we need and we shouldn’t mind all of them taking credit for their contributions.
https://www.sciencemag.org/news/2020/03/were-behind-curve-us... Q: What's the turnaround time when you have a sample on hand? How long does it take to extract RNA from the sample, put it into a polymerase chain reaction machine, etc.?
A: About 2 to 3 hours to do the testing, but because the process requires us to batch samples, you can't really just test the sample as they come in. In order to conserve reagents and labor, you' wait until you have a certain bolus of samples and then test them together. So you do testing once or twice a day.
A clinical lab that is used to running these kinds of (very common) protocols will see no difference in this protocol from others they commonly run (from a chain of custody, sample integrity, etc. perspective), just with a slight change to their reagents to specifically test for COVID. Their handling should be no different.
If they didn't make their own test, they wouldn't have fubarred the manufacturing, which would have given us more tests.
One of the CDC's main goals, if not the main goal in this, is to spread out the panic. Given this goal, the fact that we're in flu season, and the lack of test kits, I can understand the stringent testing criteria. They choose the best option in a bad situation, and I think they've done a pretty good job. However, it seems like a lot of problems could have been avoided if they just used the WHO tests.
In summary, the testing early doesn't change the decision tree, which is stay quarantined for 14 days, until symptoms present or not, so little point was seen to testing until then.
Problem is now, we can't track it, so everyone ideally should self quarantine or limit as much as possible social interaction and get tested if they have good reason to believe they've been exposed. Early testing will catch a lot of people but not all, so it will slow the spread, so at this point the decision tree changes and we need to start testing more, if nothing else to get a sample to extrapolate the true spread of this thing.
That makes a ton of sense. Thank you!
If FPR is 10% and we test 100,000,000 people in the US but only 100,000 have the disease right now, then assuming a TPR of 100% (and a FNR of 0%) we will have 10,100,000 positive results = 101 x the number of people with COVID? How is that useful?
Disclaimer: I am maybe an idiot.