Frequent, Fast, and Cheap is Better than Sensitive

A number of firms have developed cheap, paper-strip tests for coronavirus that report results at-home in about 15 minutes but they have yet to be approved for use by the FDA because the FDA appears to be demanding that all tests reach accuracy levels similar to the PCR test. This is another deadly FDA mistake.

NPR: Highly accurate at-home tests are probably many months away. But Mina argues they could be here sooner if the FDA would not demand that tests for the coronavirus meet really high accuracy standards of 80 percent or better.

A Massachusetts-based startup called E25Bio has developed this sort of rapid test. Founder and Chief Technology Officer Irene Bosch says her firm has field-tested it in hospitals. “What we learned is that the test is able to be very efficient for people who have a lot of virus,” she says.

The PCR tests can discover virus at significantly lower concentration levels than the cheap tests but that extra sensitivity doesn’t matter much in practice. Why not? First, at the lowest levels that the PCR test can detect, the person tested probably isn’t infectious. The cheap test is better at telling whether you are infectious than whether you are infected but the former is what we need to know to open schools and workplaces. Second, the virus grows so quickly that the time period in which the PCR tests outperforms the cheap test is as little as a day or two. Third, the PCR tests are taking days or even a week or more to report which means the results are significantly outdated and less actionable by the time they are reported.

The fundamental issue is this: if a test is cheap and fast we shouldn’t compare it head to head against the PCR test. Instead, we should compare test regimes. A strip test could cost $5 which means you can do one per day for the same price as a PCR test (say $35). Thus, the right comparison is seven cheap tests with one PCR test. So considered a stylized example. If a person gets infected on Sunday and is tested on Sunday then both tests will likely show negative. With the PCR test the infected person then goes to work, infecting other people throughout the week before being the person is tested again next Sunday. With the cheap test the person gets tested again on Monday and again comes up negative and they go to work but probably aren’t infectious. They are then tested again on Tuesday and this time there is enough virus in the person’s system to show positive so on Tuesday the infected person stops going to work and doesn’t infect anyone else. Score one for cheap tests. Now consider what happens if the person gets tested on another day, say Tuesday? In this case, both tests will show positive but the person doesn’t get the results of the PCR test until next Tuesday and so again goes to work and infects other people throughout the week. With the cheap test the infected person learns they are infected and again stops going to work and infecting other people. Score two for cheap tests.

Indeed, when you compare testing regimes it’s hard to come up with a scenario in which infrequent, slow, and expensive but very sensitive is better than frequent, fast, and cheap but less sensitive.

More details in this paper.

Comments

Why don't we just send them a link to a [0,1] random number generator with the probability weighted by the current case counts in their probable location? Seems safe enough, and they can take it every day!

Someone who understands the potential market place.

Though paper is cheap to print, so maybe people could buy a scratch test card with their lottery tickets, and really hit the jackpot.

In this thread prior_approval shows off his failure to grasp statistics.

But carry on, a public grudge from being fired might as well be innumerate to go with the creepiness.

Is accuracy important? It would seem to me that it is the most important thing in medical tests.

" It would seem to me that it is the most important thing in medical tests."

Pretty much with everything, the metrics are quality, time and cost.

Is a 99% accurate test that takes a week and cost $100 better than a 98% accurate test that takes a day and cost $25?

It the test is to determine the flight worthiness of a 100 million plane, then probably you'd want to go with option A, however if you are talking about diagnosing Covid19, most people would consider option B better.

There is not one optimal metric for every problem.

your mom is frequent fast and cheap

An intriguing discussion is worth comment. I believe that you need to write more on this issue, it may
not be a taboo matter but usually people do not talk about these subjects.
To the next! Best wishes!!

What i do not understood is in fact how you are not really much
more smartly-preferred than you might be now. You're so intelligent.
You recognize therefore considerably when it comes to this matter, made me in my
opinion consider it from numerous various angles. Its like women and men don't seem to be fascinated unless it's
one thing to accomplish with Woman gaga! Your individual
stuffs outstanding. Always take care of it up!

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

You are begging the question. Many of the Covid tests in use today have a 50% false positive rate. It wouldn't matter if they were free and took less than a second they would be worse than useless.

No, you just dont understand base rates.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

If you've had a medical test lately, you probably had a discussion of the relevant Bayesian statistics. Medical tests are far from perfectly accurate, so the choice of action almost always involves issues of prevalence and harm reduction. This is why tests that may lead to surgery are usually performed more than once.

There is probably some minimum accuracy at which point the test becomes worthless, but in this case, the cost of a false positive is an extra sick day, and the cost of a false negative an extra day of spreading the infection.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Yes, a coin flip would have 50% accuracy. I'm not sure a test with 80% accuracy is much help. This is why HIV tests aren't given to general populations.

Take 10,000 people, with 1 in 1000 prevalence rate, you have 10 with the actual disease. Testing them with an 80% accurate test will yield 2000 false positives and 2000 false negatives. Useless.

And given the 80% test rate, if you test daily, you'll see a false positive once a week. And then what? If you test at home, do you just take 4 or 5 tests until you get the answer you like? And if you have one positive and 3 negatives in front of you on Tuesday, do you go into work?

These tests have 80% sensitivity, but their specificity is normally greater than 95%, meaning few false positives.

That still makes the math hard. 5% is a "few" when the disease prevalance rate is so low. We are seeing 60K/day testing positive, and with a 14 day cycle from well to recovered, that suggests 840K people at any time have it.

So, a false positive rate of 5% is 20X higher than the prevalence rate.

But a false positive isn't very expensive. If you're testing daily, that person stays home one day, but tests negative the next and goes back to work. That's why the regime is important. A cheap, low sensitivity test is a poor way to sample a population for the purpose of epidemiological data gathering, but a great way to decide whether employees should go to work each day.

'that person stays home one day, but tests negative the next and goes back to work.'

Perhaps in a quasi-rational world. I'd worry that the first false positive would leave you expelled for a week or more. Abundance of caution and all that. No appeal to statistics would be tolerated.

We're talking about $5 tests that take under 15 minutes. You'd just re-test immediately. If you failed a couple then they'd send you home for a day, but probably allow you to retest again the next day if your weren't symptomatic.

If anything, the worry should be in the other direction. With employers telling employees to ignore the positive results because the tests aren't that good anyway.

Alex re-upped this on twitter, so this is the first time I'm seeing this. When women want to know if they're really pregnant, they take multiple tests, for the same reason - they aren't expensive and if you take enough of them, you get a solid statistical probability of accuracy, as long as they're not exactly 50% accurate.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

What do you think about Romer’s

“Even A Bad Test Can Help Guide the Decision to Isolate: Covid Simulations Part 3”

https://paulromer.net/covid-sim-part3/

As I recall there are several posts on this blog about that.

I do think your raise a good point though. But seems fairly easy to work out some general rules of thumb for how to interpret the results statistically.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

A lot of people would get two negative tests in a row then quit. I still get people telling me they have nothing to worry about because they tested negative in April.

While this is a problem for the cheap tests, it’s a smaller problem than it is for the PCR tests. Because with PCR tests, this behavior is harder to argue against. What are they supposed to do, schedule an inconvenient, costly test every week and then quarantine for 5 out of 7 days while they wait for the results (which are obsolete by the time they get them?) This type of user is not going to do it, nor should they. On the other hand if you tell them to take a quick, easy, cheap test at home every day, at least some of them will agree to it. Some won’t, but those would be no better off with PCR tests anyway.

Or rearrange the lawn chairs as the lava approaches, whatever.

Respond

Add Comment

Why is that an argument against trying to use the cheap home tests. Some might only do a couple of days even though the instructions say test 7 days. However, I suspect most of those most likely to stop early are in fact those most likely never to order the tests.

You're argument seems to be a long the lines of there are stupid people out there so lets not give smarter people a tool to use.

oops, that was supposed to post under Slan's comment.

Respond

Add Comment

Respond

Add Comment

How accurate are the underlying assumptions of this argument, namely that it takes a week to get a PCR test result back?

I just got a COVID test and got my results (negative) back in 2 days flat. I know another person who got it in the same timeline. Maybe this is just a great testing site, and they said they only got faster very recently, but if 2 days is doable, there's not much reason for less accurate one day tests.

Discussions around pandemic responses would be better if we could have an accurate assessment of fundamentals rather than assuming that everything is the worst possible status and arguing from there.

If I was just lucky and a week is a more typical timeline and isn't likely to improve, an 80% accurate daily test that gets more accurate in more severe cases has some use. Not sure home testing is the best use, but it would be great to see something like this at airports or work-sites that can't use telework.

Also, I was able to get my results a day earlier by signing up for an account with Quest. Entering my personal info allowed them to automatically associate my test with my account. I was able to get my results from them at the 2 day mark, but the email from my doctor with my test results didn't come until the next day.

No one told me I could set up an account with Quest to do this, I just happened to have some free time and wondered if it was possible. Telling people that 5 minutes of work can shave a day off getting results seems like it would be worth it.

Respond

Add Comment

An daily immediate test with just 50% sensitivity will give you your true positive faster on average than a daily 100% sensitivity test that has a 2-day delay. There is a formula for this here https://chris-said.io/2020/07/15/why-inaccurate-antigen-tests-are-better-than-slow-pcr-tests/

Only true if the false negatives are random. If the false negatives occur because the specifics of a person's disease, enzymes or other compounds in their saliva, poor execution of the test, or poor storage of the reagants or strips this won't be the case.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

My wife just got tested yesterday in New Mexico. Results in 7-10 days. Maybe it's due to the surge, but 2 days is not close to realistic here right now. So, what am I supposed to do? Stay home from work for a week until we get her result? Then, if it's positive, I need to get tested and wait another week. I am not symptomatic, so they will not even give me a test right now due to high demand.

Is your wife symptomatic? See CDC guidance on isolation. Cough, shortness of breath together. New onset of loss of taste/smell is 95% specific to COVID-19, so if you have that, by all means isolate! If she is, then yes, you belong in quarantine as a close contact. 14 days, regardless of any test result you get.. Understand that even if you test negative, your incubation period is 14 days, and you are most contagious during the 24 hours before you get symptoms. It’s all transactional. Your wife should isolate away from everyone, with her own BR/BA if possible. If you isolate away from her, then she stays isolated for 10 days after illness onset, and you quarantine for 14 days after your last exposure to her during the isolation period. Otherwise, she recovers in 10 days (longer if fever continues) and you still have 14 days of quarantine, again, counted from your last exposure to her.

Above is both CDC and San Diego County Public health guidance. I was a contact tracer for several weeks. Sorry for bad news!

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

"A lot of people would get two negative tests in a row then quit."

Exactly. But with a cheap test you can have everyone in the office take three times a week. Or everyone in the class. Etc. You can just do it at the start of every day on-site.

But 20% of your office will test positive every day. Everyone in your office will test positive in a given week if they are taking daily tests. How does that help?

That’s not how specificity works

+1

E25Bio is talkign about an antigen tests. Plenty of those have been found to have serious problems during covid AND historically. I'll agree if the specificity is 99.5% or higher, then it's very useful and you can fix any problems with more tests. But if specificity isn't that good, the worth of test quickly tends towards 0.

" I'll agree if the specificity is 99.5% or higher, then it's very useful"

What's magical about 99.5%? Yes, 20% false positives would be problematic, but 10% false positive might be fine if the test is $5 and takes 15 minutes.

You could reasonably have everyone who tested positive re-test when you are finished with the group. If the false positives were random, then the combination of the two tests would be 99% accurate.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

I don’t think that’s accurate (but I don’t know). My understanding is that the false positive rate of these tests is no higher than that of PCR tests, but the false negative rate is higher as they are less sensitive. Can someone who knows comment on this?

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

That really depends on who's administering the test. If the person is doing it at home... then yeah... they might stop after a couple of negative tests. If it's a requirement to get into one's workplace or school and is administered upon entry, then the number of tests is just limited by the supply and the organization's budget. If that's what got my kids' schools to open up, I'd pay much more than $25 per kid per week for my kids (and a few others if those families can't afford it) to get tested everyday.

So, a high school with 1000 kids....every day you test everyone and it reports 200 per day have it when you know only 1 or 2 does Then what?

That’s not how specificity works

Respond

Add Comment

"every day you test everyone and it reports 200 per day have it when you know only 1 or 2 does Then what?"

For a $5 15-minute test?

Well, then you retest the group of 200. Then you retest the group of 40. Then you send the 8 home and tell them to get a full test.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

If this test was available I am absolutely sure I would be taking one every day on my way into work.

Really a rapid test should be compared to the accuracy of taking your temperature to detect Covid. That's what it would be replacing.

So, better than nothing to close to worthless after a few months experience of using temperature as a screening method?

Shrug, I don’t see why they would be approved in the first place.

If/when they are approved I doubt many people would even bother to purchase them.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

This type of test would be great for dental offices. Using screening questions in conjunction with a rapid test would help keep infectious patients out of our chairs. We use high levels of infection control but it is very disconcerting to get a call 48 hours after you see a patient that they tested positive. Because of the cdc recommendations for health care workers, we don’t have to quarantine or test as long as we were utilizing n95, eyewear, shields etc. However, I would prefer not treating anyone who is infectious. This test would allow us to test 15 mins before the appointment. Anyone positive would be rescheduled. We would have them test in their car.

Respond

Add Comment

Respond

Add Comment

>really high accuracy levels of 80%

Oh, FFS. That is pathetically low.

That’s NPR for you. And dutifully vomited up here, of course!

Alex accepted that criticism and argued that even at crappier levels of accuracy you are still finding obviously infectious people.

That’s all that matters. Keep the messaging simple, it’s not even about finding who is infected. It’s simply about finding who is infectious.

Also with at home tests people will test constantly.

The current threshold for testing is absurdly high.

Respond

Add Comment

Randy is just anticipating the reactions of the teachers. If they're already utterly terrified of interacting with children, whose susceptibility to transmission of the virus is minuscule, they're not going to agree to reopen schools based on a test that doesn't meet the highest standards of accuracy.

Not so for older kids. A middle or high school teacher might interact with 125 different kids at work; that should frighten any teacher not young or in excellent health.

But with older students, the teachers' interactions can be limited much more easily to just standing in the front of the class, several feet away from the nearest student. It's only the little kids that require close contact.

The experience in Israel, where they shut the schools down again when the infection rate started rising, the younger kids will obey orders and the older kids can figure out that they don't want to get sick, but the middle school kids are impossible.

I can't blame teachers for being afraid. They're always coming down with the kiddy crud. Children are walking repositories of germs. Kiddy crud is annoying but not dangerous. COVID is dangerous.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Tabarrok's argument for the cheap test has a flaw: his argument is that it will reduce the spread because it will allow for more frequent tests, the flaw being that an unreliable test will encourage the person tested to go about her business believing she isn't infected, the false belief encouraging more promiscuous behavior than if she didn't "know". An unreliable test is worse than no test at all. I'm reminded of the argument for wearing a mask (other than N-95), the ambiguity about who would be protected believed to induce more people to wear them if they falsely believed the mask would protect the wearer. The flaw I have witnessed first hand among my many evangelical friends: their false belief induces them not to wear a mask and to put their faith in God instead. Readers might believe this is crazy, but unless one spends time among evangelicals one cannot appreciate what motivates them to do what they do. That they don't "know" the mask is intended to protect others from the wearer is what I call motivated ignorance. Last month I spent some time with a group of evangelical friends in the low country, none wearing a mask. When I asked why not, the uniform answer was they put their faith in God. And when I tried to correct their false belief about who is protected and informed them that the mask is intended to protect others from the wearer not to protect the wearer, they looked at each other, rolled their eyes, and almost in unison said that's more fake news. Their piety is more important to them than the health of others.

I really cannot believe that you still cannot acquire the knowledge that masks protect the wearer.

https://www.cleveland.com/coronavirus/2020/06/masks-may-protect-the-wearer-from-the-coronavirus-more-than-originally-thought-new-research-suggests.html

https://www.npr.org/2020/07/20/893227088/growing-body-of-evidence-suggests-masks-protect-those-wearing-them-too

https://www.dailymail.co.uk/news/article-8543869/Face-masks-help-protect-wearer-infectious-disease-experts-claim.html

Great sources--there was almost enough information in them to actually find a research paper (for non-hamsters) that demonstrated the amount of protection conferred by masks. Why not post the studies themselves rather than point to rambling articles that don't link back to any primary sources? Our (media) institutions are failing us (thanks, Obama).

In the age of "person, woman, man, camera, tv" these are probably weighty enough.

Respond

Add Comment

"Our (media) institutions are failing us (thanks, Obama)."

So, I'm going to stick up for Obama. It was the crappy (media) institutions that brought us Obama, not the other way around.

Respond

Add Comment

Respond

Add Comment

Raymond: Next time you see your Evangelical friends, could you pursue your questioning further. Do they go through stop signs, not get vaccinations, not go o doctors because they "they put their faith in God." And when they say it is fake news, ask them how they know it is fake news and what news would change their mind. You challenge arguments on this website, why not do it with your friends. I personally would like to know how they defend themselves layer after layer.

Respond

Add Comment

Respond

Add Comment

Still, the best and simplest message is "when we wear masks we reduce disease transmission."

There is no real need to go into ratios of protection, me vs you, because ideally everyone is wearing a mask when interacting with others.

Respond

Add Comment

If my evangelical friends "knew" that wearing a mask protects others, they would wear a mask. As it is, they don't even wear masks when around seniors. I know, wearing a mask to protect others defies the fundamental law of economics that one should focus solely on oneself (greed is good), and that doing so enhances the well-being of everyone. I really can't believe that one doesn't understand and appreciate this fundamental law of economics.

You don't have friends.

He thinks he has friends on MR.

Respond

Add Comment

Respond

Add Comment

While we are inventing non-existent fundamental laws, is the fundamental law of law that all lawyers who are not economically illiterate pretend to be so?

Respond

Add Comment

Respond

Add Comment

‘the flaw being that an unreliable test will encourage the person tested to go about her business believing she isn't infected, the false belief encouraging more promiscuous behavior than if she didn't "know".‘

1. These people are already going about their business.
2. If we can make it safer, going about your business has value!

Respond

Add Comment

I don’t follow how your evangelical friends not wearing masks because they think God will protect them proves that people do more risky things if they wear masks. Or get negative tests.

Respond

Add Comment

Will a false negative test cause some people to be more risky than they would otherwise?

Maybe a little bit.

Will this outweigh the massive benefits a testing regime provides?

No way, not even close.

It’s kind of like masks. They maybe make people behave slightly more risky, but nobody serious believes they have caused a net increase in viral transmission. I believe the story about your relatives but I’m not sure I see how it disputes this point.

Respond

Add Comment

Respond

Add Comment

Again at crappier efficacy levels it will still detect infectious ness. With the way this virus spreads via super spreaders, a linear decrease in super spreaders can create exponential decline in this virus.

Respond

Add Comment

I take the attitude that it's fine to push the FDA to play faster and looser. We just don't want that to be the only argument, and we want final policy to balance risks.

As we saw with the hand sanitizer recalls, sometimes fast and loose overshoots the goal.

Respond

Add Comment

Medical and care facilities come first, and an 80% accuracy rate is already a bit of a joke. Going below that makes the stylized example more than a bit of a joke. Let us use 66% accurate - 3 infected people take the test on Monday and go to work. On Tuesday, 3 people take the test, and one person goes to work.

Sadly, the continuing failure of testing in America is the sort of thing that leads some people to keep advocating inferior methods of testing to control a pandemic. And highlighting that failure as the reason to adopt inferior methods would be tragicomic in another context.

Absolutely horrendous comment. A complete no-nothing arguing against a testing regime he is absolutely clueless about.

Respond

Add Comment

Respond

Add Comment

... Covid-19 "Testing" has become an end in itself -- quality & accuracy of test "results" has become a minor consideration versus the fast quantity of 'reported' results.

Mindles panic in government actors and their media coaches has erased most rational analysisof the problem.

Respond

Add Comment

What's the specificity for the E25Bio test? I don't see that info anywhere. If it's low enough, people may stop using it if they needlessly have to quarantine multiple times.

Respond

Add Comment

Alex, it's not a mistake. It's politics. Please talk to your Public Choice colleagues.

Do you want to flesh that out? Why is it "politics" or even "public choice" to have a bad COVID response?

At a really basic level politicians want to be seen winning, and citizens want to be not dying.

"At a really basic level politicians want to be seen winning, and citizens want to be not dying." - But if the two come into conflict, the citizens must go. See Cuomo, Andrew.

The two are not really in a conflict.

(It is really an in-the-weeds discussion about nursing homes. But I guess when it's all you've got we'll hear it a thousand times.)

It’s an open question whether death rates actually have a positive effect on polling at the state level.

But yes, the thousands of dead nursing home patients is more of a “targeting the seen metric” story and how quickly that leads one astray.

Respond

Add Comment

Respond

Add Comment

Fwiw, on what I meant by weeds, the transfers to skilled care facilities are happening all over the country, whether it is done quietly or out loud. Here in Orange County CA:

"Dr. Paul Yost, who’s an anesthesiologist at St. Joseph’s Hospital in Orange, said it would be helpful to know how many hospital cases are transferred to skilled nursing facilities so the overall picture of the county’s healthcare system can be better understood.

Yost, the CalOptima board chairman, also noted the current system wasn’t designed for a pandemic.

“Our whole healthcare system is not designed something like this — a pandemic — that strikes a large percentage of the population,” he said. “It’s designed around providing high quality care around things like heart surgeries … but a global pandemic, it’s not designed for.”"

Two factors to drive this, even without mandate, is that hospitals want to free up beds, especially in the crunch, and insurance companies want a lower cost per day.

https://voiceofoc.org/2020/07/is-orange-county-turning-the-corner-on-coronavirus-or-headed-off-a-cliff-a-closer-look-at-the-numbers/

Respond

Add Comment

Respond

Add Comment

anonymous, I think you don't know the research on bureaucracies done by Public Choice scholars, starting with Gordon Tullock in the 1960s. Given what has happened with several federal agencies in the past 5 years, you should be able to understand why today public-health bureaucrats' interests are hardly aligned with those of the elected politicians (particularly, in an election year).

If you have any doubt about what I mean, I'm very much against the qualified immunity protecting bureaucrats suspected of committing crimes in 2016 and earlier.

The problem for me with that kind of general argument is that it seems to justify any outcome.

Wherever you go there you are.

Why? Please read Angus Deaton's column circulated a week ago:

https://www.project-syndicate.org/commentary/us-connecticut-compromise-1987-and-failed-covid-response-by-angus-deaton-2020-07

Forget about his first line and focus on his main argument to claim that the U.S. response to the pandemic has been an unmitigated economic and public-health disaster. It is a reasonable argument based on the divided power that the U.S. has had since 1787. You have to be careful, however, on how this division has worked and why one could have expected not to work well under the unusual characteristics of the virus, disease, and pandemic as well as those of the political context in which the pandemic takes place. Although Deaton's argument is reasonable, he makes the mistake of pointing to one side for the chosen policies when politics is a game and you cannot ignore the other party (in addition, I think Angus is wrong about some facts). He's upset because Dem proposals were ignored and so he calls for constitutional reform (he looks like Hillary blaming the Electoral College for her loss).

Also, bureaucracies are needed to design and implement elected politicians' decisions. That is, effective administrative units staffed by people with incentives aligned with the politicians'. I hope someone will look at how the relevant units have acted and performed during the pandemic, starting with their incentives to collect and process relevant information for good decisionmaking, and then with their incentives and actions to execute decisions (btw I think Dr. Fauci has failed terribly on both accounts).

From that:

"The problem is the lack of a central, enforceable national strategy in a country with a federal system that is ultimately controlled by local authorities responding to their own needs and perceived risks."

Sure, sure. That's why the US couldn't mobilize for WWII, except we did.

That quote denies national policy options, and just says wherever you go there you are.

It's the ultimate fig leaf for the Trump administration, that the constellation of other choices would have ended up the same.

It's the tautology that pandemic mobilization was impossible because we didn't do it.

I have no idea how bright people can believe such things.

Respond

Add Comment

To name one crucial national executive decision, the defense production act was not invoked for PPE.

What would you have expected to occur via the DPA?

In the WWII model, a lot of companies "not in the business of X" were mobilized for production.

https://www.popularmechanics.com/military/weapons/a23340620/singer-sewing-machine-company-45-pistol-gun/

So what would you have expected to occur via the DPA?

It's a dull game, but I can play. Keurig converts from K-Cups to masks, everybody wins.

The 510(k) approval process is about 6 months, so maybe by September Keurig would be up and running with their factory retooling complete and supply chains built. That gives you March of 2021 for final approval of the 510(k) and selling of masks.

I know the 2 packs of 50 masks at Costco per customer rule is annoying, but...

Not seeing how that changes anything

Was that the WWII rate, or is it you version of "no we can't?"

This entire blog post is about the FDA not changing its criteria even in the event of a pandemic. We'll know soon enough.

I'll bet against.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Maybe I should back up and explain what I think is really going on here . We have a right-of-center government which has totally messed up the response to a national emergency.

A certain sort of right-of-center person can't deal.

And so they will make various sophistic arguments that it could never have been another way. Thus, they are free of responsibility.

But what if that's false? What if it was in fact the right-of-center theory of government which blocked action? What if they could not have a robust national response because that is not what they saw the role of national government being?

In that case right-wing ideology produces the result, which corrupt right-wing handweavers will try to bury.

("Blame Fauci" is the perfect thumbnail for that. The guy was never given authority or control, but "blame him" as a distraction.)

+1

That's an interesting theory, because it allows for specifics.

What if it was in fact the right-of-center theory of government which blocked action?

This is refreshingly honest(!) because it creates a space for rigor and falsification. You have a clear falsifiable hypothesis:

Effective anti-pandemic action was blocked because of right-of-center theory of government.

Now we just need to test it.

Respond

Add Comment

As an example of your confusion, let me tell you that Dr. Fauci never was supposed to be responsible for deciding strategies and policies. Please read these two entries en Wikipedia

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

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

I think it was a mistake to establish a task force (remember it was established in late January). Read carefully what the entry says about it. Then think how and why its responsibilities changed and more importantly ask yourself why Dr. Faucie transformed it into a one-man-show (the second entry refers to him as a lead member).

He has been 50 years working as a bureaucrat, sometimes as a scientist (participating in group research) but mainly as a manager to execute the NIAID agenda. Read the Wikipedia entry

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

and you will find no reference to any assessment of this agency's performance since 1984 when Dr. Fauci was appointed Director (I'd appreciate it if someone refers us to a report assessing the agency).

Today you can find in mass media comments made by Dr. Fauci about the lockdowns. He has been talking about them for more than 4 months. That's beyond the scope of his agency and his personal work in the past 79 years.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Seriously???? I looked at the website Alex links to and there is almost nothing there. Is this just another hoax? Not a single word on how this technology works.

Of why we don't need the FDA to interfere with getting America back on its feet.

Respond

Add Comment

Antigen tests have been around for years and are well understood. LAMP tests are another option. The NYT has an overview of the technology, new and old, here https://www.nytimes.com/2020/07/06/health/fast-coronavirus-tests.html

Respond

Add Comment

Respond

Add Comment

Not saying this is necessarily your position, but it isn’t the administrative state failing here. I really wonder what deaths per million it will take to make people take things more seriously. People should just assume they are asymptomatic and shredding and act accordingly.

Looking at estimated K values, it seems like stopping super spreading events should be the focus. I don’t see how less accurate tests help in that regard. And I can imagine ways it could make super spreading events more likely.

If the estimated K values were higher, maybe this would make sense. The fact that K value estimates are so low and super spreading events happen before people even realize they have a problem and should get a test, makes me think testing of any kind isn’t going to stop this thing from burning now that it has gotten so big.

I think you need to start quarantining neighborhoods at the very least. If you don’t, you’ll just going to end up locking down entire cities. Once numbers drop, maybe you can take a more granular approach. Waiting for a positive text of whatever kind strikes me as just too flat footed.

Wouldn't it be progress if even a portion of superspreaders discovered their superspreadiness and stayed home, even if only for a fraction of their superspreadhood?

Respond

Add Comment

Respond

Add Comment

I guess it is obvious that regulations is not one of Tyler's areas of expertise. I guess he has never heard of accreditation bodies, liability attorneys, or false negatives.

Respond

Add Comment

“ The fact that K value estimates are so low and super spreading events happen before people even realize they have a problem and should get a test,”

Tests like these enable regimes where people are tested all the time (for example, twice weekly), so that would catch more superspreaders. With $5/test you could do this for $40/month/person.

Respond

Add Comment

Anyone who has studied Bayesian probability theory would know that even a 1% false positive or negative, in the real world, has something like a 50% accuracy when detecting rare events, which, if you are targeting superspreaders, they are by definition uncommon in the general population.

An 80% test is little better than a random number generator.

+1 for noticing false positives matter. But even if ten people have to stay home when only one of them is a superspreader that's a price well worth paying.

Respond

Add Comment

+1
There it is.

Respond

Add Comment

Is this argument functionally different from "everyone should just stay at home"?

Respond

Add Comment

"An 80% test is little better than a random number generator."

You are confusing the sensitivity of the test with a random number. Those aren't comparable. The test doesn't randomly miss people 20% of the time. It is sensitive to the quantity of the virus in the sample. The test misses people who have the virus but aren't shedding a lot of the virus at the time.

But flagging people who are in the highest 80% of virus shedders with a cheap, fast test is not equivalent to random signal noise.

Indeed, this type of test would most likely have a very high rate of detection for "Super spreaders" because they would have a high viral shedding load.

+1

You beat me to it.

Respond

Add Comment

Respond

Add Comment

Yes at some point this is angels dancing on pins, none of this is going to happen.

But for the sake of math:

Imagine you could pump these out to everyone. To get inside any retail or work space you would have to provide a time stamped test from within 24 hours prior. So everyone is taking these essentially daily.

The accuracy isn’t random, it’s directly correlated with viral shedding which is directly correlated with likelihood of spreading.

In that scenario the fact that you will get few false positives and some false negatives doesn’t actually matter that much. You’re going to prevent a lot of transmission regardless by keeping the most contagious people out of circulation while they test positive.

But none of that is going to happen, so...

'To get inside any retail or work space you would have to provide a time stamped test from within 24 hours prior. '

Considering how that has worked out with wearing masks, how realistic do you consider that idea?

He cannot imagine that idea happening either. To be honest, he is utterly skeptical of anything going right in the U.S., the reason for saying he will move to Taiwan if Trump is re-elected.

Respond

Add Comment

Yes at some point this is angels dancing on pins, none of this is going to happen.

But for the sake of math:

But none of that is going to happen, so...

Thought saying it twice was enough

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

The post mentioned this, but it is worth repeating:

Even if these tests were just as expensive as PCR (they aren't) and even if they had low sensitivity during the infectious high viral load period (they don't), they would still be massively superior due to their high turnaround time. I did an analysis showing how speed is far more important than accuracy, at least for routine mass testing.

https://chris-said.io/2020/07/15/why-inaccurate-antigen-tests-are-better-than-slow-pcr-tests/

Respond

Add Comment

Thank you

Respond

Add Comment

We'll be sitting around for months with our thumbs up our fundaments if we wait for the DAD to do something. Please E25Bio just market these as novelty items and let people have an option to use these at home. Far better than having to risk catch covid waiting in line at testing centers.

More on paper strip tests here: https://www.nytimes.com/2020/07/03/opinion/coronavirus-tests.html

Quick - do something even if it's wrong!

Authoritarianism now. Authoritarianism tomorrow.
Authoritarianism forever.

Never, never, never allow the little people make their own trade-off calculations.

You know best.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Most of the 'advantages' described in this article rest on the assumption that the tests are accurate... even after admitting that they don't meet FDA standards.
Given that, I'd much rather not see arguments from narrative - storybook arguments - but would want to see some math about the effects of various false positive and false negative rates.

The post links to a paper showing, quantitatively, that speed trumps accuracy. https://www.medrxiv.org/content/10.1101/2020.06.22.20136309v2

Respond

Add Comment

Respond

Add Comment

Need to separate out two goals: quick tests to decide if you can safely form a temporary bubble with high risk individuals like your grandparents, low sensitivity tests very very bad. Want to reduce R0, quick low sensitivity test very very good.

Respond

Add Comment

I feel like everyone in the comments criticizing this idea is assuming that the 80% accuracy number reflects that BOTH 20% of people who have the virus will get false negatives AND 20% of people who don’t have the virus will receive false positives. I’m not sure but reading the article, it doesn’t seem like that’s the case. The accuracy rate seems to be a result of the test not being able to detect low levels of infection. So it will be 20% inaccurate for people with low levels of the virus in their system (which also means they’re likely not very infectious). If that’s true (and I grant you we need clarification on that), I think most of the criticisms in the comments here just wouldn’t apply.

+1, there's no evidence that the test provides 20% false negative. If so that would be problematic. But arguing that fact in the absence of any evidence seems silly.

Respond

Add Comment

Selectivity and Sensitivity are what classical statisticians term Type I and Type II Errors. (Errors of inclusion and exclusion).

Respond

Add Comment

Respond

Add Comment

Health care authorities prefer to have people in queues and waiting areas instead of getting tests and medicines at home and using telemedicine.
I think I read somewhere that this is very pronounced in Italy which is why the first European outbreak of Covid-19 was so severe there.

Respond

Add Comment

Reminds me of that old cartoon you'd see on the wall in accounting departments: something like....

Fast Answers: free.
Correct Answers: $20.

Respond

Add Comment

Not a slogan for budget condoms, it turns out.

Respond

Add Comment

I can easily think of a variety of scenarios which you claim are "hard to come up with". This assertion is so ludicrous on its face that you should be embarrassed. I must be a gol-darn genius!

Not that that fact demonstrates the proposition (that fast is better than good) is false. I can think of settings where fast and frequent are better than good, expensive, and slow but wonder how many home testers would actually test more than once or twice. This is reminiscent of the sci-fi where people's toilets analyze their morning ablutions and eliminations and pronounce them healthy or not. Sure, it could be used well, question is would it likely cause more harm then good on a population basis. Should we assume people are lazy or that we are dedicated to protecting others (since there's no personal benefit in knowing you're infected earlier than when symptoms occur afaik (HCQ aside...).

Respond

Add Comment

Respond

Add Comment

The upshot: People with knowledge of the technical issues involved with these tests say that the economists are greatly oversimplifying things and that claims made about these tests are wishful thinking.

One person who has been involved with the FDA indicated that the FDA's mistake has been to permit a lot of very ineffective tests that have created confusion.

Spit tests have significant technical challenges due to enzymes in saliva. Self administered tests require that enough people do them correctly. Who would pay for the repeated tests? The prices claimed come from people who are looking for investors and support. They are the "best" case scenario. You can't base policy on the wishful thinking that entrepreneurs use in raising funds.

Economists: okay, first we assume a rational person....

Epidemiologists: um, you lost me right there.

Economists: then we assume full information...

Respond

Add Comment

Respond

Add Comment

https://www.cell.com/cell-reports/fulltext/S2211-1247(20)31001-9

The influenza virus hemagglutinin (HA) and coronavirus spike (S) protein mediate virus entry. HA and S proteins are heavily glycosylated, making them potential targets for carbohydrate binding agents such as lectins. Here we show that the lectin FRIL, isolated from hyacinth beans (Lablab purpureus), has anti-influenza and anti-SARS-CoV-2 activity. FRIL can neutralize 11 representative human and avian influenza strains at low nanomolar concentrations, and intranasal administration of FRIL is protective against lethal H1N1 infection in mice. FRIL binds preferentially to complex type N-glycans, and neutralizes viruses that possess complex type N-glycans on their envelopes. As a homotetramer, FRIL is capable of aggregating influenza particles through multivalent binding and trapping influenza virions in cytoplasmic late endosomes, preventing their nuclear entry. Remarkably, FRIL also effectively neutralizes SARS-CoV-2, preventing viral protein production and cytopathic effect in host cells. These findings suggest potential application of FRIL for prevention and/or treatment of influenza and COVID-19.

The terms glycan and polysaccharide are defined by IUPAC as synonyms meaning "compounds consisting of a large number of monosaccharides linked glycosidically".
-----
Thinking out of the box.

Attack the supporting sugar molecules that coat the virus. Anything to weaken the sphere. Starve it, crack it, coat it, poke it, whatever it takes. Our white cell overlords demand we do something or they will find other bioskin hosts and leaving us rotting corpses.

Respond

Add Comment

This is obvious to everyone who has a minimal grasp of the issues and basic mathematics, except the FDA bureaucrats. What does this say about the FDA and its staff?

What are the names of the bureaucrats responsible for this and other decisions that are responsible for creating hundreds of thousands of death? As it is now, saying NO just means taking zero risks of being possibly wrong while killing thousands of people with delays.

Note the FDA added 30 days delay for allowing the approved WHO and German testing for the US, while the virus was doubling in 3 days, which increases the problem by a factor of over 1000 times, eliminating the possibility of getting this under control. Who was the person that was more concerned about his bureaucratic fiefdom than the lives of the citizens?

What is the difference between these FDA bureaucrats and the Nazi bureaucrats in the SS that just followed orders, policies, and procedures? With the president calling this virus a WAR, perhaps we need a tribunal after it is over and we convict the bureaucrats for crimes against humanity.

My guess is that Trump will lose the election and that the Biden administration will ensure that there will be no deep investigation into the FDA and CDC. The narrative will be that this is completely Trump's fault, so that's where the investigations will be targeted. Any investigations into the FDA/CDC will be pushed by Republicans and the Democrats will attempt to minimize their scope or stop them completely to keep the narrative intact.

Thus nothing will change. Outside of reputation of course.

I hope I'm wrong. I wouldn't mind a thorough investigation into everything that the Trump administration did, the FDA did, the CDC did, TSA (why were there no temperature checks at airports), etc.

Respond

Add Comment

The politicians have the power to override the bureaucrats, and they often do. That's what the whole "deep state" argument is about; it's an excuse to override the bureaucrats. Chloroquine anyone?

This may be an argument for increasing the quality of our bureaucrats. Maybe we need to be paying them more and providing them with more resources.

Respond

Add Comment

I totally agree with Dr Weaver.
All bureaucrats and people in power should be made to put their names to decisions and be held responsible for them.
In Italy engineers who build bridges that collapse killing people face serious punishment. Bureaucrats who make decisions that delay perfectly good treatment to thousands of people resulting in their deaths should face similar penalties.
A sensible answer would be to advise people in need of treatment that certain treatments haven't been thoroughly tested, but have a probability of so much of working, and let them take the decision for themselves.
But of course that isn't nearly as much fun for the health authorities, who would lose a lot of their dictatorial power over people they like to call "patients". I hate the implications behind that word.
Writing in _A Fortunate Man - The Story of a Country Doctor_, John Berger says:
"We give the doctor access to our bodies. Apart from the doctor, we only grant such access voluntarily to lovers - and many are frightened to do even this. Yet the doctor is a comparative stranger."

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment