CDC reviewing ‘stunning’ universal testing results from Boston homeless shelter | Hacker News Reader

CDC reviewing ‘stunning’ universal testing results from Boston homeless shelter

boston25news.com

53 pointsg42gregory42 comments

https://news.ycombinator.com/item?id=22880335

Some possibilities: the number of asymptomatic carriers is higher than expected, these people were mostly infected at the same time and just haven't shown symptoms yet, these people were infected by a less severe variant of SARS-CoV2, the tests are simply incorrect, or the inhabitants of that homeless shelter are in some way non-representative of the population in general (e.g. they're all in their mid-20's).

They imply that those asymptomatic cases didn't all stay asymptomatic and that one was hospitalized, so I'm guessing the most likely explanation is just that most of them were infected around the same time, and the test just caught them before they started to show symptoms.

And another:

The infection is far more widespread than previously believed because a larger percentage of people are asymptomatic than previously believed.

That's mentioned in the first setnence of the comment you are replying to

Wow. So it is. I guess I skimmed over that and missed it on multiple rereadings. Dang.

I hadn't thought of that possibility. I'd thought that maybe the people with severe symptoms had gone to the hospital instead of the shelter.

I would love to see a follow up on their symptoms in 2-3 weeks. Not sure we'll ever see that info though, follow ups don't seem to sell eyeballs these days..

italy did this in a small town, testing weekly, by week three 80+% showed symptoms, and concluded they would all eventually show symptoms.

The CNN article has a picture of what a Boston homeless shelter looks like:

https://cdn.cnn.com/cnnnext/dam/assets/200417103746-01-pine-...

So yeah, it's a bunch of tightly packed beds. One or two asymptomatic carriers can probably infect hundreds in just a night.

Those appear to be just PCR tests that tell you if the person has an active infection or not. Those that are negative should get a serology test to see if they were previously infected. The combination of both of those tests would be a true indication of how far this has spread.

Of course, given how infectious this virus is, it's no wonder that half have been infected in the last 1-2 weeks at the shelter. There's probably little to no safety measures in place to help isolate homeless people at that shelter, so it would spread like wildfire.

Yeah, it seems like it spreads to everyone in every nursing home it hits, so it'd be shocking to see it fail to spread in a homeless shelter.

In a nursing home I would've expected nurses to be transmitting COVID-19 between patients.

But merely having a lot of people in a room be enough for widespread transmission is a bit of a wake up call. I can't understand how given that you could ever open up bars, restaurants etc.

Well, a lot of people in a room for at least 8 hours, and unless this is a one-day shelter probably iterated over quite a while. I don't think this suggests a story where you sit in a restaurant for 2 hours and then everyone's infected; there's only been one report of a case like that, and multiple contact tracing reports from other countries where a carrier only infected a few other people in their bus or restaurant.

Transmission via asymptotic carriers isn't a new thing. It's been known for months (e.g. the analysis of the first cluster in Germany) and is one of the main reasons this virus is so problematic.

I am actually surprised people aren't more concerned about the long term effects of COVID-19.

We have seen patients who got through it but were left with persistent neurological effects i.e. loss of smell, heart damage, weakened immune systems etc.

If people are asymptotic but have reduced life expectancy then that is a really big problem. Especially if we are talking about just letting COVID-19 basically run wild through healthy people.

> We have seen patients who got through it but were left with persistent neurological effects i.e. loss of smell, heart damage, weakened immune systems etc.

Not a significant percentage of patients by any means.

Lasting lungs damage is expected in ~5% of hospitalized cases (not counting ICU/reanimation protocol cases) in eastern France. I'd says it is a significant percentage. We will need to wait a couple month to be sure it is irreversible (pulmonary fibrosis are irreversible for the moment, as we can't chirurgically extract damaged lung tissue yet) and also the extent of this damage.

A young lady got asthma for the first time in her life last sunday, asthma that did not left her at least until wednesday (when my cousin told me about it, no news since). As nurses and doctors there now can afford to take vacation now (two-day vacations are vacations when you're working everyday since early march), they will probably have some time to start evaluating the non-deadly damages sometime soon.

> A young lady got asthma for the first time in her life last sunday, asthma that did not left her at least until wednesday (when my cousin told me about it, no news since).

Anecdotal. While you say lasting lung damage is expected, you also say that a couple of months are needed to be able to prove the expectation.

Have there been any studies on fertility?

We had to do IVF, but aren't there movies about an 'innocent' low-symptom virus that ends up leaving half the world or more unable to bear children?

That could be an issue if we want to continue to exist as a species. Good for the environment though!

> “It tells you, you don’t know who’s at risk. You don’t know what you need to do to contain the virus if you don’t actually have the details or facts,” said Marty Martinez, Boston’s chief of Health and Human Services.

Plans to reopen states are conditioned on testing, but how long --and will we ever-- have enough testing data (among the general population, not just those showing symptoms) to make the informed decision?

If we can't get the cotton swabs, the reagents, with the ability to test nationwide at scale, health policy experts will not be positioned to green light re-opening.

Well to do folks can work from home, or not work from vacation homes, have all their needs delivered, and not catch this thing. All the "essential" people will catch it and some will die, and the well off will consider this very cheap, good, proper, and mandatory for economic reasons.

It's hardly worse than the exploitation for profit of the tobacco industry. Known for decades it would maim and kill, but millions invested in, were shareholders, were owners. And they've never been sanctioned. This will be no different. The lower class will take one for the team, again, for the greater good of society so that the better classes can live with minimal inconvenience.

By all rights, essential workers should quit. But they have a boot on their head: landlords and debts. It's economic coercion.

It's been well known for over a century that smoking is a slow poison. The well off smoked as much or more than the not so well off.

My dad, who went to high school in the 1930's, said they were known as "coffin nails" then. Any doctor who cut open a smoker knew what smoking did to the lungs.

That's a very cynical view. Doctors are by no means lower class. Most if not all of them can quit, and are not subject to "economic coercion". Yet almost of them choose to work and overwork.

I am not an "essential" and am working from home, but I donate and try to help my local community, and if I am called to help in other ways, I will.

And no, I don't actively invest in tobacco.

Many doctors have a substantial amount of debt from med school that they took on with the assumption of having a doctor's income.

I think there's a common misunderstanding here. The value of testing is that it enables effective contact tracing. (Ideally, you want quick and regular tests for every identified contact, so that if they do have it they can be isolated before they start coughing.) There's no data-driven strategy to compute how safe it is to open up - that's determined subjectively based on general trends.

Soon there'll be a different kind of pressure. The Smithfield pork processing plant in Sioux Falls, SD is closed because of 500 coronavirus cases amongst the employees. Several other plants are idled now because they are downstream of the South Dakota plant. This is a cut of 5 % of the pork supply in the United States. Several Tyson chicken processing plants in Georgia also have coronavirus outbreaks.

Everyone thought that food shortages due to the COVID-19 pandemic were a very remote possibility, but it seems they are coming, and the government can't permit that. Food riots are precursors to revolutions.

I wonder what the age breakdown was of those tested. It seems like there have been a couple of (very preliminary) indications that prevalence may be higher (and thus case severity lower) than we thought. It's probably too early to be optimistic, but I'm looking forward to getting more and more cases with better population samples than "people who think they have it", to get a better sense of what the rate of symptoms is.

These articles really need to state what kind of testing was done: RNA or serological.

Very different conclusions if 50% test positive for one vs. the other.

Probably better off on the street in the open air, than a 'shelter' full of asymptomatic COVID cases...

The most stunning testing result is the recent Stanford blood study in Santa Clara showing that potentially 50 times as many people have been infected with Covid as previously thought.

We might already have 50 million Americans that have had Covid before another couple weeks is up.

Of course the media won't cover this story widely, as it won't sell properly, there isn't enough sensationalism. It lowers panic (it would mean the Covid mortality rate is radically lower than previously feared, even versus common low estimates), rather than raising panic.

It's crazy that the US isn't putting a lot of manufacturing and money behind doing widespread blood testing to see how common Covid actually is. This should have been done a month plus ago. Based on deaths, 10-15% of NYC probably has already had it at this point and if so, we need to know that asap. I've been talking about this for months now, it's not a premise requiring any kind of specialized knowledge to grasp (obvious question: how widespread is the virus), and I've yet to see any media interest in it despite how blatantly obvious it is.

It's almost like there is some interest in ensuring we don't find out how far Covid has already spread. The first thing they should have figured out, is the last thing they're apparently going to figure out.

The Santa Clara study has gotten a fair bit of press and attention but it's not clear whether the results will hold up. The antibody test is new and they're assuming very few false positives, and it's a non-random study (recruitment via Facebook).

Here's some more criticism: https://medium.com/@balajis/peer-review-of-covid-19-antibody...

It's why we should be doing massively increased blood testing and test manufacturing to resolve that question. For the last few months it has been the single most important question and they're putting very few resources behind trying to answer it.

We can't test blood properly? Of course we could. They are choosing not to, is the actual context. It's a large mistake.

How many infections have you already had? How widespread is it really? No answer, no attempt to answer at scale. At its best, it's immense incompetence in action by the authorities in charge of dealing with this mess.

> It's why we should be doing massively increased blood testing and test manufacturing to resolve that question.

Yes, literally everybody agrees with this.

> For the last few months it has been the single most important question and they're putting very few resources behind trying to answer it.

This is not true, every available resource is being deployed to expand testing capability. Again, literally everyone wants to.

> We can't test blood properly? Of course we could. They are choosing not to, is the actual context. It's a large mistake.

No one is choosing against testing today. The large mistake was 3 months ago when the federal government should have started activating all of this.

> How many infections have you already had? How widespread is it really? No answer, no attempt to answer at scale. At its best, it's immense incompetence in action by the authorities in charge of dealing with this mess.

Totally agree on the incompetence, but again, the critical moment passed months ago. Everyone is desperately trying to catch up now.

You seem to conflate antibody testing with viral rna testing. The former is somewhat difficult because you can't make the reagents necessary overnight, and the reagents are unique for the virus. Essentially, you have to clone the gene for an immunogenic portion of the virus (the spike protein most likely), express it in a recombinant host (which may require extensive optimization), purify it, and then validate that it properly detects blood of people who were confirmed positive for viral RNA and then recovered. That all takes time, and then you have to scale up manufacturing. In contrast, the the viral rna (RT_PCR) test takes about a day to design and the reagents are synthesized overnight and then you can run the test in the morning of the second day. Scaling that up to 1 million viral RNA tests a day should add at most a month, but our country already runs this number of tests a day if you count every academic, research, hospital, and CLIA lab, so the manufacturing capacity already exists.

It's important to acknowledge, though, that most of the criticisms in this article apply even more strongly to the official reporting model. (For example, the study might have accidentally skewed towards people with symptoms, while the official numbers are intentionally and maximally skewed towards them.) So the criticisms don't argue that the real number is closer to the 1,870 on the county's website; they argue that we really have no idea how many people have been infected.

Yes, we have little idea about non-symptomatic carriers.

However, combining a prescreen with a test does significantly reduce the chance of false positives.

1. That's just one persons opinion. Don't treat it as fact.

2. Read the comments in that article for some counter-points.

The point is, this can't go on and thank god we're opening up everything again.

> Of course the media won't cover this story widely, as it won't sell properly, there isn't enough sensationalism. It lowers panic

Actually it is being covered in the media (e.g. https://www.cnn.com/2020/04/17/health/santa-clara-coronaviru... ) and the reason it’s not being trumpeted everywhere is because there are very serious statistical issues with the study that make it irresponsible to draw the conclusion that 50-85x the number of people are infected as previously thought. https://twitter.com/AlanMCole/status/1251493794039705600?s=2...

"50 times as many people have been infected with Covid as previously thought.

That is a stunningly incorrect analysis. The results suggest (and I use suggest deliberately, because there are a lot of sample bias/ data analysis issues here), that potentially (again, deliberate as this is a ceiling, not a floor) 50 times as many people have had COVID as confirmed cases. Well. Duh. We've known that asymptomatic carriers exist since before it even hit the U.S. Moreover, please note that this "50 times" number equals 5% of the population. We have 95% more to go before it burns out. There is plenty of reason to panic. If we opened up again and went back to our regular lives, we'd have about a million dead in two months.

Of course, everyone is desperate to figure out how far COVID has spread, but we just don't have enough testing capacity right now. Because the government has just fallen flat on its face. We can now begin to understand why. As Fauci said in the Friday press-conference, from the beginning they trusted private industry when they promised they would have millions of tests a day capacity within weeks. I have personally run over 5,000 samples of this kind of test (not covid specifically but viral RNA RT_PCR assays) in one day. It's no rocket science. It's trivial. The only reason we are STILL?! at this laughable number of 150k/day tests is because our country is currently run by the most incompetent administration in a century.

# of deaths doesn't even matter really... what matters most to me - is what this would do to service workers, healthcare workers, transit workers and the ICU's. Even if this proves to be as fatal as the flu, it's the long ICU stays and the toxicity of higher viral load on those who are more in contact with people.

You can't deny what's happening in NYC right now. They're under attack. This would be every city in America, we'd maybe lose 5-10% of our hospital staff across America. Schools are closed now, what about med schools? Will new doctors come out ready to work next year or the year after or will they be set back and there be a shortage?

Lots of systemic damage beyond just the fatalities could happen as a result of negligence. We need to be extremely careful.

> Of course the media won't cover this story widely, as it won't sell properly, there isn't enough sensationalism. It lowers panic (it would mean the Covid mortality rate is radically lower than previously feared, even versus common low estimates), rather than raising panic.

I think there is a common misunderstanding here, which is that the mortality rate of COVID-19 is an important factor in policymaking.

It’s not. The actual number of illnesses and deaths is what is important. That’s what maps onto healthcare capacity, which is the limiting factor.

And it is what people really care about. I might find it interesting that there are a large number of asymptomatic carriers, but what I really want to know is whether my friends and family are safe.

You’re right about mortality rate, but the percent of the population that has been exposed is critical for policy, because it tells us how big any upcoming surge could be. And it’s looking more and more like we got that one badly wrong, so we’re going to end up with thousands of outdated ventilators, a boatload of new national debt, and an even more jaded culture that totally ignores policy folks. Could be worse.

Reply on news.ycombinator.com