Hacker News Reader

Comment by dboreham | Hacker News Reader

>This virus is NOT currently spreading widely in the United States.

Curious how this is known given practically zero tests have been done and numerous cases with no contact external to the US have been discovered.

View on HN

Oh really: https://twitter.com/trvrb/status/1233970271318503426

"This strongly suggests that there has been cryptic transmission in Washington State for the past 6 weeks."

You mean as described in, and entirely consistent with the CDC summary?

Yeah, this is absurd. Surveillance testing isn’t even being done at anywhere near the levels to make this claim, nor are any of the only known effective mitigation strategies, like bans on mass gatherings, being put in place.

Compare this to what is being done in Singapore with effective quarantines and extensive contract traces, or Korea with on-demand, drive-thru testing, and the complacency is staggering.

The CDC isn’t even releasing treatment details for recovered patients so doctors and hospitals can prepare to respond:

https://amp.cnn.com/cnn/2020/03/01/health/coronvirus-patient...

Before you advocate 'surveillance testing', you should provide a link to hard data showing what the false-positive rate is.

Not assurances of how accurate it might be...

I want a definitive false-positive rate.

Until we have that, surveillance testing may be more harmful than beneficial.

That’s insanity. The WHO themselves insisted that aggressive surveillance testing is necessary to formulate an adequate response and it’s what Korea and Singapore are doing right now. There have been no reports of widespread false positives, only false negatives, and even in that case, testing was ramped up because it’s better to catch some than none.

Sigh...

I wrote: "Not assurances of how accurate it might be..."

Yet, that's exactly what you responded with...

Once again, what is the false positive rate?

If you don't know, so be it. But, if its 5% or so (like many similar viral tests), you may wish to give some thought as to what you are going to do with the tens of thousands of healthy 'false positives' your proposed surveillance testing will uncover.

This will be based on models using what we currently know about the time it takes patients to show symptoms, the number of people currently detected showing symptoms, and the expected rate of detection of symptomatic patients.

'Numerous' is a bit of a loaded term, there have been cases of community spread in three areas so far, as described in the CDC summary.

> The coronavirus has been circulating undetected and has possibly infected scores of people over the past six weeks in Washington state, according to a genetic analysis of virus samples that has sobering implications for the entire country amid heightening anxiety about the likely spread of the disease.

https://www.washingtonpost.com/health/coronavirus-may-have-s...

Yes, correct, scores of people in the areas the CDC report there being cases, and in which they say people are at increased risk. That's not 'widely in the united States'.

They also say that widespread transmission is eventually likely to occur. They are simply, and as far as I can tell very clearly and responsibly, drawing a distinction between what is happening now and what is likely to occur in the future.

As there has been community transmission and aren't actively testing symptomatic people (unless in known direct contact with someone who has traveled to an affected area, or traveled there themselves), it's questionable if they have a sufficient grasp on "what is happening now".

>and aren't actively testing symptomatic people

The article even has pictures of the test kits they've been distributing.

“Widely” is sort of a vague word.

It might be only in a few states but Florida, California, Washington, and New York could have a lot.

It is suspected that in Washington the virus has been spreading undetected for 6 weeks.

There's a model floating around out there now [0]. I think the main point of it is the transmission efficacy. The Gates article recently published in the New England Journal of Medicine states the average person infected will go on to infect two to three additional [1].

I'm of the opinion that, while China appears to be the root cause, the advantage they have is a very authoritarian control on the response. People were literally welded into their apartments [2]. In the US, for example, there are many people that will blindly ignore all requests by politicians, local, state and national government. Many Chinese don't have that option for fear of strong repercussions. While I'm not condoning that use of force - I'm just stating as an observation. Beyond that the US seems to be in an odd spot given the upcoming election and divisive divide along party lines.

[0] https://drive.google.com/file/d/1bS6S4oNRuMHD4ezh177QSf3UU6a...

[1] https://www.nejm.org/doi/full/10.1056/NEJMp2003762

[2] https://www.cbc.ca/player/play/1703503427818

The question about China's draconian response is, was it an overreaction? They are doing a lot of damage to themselves by welding people in to their apartments and shutting everything down. Maybe it will turn out that the Korean approach of testing, canceling events, and you know, not putting people in camps, will end up with less overall damage.

This was my first thought! My child here in Oregon has all the symptoms (fever, cough, and respiratory issues) and was exposed to a VERY close proxy of the person here in Oregon that recently tested positive. The doctor at the urgent care (we've gone twice now) even heard crackling in the lungs. They labeled it as the flu (note: my kid was vaccinated this year) but they didn't test for the flu and they only prescribed steroids to clear up the lung issues.

Given the mismanagement I've seen first hand, at this point I am reasonably sure the virus is indeed wide-spread.

EDIT: Just to clarify in case anyone thinks I'm being overly paranoid: the person that tested positive in Oregon is a school employee and acquired it via community-based transmission. The kids that my child goes to school with have siblings that were directly exposed to that person. The time-frame fits within the two week incubation period as well.

Criteria as set out by the CDC: https://www.cdc.gov/coronavirus/2019-ncov/hcp/clinical-crite...

> Fever or signs/symptoms of lower respiratory illness (e.g. cough or shortness of breath) AND Any person, including healthcare workers, who has had close contact with a laboratory-confirmed COVID-19 patient within 14 days of symptom onset

For healthcare professionals to brush off these facts as "the flu" without first performing a flu test to validate such a claim is irresponsible and indicative of an issue where this virus has gone undetected with valid cases being mislabeled as the flu despite evidence suggestive of Covid-19. Further supported by: https://www.washingtonpost.com/health/coronavirus-may-have-s...

One thing you can be sure. This will be a political football played aggressively by both or multiple sides trying to get an edge...

And the news of course will amplify things —they have since it emerged in China where they vacillated on whether it’s an overreaction or whether its a economy destroying pandemic.

Especially with news like this:

https://www.expressnews.com/news/local/article/CDC-Coronavir...

It is Coronavirus pandemic only if you are testing.

If you are not testing - it is just a regular flu.

That is probably why they capitalized "NOT": To deflect from the fact that they do NOT know anything.

And for extra security they added "widely", which makes this statement stretchable after-the-fact to fit any imaginable reality.

They are using a high threshold of "widely", ignoring the math of exponential growth and what that says about the next several weeks.

Easy. %20 of the cases require some type of hospitalization. %80 are mild and don't require medical attention. If you have Covid-19 rampant in your town - the hospitals will fill with persons experiencing a bad flu - that %20.

https://www.businessinsider.com/coronavirus-80-percent-cases...

>%20

>%80

that's not how percentages are written, ever.

Is it still the case that you have a two week latency before symptoms? If so, that would make detecting the early stages of an outbreak without testing exceptionally hard.

Last I read it was anywhere from 2 - 14 days.

> Current estimates of the incubation period range from 1 to 14 days with a median of 5 to 6 days, although recent case reports suggest that the incubation period may be as long as 24 days, which is longer than the 14 days that WHO and the US Centers for Disease Control and Prevention (CDC) have been using to inform quarantine policies.

https://jamanetwork.com/journals/jama/fullarticle/2762510

Here is some data from 1100 early cases in China. Median incubation period is 4 days, interquartile is 2-7 days. 14 days is a very distant outlier, and it's possible there was an intermediate carrier or two in those cases.

https://www.nejm.org/doi/full/10.1056/NEJMoa2002032

I think that's true only if transmission is possible before noticing symptoms.

Relatively rare but transmission between asymptomatic people has been detected/confirmed for this virus.

Yeah, this is more serious than the flu, but not as serious as MERS or SARS. Like the seasonal flu, the main worry is older people and/or people who are already respiratory compromised. I don't want to catch this much like I don't want to catch the flu, but my main worry is for my parents and grandparents.

This is way more serious than the flu in multiple ways. First of all, estimates show that the case fatality ratio (CFR) is around 2%, which is 20x higher than the flu. This number is likely to rise when ICUs are becoming over-constrained and cannot take new patients. This is already happening in the north of Italy as we speak.

Second, the transmissibility of the SARS-nCov-2019 (the virus causing COVID-19) is much higher than the seasonal flu. This means the period within which a demand will be made on healthcare services will be much shorter and much more intense.

Third, there are reports showing lasting damage to kidney and lung tissue (sponge-like structures). Even when surviving COVID-19, previously healthy people might become chronically ill.

Fourth, in order to prevent at least part of what I am describing, very stringent measures are required. These already have a historic impact on our worldwide economy and will hit developed countries the most (since they require a worldwide infrastructure). It is very likely that the impact will become unprecedented as various continents are considering and implementing large-scale quarantine measures.

Concluding, comparing COVID-19 with the regular flu is both simplistic and dangerous. Prepare yourself and your loved ones for a bumpy ride.

Care to explain the down-voting?

I'll take a stab at why you're being down-voted.:

>case fatality ratio (CFR) is around 2%, which is 20x higher than the flu.

[Citation needed]

>This number is likely to rise when ICUs are becoming over-constrained and cannot take new patients. This is already happening in the north of Italy as we speak.

[Citation Needed] I can find citations for crowded hospitals in popular media, but nothing about the death rate increasing.

>Third, there are reports showing lasting damage to kidney and lung tissue (sponge-like structures). Even when surviving COVID-19, previously healthy people might become chronically ill.

[Major Citation Needed]

>very stringent measures are required

[Citation Needed]

That's probably why. There are 4 claims you've made that need to be backed up with facts.

HN is not Wikipedia. I am more than happy to provide evidence for my claims, but I'd be downvoting almost every single comment on HN if those were the requirements.

The CFR is easy to find with a Google search. There are multiple studies currently providing strong lower and upper bounds on CFR.

My second statement has the word 'likely' in it. W.r.t. Italy, I am talking about an Italian media article describing over-constrained hospitals and the provisioning of military field hospitals.

Third statement has some evidence in other parts of the thread.

Fourth statement: many, many experts are warning that we are doing too little to late. The need for a response is logical given the high R0, CFR and needed hospitalization. If needed, I can provide some links, but going to my next meeting right now.

HN is not wikipedia, correct. But there is a baseline for expectation, from what I can tell. If you're going to make claims of fact, especially for something controversial, you need to support with facts. That's your job as the person making the claim.

Otherwise, you need to put a sentence in there that all of these things are just your opinion.

And your response to me is just more hand-waving "i'm right i'm right" with zero content. You're still doing it.

> Third, there are reports showing lasting damage to kidney and lung tissue (sponge-like structures). Even when surviving COVID-19, previously healthy people might become chronically ill.

you can't say that and not point to a source.

Well, this [1] is an article describing extensive GGO [2] and subpleural parenchymal bands in recovered patients. I have read other reports from Wuhan doctors describing these effects, but some search terms seems to be censored these days. I am not a specialist, but it was described as lasting damage.

[1] https://pubs.rsna.org/doi/10.1148/radiol.2020200370

[2] https://en.wikipedia.org/wiki/Ground-glass_opacity

Here's an article from the Lancet describing patients with high rates of organ function damage. It's not as useful as the one linked by edejong because it's a very early study. It shows that quite a few patients already had damage on admission, more than those that subsequently died so clearly many patients survived with organ damage, but it doesn't follow up on how many more may have suffered such damage after admission. So it's evidence, but not as strong.

https://www.thelancet.com/pdfs/journals/lancet/PIIS0140-6736...

> First of all, estimates show that the case fatality ratio (CFR) is around 2%, which is 20x higher than the flu.

These estimates are mostly from China. Obviously likely to be higher for a respiratory condition in a place with some of the worst air quality in the world. There is also some question about whether the less serious cases are even being identified -- if your sample is only of the people with severe symptoms, you get the wrong number for the percentage of people with severe symptoms.

> Second, the transmissibility of the SARS-nCov-2019 (the virus causing COVID-19) is much higher than the seasonal flu. This means the period within which a demand will be made on healthcare services will be much shorter and much more intense.

Fortunately flu season is coming to an end, so there will be less contention for resources there and the same reasons people get the flu less often in April than January will help to mitigate this.

> Third, there are reports showing lasting damage to kidney and lung tissue (sponge-like structures). Even when surviving COVID-19, previously healthy people might become chronically ill.

That can also be true of the flu, isn't it?

> Fourth, in order to prevent at least part of what I am describing, very stringent measures are required. These already have a historic impact on our worldwide economy and will hit developed countries the most (since they require a worldwide infrastructure).

This is probably just not going to happen. They'll try to contain it using strong measures when the number of cases is small enough that there is some hope of that actually happening, but strong measures aren't that problematic when they're only applied to a small number of people.

If they fail in the early stages and millions of people become infected, the strong measures can't be justified at a large scale because once it's that widespread they won't be effective anyway. The point of the strong measures is to prevent that from happening. They aren't worth continuing after you've already lost.

The estimates seem to be reflected by international numbers to a certain degree, ranging from 1%-3%. With the disease in exponential growth, dividing #death/#cases is not the right calculation to make, as it might take 3-5 weeks before a patient dies from the disease. The Princess Diamond ship is currently at 1.2% iirc.

Your factor of under-representation of cases seems to cancel out the delayed onset of mortality at the moment.

The end of the flu season is incomparable with the demands on the health-care system due to COVID-19 since flu-demand is at least 1-2 orders of magnitude smaller. The advantage of the end of the flu season is that we can identify cases quicker.

W.r.t. lesions caused by COVID-19 comparable to the flu. Yes, both cause permanent damage. However, COVID-19 seems to attack more of the lungs and in general, both lungs, whereas influenza often is isolated to one lung. I am not a specialist, but I have read scary descriptions of COVID-19 lungs from specialists.

The stringent measures are already happening and the effects are already being felt. The stock market has responded by evaporating 4000 billion in value. Given the past lackadaisical response of government, it is likely we will see further effects either way: response or not.

If nothing is done and millions of people will get infected (think: 2 billion people) in a short amount of time (1 year), we will see widescale breakdown of services, global turmoil and panic. Hopefully, we will be able to slow the spread, develop a vaccine and 500 million people will get infected instead.

There’s something of a social stigma around treating this thing seriously but alarm bells should be ringing at this point.

We have, essentially, a hurricane off the coast (so to speak) and no one can say with certainty whether it will blow over or whether people are going to take a direct hit.

Like with storms, there’s the fear that hyping it up too much will lower trust in the future. However, not preparing people and then having the worst happen is many times worse.

This thing should absolutely be taken seriously. The responsible thing at this point is to prepare for worst and plan. The worst being an overwhelmed healthcare system because too many people require hospitalization all at once.

If the hurricane blows over, we’ll have stocked up our emergency kits that we, frankly, all should have had in the first place.

But we should really stop comparing this to the flu.

I'm not sure we have the confidence to run the logic that way right now. There are still some ways to harmonize "this hit China really hard" and "this won't impact much of the rest of the world" at this point. I've seen it hypothesized since before it even spread out of China that this may happen to target some people more than others based on different levels that people express the receptor that this virus targets. There's also some reason to wonder if China is getting hammered due to widespread smoking [1], and even more widespread exposure to pollution levels that are not entirely dissimilar to smoking. If, perchance, all of these factors are true at once, then it could be that what was in China was a perfect storm for them, and it'll be just a cold for everybody else.

We still don't really have the evidence in yet to know. At the moment we can fit the facts both to "everyone's in big trouble" and "it isn't that big a deal anywhere else". In some ways, the best possible news would be that it's been spreading in the US for weeks... and we didn't notice, because it's literally not noticeable because it's not that big a deal here. I'm not pitching that as "truth" at the moment, but it's on the table.

[1]: https://www.preprints.org/manuscript/202002.0051/v1 - note this preprint disagrees with the "differential expression" theory; given it's "early days" I'm taking this merely as a data point, not proof yet. This could explain how it's a catastrophe in a high-pollution city and just "a cold" everywhere else. This scenario raises the interesting possibility that covid-19 could have started anywhere, and it simply wasn't noticed until it happened to get to Wuhan through normal mechanisms. What if "patient 0" turns out to be an asymptomatic business traveler from $NOT_CHINA? I'll call this a low-probability outcome, but what if we start widespread testing in the US only to discover that almost everyone already has it?

You should back up your claims.

The CDC reports that over 470 tests have been done to date. Is that practically zero?

https://www.cdc.gov/coronavirus/2019-ncov/cases-in-us.html

"numerous cases" You'll have to back that up as well given the linked article clearly states the cryptic transmission in Washington is accounted for. They call it "Community Spread."

Yes, that's practically zero, when compared to other places doing tests.

not even a rounding error in the grand scheme of things.

Switzerland has done 1300 tests to date and has the capacity for ~1000 tests a day: https://www.swissinfo.ch/eng/covid-19_switzerland-confirms-s...

So, yeah, I'd call that practically zero for a country with 38 times the population of Switzerland.

https://www.cnbc.com/2020/03/02/coronavirus-new-york-city-do...

Reply on news.ycombinator.com