Hacker News new | ask | show | jobs
by chrisseaton 2280 days ago
> 15-20% of infected people require hospitalization for respiratory problems

This is wrong - it's 15-20% of identified, diagnosed and subsequently monitored infected people, isn't it?

I thought there was a mass of unidentified infected people, and even basically diagnosed but told to just deal with it at home with no further contact as they're low risk and minimal symptoms, and (obviously) 0% of these groups are going into hospital? This is what Wikipedia says at the moment.

Or am I wrong?

6 comments

You're absolutely right. But does it really matter ?

A lot more ventilator are going to be needed, not in % but in hard cold real absolute number. Isn't that more important ?

> But does it really matter ?

Yes - I think we should challenge misleading information wherever we see it in this situation. Fighting panic is part of the problem and bad numbers cause panic.

I mean, if we don't really care that the numbers aren't accurate because it's more important to emphasise why the project is important, we might as well go all the way and say 99% of people need a ventilator and really sell the project.

There was a news report recently implying a 50/50 survival rate, due to this same kind of assuming everyone realises that you're talking about some group that's already in a bad way, but not actually saying that in the text.

I don't think this project wants to create panic. To me it rather is trying to help and be prepared. I'd rather have too many than too few.

Plus, we don't know when the site was set up. Two weeks ago ? Four weeks ago ? Our collective knowlegde is changing every day. Could just be they have been busy and did not find the time to update it ?

And finally, blueprint for a cheap OSS FEV will always be useful. COVID or not.

They’re not arguing against having an open source spec, they’re just pointing out that you can do the open source ventilator thing while not spreading misinformation.

Please don’t confuse the two. It’s hard enough to fight the misinformation as is without well-intentioned people such as yourself introducing red herrings.

Again, your point is coming from a good place. But we need to be really careful about not accepting misinformation.

This is one line on a whole page. And if usually people should cross-reference at the very least 3 reputable sources before accepting an informations as true. Maybe it is a good time to ingrain that message along the "wash your hands"
Yes, I’m sure lots of people in Italy, Paris and Madrid wishes they had spent more time/energy challenging the notion that ventilators are important as they are letting hundreds of people die every day because of the lack of ventilators. /s
I can't understand this point of view. It's more important than ever to put out good, accurate data and to help the public understand what is going on.

Also - what happens if something changes and now literally 20% of people do require ventilation? You won't be able to get that message across now because that's what they already think and the message won't be any change to readers!

Yes - I think we should challenge misleading information wherever we see it in this situation

Indeed, and your information is wrong and misleading, stop it. You listed no sources and are going on "I thought..." You thought wrong.

Actually, the biggest problem for a very long time has been people downplaying the risks and dangers of this pandemic, questioning the numbers, wondering whether it’s really that bad, and castigating any amount of preparation as “panic”, a word which has become meaningless in its overuse.
But we know we're creating more artificial problems for ourselves!

There was no actual food supply issue. But people have panicked due to unchecked bad information and now we do have a real food supply issue, at the very worst time to have one!

Maybe if someone had said to people 'hang on that's not quite right there's plenty of food being supplied' we'd have one less problem.

I don't get this idea that "panic buying" is necessarily bad. We are moving goods from communal locations to people's homes. We aren't destroying goods. Once/if the virus does arrive in a large volume at the location we would greatly rather that people stayed at home and ate food they had stockpiled than that they then went to the grocery store. Dealing with shipping extra products now (while a very small fraction of people are infected), or just having shelves in stores be slightly bare, seems like a worthwhile tradeoff.

There are some questionable cases, like people hording years worth of toilet paper (which can cause real temporary shortages and actually significantly inconvenience people), but everyone stockpiling a months worth of food seems like a good thing.

> I don't get this idea that "panic buying" is necessarily bad. We are moving goods from communal locations to people's homes.

Not to everyone's homes. For example: old people who can't rush to the shops and elbow their way through the queue may get nothing.

Having more food at home and therefore not needing to go out as often is good for the people that are physically and financially able to. Those that can’t - people on low or no income, homeless, elderly, disabled and sick are some of the groups that are most vulnerable to this virus. For them, other people’s panic buying has caused more than an inconvenience.
Panic buying also usually implies long lines of people waiting in close proximity to each other leading to additional vectors of potential infection. So yeah, no.
Well a few people moved some of the goods to their homes and today when my wife went to do the shopping she phoned me in tears saying the entire vegetable section was empty. Same with the canned goods section. Same with the bread.

So... We couldn't get any of those to our home. That's kind of a problem.

People just need to chill.

The Imperial study estimated a need for 100k ventilator intensive care beds at peak in the UK, currently there are 5k beds, and they’ve managed to find another 5k ventilators using spares and old models. The government is planning on using ventilator technicians (usually 1 per bed) to manage many beds, with newly trained junior staff managing each bed. So assuming no treatment breakthrough they will need something on the order of a ten fold increase in ventilators and a 20 fold increase in staff numbers, on current trajectories within a month or two, and with full suppression maybe in the Autumn.
A lot of ventilators and more importantly people to administer them. This project is not providing a ventilator, and it's not providing people the special training required to administer them. If we need ventilators, we can have ventilator companies produce them using emergency powers every developed country has.
Well, not everybody is The First World and have ventilators plants in their border. But some place will love to have access to free and unencumbered blueprints to be able to do something locally with what they have at hand ?

But you are also right that people are needed to administer those. No doubt about that. As is also true a person can administer several of those machines.

And, just because something tries to address A only (and not B and C), does not mean we should not do it because B and C. Separate issues. Beside, to train people, you need spares to train on.

No, he's wrong and you're wrong.

There's no evidence supporting the theory that large numbers of asymptomatic people offset the figure of 20% of patient being severe cases. Hospitalizations and death skyrocket in Covids infected areas, we know what this thing looks like at scale. Plus Who report, pattern of infection, China and Korea eliminated visible cases and haven't seen many more etc.

You are wrong according to the WHO investigation of the events in China.[1]

You are wrong according to the statistics that came out of Korea - if there was an invisible group of asymptomatic, Korea's infection rate couldn't have been controlled. [2]

This destructive belief has persisted for a while because it made sense for various flu epidemic and gave the comforting idea most infections would be harmless. But is now with us at scale and all the evidence points to a rough 20%, 1-in-5 hospitalization rate [3]. I wish actual authorities would spend more time debunking this (even get fully clear on it themselves).

[1] https://www.who.int/docs/default-source/coronaviruse/who-chi...

[2] Look at covid19info.live and look at the South Korean statistics. There's reason to think Korea found most if not all infection. Similar reasoning also applies to China.

[3] Edit: Discussion of CDC study: https://thehill.com/policy/healthcare/488325-cdc-data-show-c...

> There's reason to think Korea found most if not all infection.

This is beyond ridiculous and you have no basis for making that assertion. As of last Saturday, In South Korea, as of the weekend only 248,000 people out of a population of 50,000,000, with 8,086 +ve cases and 72 deaths.

There is significant evidence that not only are most cases mild, but often asymptomatic.

https://www.sanitainformazione.it/salute/scovare-i-positivi-...

In English:

https://mobile.twitter.com/andreamatranga/status/12397748625...

> According to Crisanti, the director of the virology lab of U Padua, as little as 10% of #COVID2019 carriers show any symptoms at all. He sampled repeatedly the entire 3k+ population of Vo ', one of the initial clusters.

https://grapevine.is/news/2020/03/15/first-results-of-genera...

> 700 have been tested. Kári says that about half of those who tested positive have shown no symptoms, and the other half show symptoms have having a regular cold.

https://www.repubblica.it/salute/medicina-e-ricerca/2020/03/...

> "The vast majority of people infected with Covid-19, between 50 and 75%, are completely asymptomatic but represent a formidable source of contagion". The Professor of Clinical Immunology of the University of Florence Sergio Romagnani writes

> But is now with us at scale and all the evidence points to a rough 20%, 1-in-5 hospitalization rate [3].

No. It doesn't. That link doesn't say why they were hospitalised. In America if your insurance is good enough you can be referred for little to no reason.

The one credible source among your links talks about surveying a population and seeing of those testing positive for Covid are asymptomatic or have cold symptoms.

But this finding is not extrapolated to mean that the vast majority won't require hospitalization. There's a reason. When the virus is growing exponentially, most people have just gotten the virus and haven't gone the 2-3 weeks typical for becoming so sick that you require hospitalization. Exponential growth means 3-week old cases are rare. A weekly doubling time 1/16 of the cases of the cases are three weeks old. If 1/5 of those cases require hospitalization eventually, you will wind-up with only 1/80 of those cases seeming to require hospitalization if you're just taking a survey.

Some of my references are extrapolating things (correctly) but others are citing recognized authorities. Your entire argument is basically incorrect extrapolation based on not taking into account exponential growth.

This article widely read article summarizes the quandary we're in and how to extrapolate the current data.

https://medium.com/@tomaspueyo/coronavirus-act-today-or-peop...

People need to read it and stop with the destructive misinformation.

https://medium.com/@tomaspueyo/coronavirus-act-today-or-peop...

> If 1/5 of those cases require hospitalization eventually

They won't, they don't, and you have no basis for making that claim. I don't know what your agenda is here but it is entirely clear you have no desire to honestly engage regarding the facts. Certainly a complete misreading, at best, of data presented.

It clearly deals with the symptoms during the while life cycle of the disease.

What precisely is your goal with this misinformation?

They won't, they don't, and you have no basis for making that claim.

All the links in my original post are the basis of my claim - the WHO finding in China is very plausible and says exactly what I say - so saying I have "no basis" is clearly misrepresenting my above post.

I believe I'm characterizing your claim and their links as well as I can while vehemently disagreeing. As far as I can tell, you cite a survey finding many asymptomatic cases and think that proves things will stay that way but fail to consider the properties of a growing infected group. I'm entirely hostile to your position but I know only substantial arguments can help here.

My main goal is to make clear the urgency of this situation. There's a debate about whether the virus needs to be actively suppressed and I want to make it clear that this is indeed necessary. Basically, not seeing the American Health Care system collapse and hundreds of thousands of people die is my motivate. For that, we have to realize how many people will be coming in (though that's visible in Italy).

You talk of "engaging with the facts" but you don't present either facts or arguments in this post - plus alleging motives, etc.

Edit: Looking further at your link, you're describing the (important testing approach in the village of Vo). You can say "as 10%" were symptomatic but this is in the context of the virus being spread by them, again, not in the context of the people not getting sick later. There's really no reference to exactly what percentage of people go seriously sick.

The Imperial College study, which seems well-received, and which caused the UK government to change strategy, estimates a hospitalisation rate of 4.4%.

This is on page 5 of the paper.

Edit: link https://www.imperial.ac.uk/media/imperial-college/medicine/s...

> My main goal is to make clear the urgency of this situation.

It’s not ok to deliberately over-estimate numbers in order to achieve this goal. That was my original point at the top of this thread.

You may think it’s ok because the ends justify the means, but it’s still wrong and dangerous.

> You talk of "engaging with the facts" but you don't present either facts or arguments in this post - plus alleging motives, etc.

You are clearly or for an argument. 1/5th cases are not as you say. That's a lie. Plain and simple

The WHO conclusions have been widely challenged.

The cruise liner and the 3000 pop Italian village are the well studied exposed populations so far I think and they indicate a big asymptomatic percentage.

The cruise ship showed a ~50% symptomatic rate, so any stats based on symptomatic patients are probably only off by a factor of 2.

Note that both South Korea and China outside Wuhan do extensive contact tracing and testing of people an infected individual can be determined to have interacted with, so they pick up a good deal of asymptomatic cases too.

The cruise ship demographic is extremely relevant.
Can you show me a link to what percentage of those exposed in the ocean linear needed hospitalization? I haven't seen any direct discussion of this and that the situation in question. Sure many can be asymptomatic but that doesn't imply the symptomatic group doesn't tend to get very ill.

Edit: I should have said "a large enough group of asymptomatic to push the fatality and sickness rate way".

Yes, there can a majority asymptomatic but that doesn't mean that 20% of the overall don't wind-up needing serious medical attention also.

Hopefully, you can read the comment I replied to and see the context

Sorry, on mobile, hopefully someone else can dig it up.
I've read several articles and none talk about the hospitalization rate. Two people died, which akin to the fatality with medical care seen elsewhere. That would seem to imply a similar rate of getting sick since the disease pattern is that with reasonable care, only small-ish portion of those getting sick die.

I mean, understand. Lots of people asymptomatic, a few quite ill, 1% die, sounds not terrible but it's very, very bad for it's health care overwhelm effect.

https://www.statnews.com/2020/02/25/new-data-from-china-butt...

> But on Tuesday, a World Health Organization expert suggested that does not appear to be the case. Bruce Aylward, who led an international mission to China to learn about the virus and China’s response, said the specialists did not see evidence that a large number of mild cases of the novel disease called Covid-19 are evading detection.

> “So I know everybody’s been out there saying, ‘Whoa, this thing is spreading everywhere and we just can’t see it, tip of the iceberg.’ But the data that we do have don’t support that,” Aylward said during a briefing for journalists at WHO’s Geneva headquarters.

Your own link says that this is a highly controversial point of view, not agreed with by other experts.
The point is it’s apparently not a settled matter.
Yes, that number is negligently incorrect.

It's like saying 90% of basketball players require casts because, from the set who end up in ambulance, 90% of them have a broken arm. That doesn't mean 90% of basketball players require casts, and it certainly doesn't mean they need them all at once.

There was a study posted here recently that said as many at 86% of people were asymptomatic, then only some sliver of those with symptoms end up needing to go to hospital in the first place -- and 20% of that group that tests positive for the virus ends up needing a ventilator and 5% of them end up dead.

Net-net close to a 0% fatality rate under 29, 0.1% under 49.

Still gonna feel like crap tho.

Nobody knows the asymptomatic rate for certain, one way or the other. We don't know the rate of false positives with existing tests.
Fine, but we absolutely do know for an absolute fact that -- it is not true 15-20% of people with nCoV-19 need a ventilator. We don't know what the number is but we 100% with absolute certainty know it is not 15-20%.

If we acquiesce to 70% of the US getting nCoV-19 as the epidemiologists are suggesting that would require 50 million ventilators. There are about 70,000 in the US. So we'd need almost 1000X as many ventilators as we have.

If that were true we've have the national guard locking people inside their houses, and the UK wouldn't be contemplating giving nCov-19 to everyone young to foster herd immunity.

Well in China they seem to have forced everyone to stay home?

The UK appears to have decided allowing 500,000 of us to die was a bad idea and we're now on "lockdown". At least everyone is _advised_ to socially distance, because - it seems - then businesses can still fire people for not turning up to work, and insurers can avoid paying out ("you chose to stop the event, you weren't obliged to").

I'm not sure we can tell what the rates are, what's the testing false positive rate? UK gave up testing a while back (except emergency hospital admissions).

For the last week, at least, all new cases here are in theory emergency hospital admissions. 700 cases per day (and rising), 10% of our normal number of intensive care beds.

> Well in China they seem to have forced everyone to stay home?

Well, it's China.

> The UK appears to have decided allowing 500,000 of us to die was a bad idea and we're now on "lockdown".

500,000 people dying wasn't going to happen. Korea's death rate is closer to 0.4%, almost entirely the older folks who were to be quarantined at home during this process anyways. Korea's death rate for under-40's is 0-0.1%, so at worst, ignoring that vulnerable folks in those demographics would also be quarantined, the death toll would less than 50K -- probably much, much less, and not drastically out of line with a bad flu year.

> I'm not sure we can tell what the rates are, what's the testing false positive rate? UK gave up testing a while back (except emergency hospital admissions).

Supportive treatment is the only thing you can do anyways. Beyond that PCR tests will only tell you if you currently actively have the disease not if you had it before and recovered. We need antibody tests for that.

Where are you getting this 0.4% number from? Korea is testing extensively, so CFR is very close to IFR, and CFR is above 1%.
I can probably derive everything you believe from your flagrant mischaracterisation of the UKs policies.

It might help if you don't get all your news from Twitter

Got to love hackernews, in a discussion of tools to fight global pandemic, the top comment nitpicks an number that everyone knows is imprecise, and offers nothing constructive.
> everyone knows

I found his comment helpful because I wouldn't necessarily have thought of that.

Yours, however, adds nothing useful to the conversation.

It's not nitpicking to point out the statistic is wrong by an order of magnitude
I believe you are correct.