Hacker News Reader

Comment by S_A_P | Hacker News Reader

This virus seems to be right at the cusp of something we should really fear. ~3% mortality rate and at least semi easily transmissible. The fact is that this could be a lot worse and we are already seeing some ugly behavior from it. I imagine chaos if mortality climbs by an order of magnitude. We need to keep our heads here and figure out how to stop this thing.

View on HN

> At present, it is tempting to estimate the case fatality rate by dividing the number of known deaths by the number of confirmed cases. The resulting number, however, does not represent the true case fatality rate and might be off by orders of magnitude...

> A precise estimate of the case fatality rate is therefore impossible at present.

Taken from https://smw.ch/article/doi/smw.2020.20203

mortality depends on the country in which the patient is treated in my view.

according to WHO's situation report, Singapore has 91 confirmed cases and 0 death so far. in most developed country, mortality rate seems to be on much lower side. Of course these numbers can change at any minute.

https://www.who.int/docs/default-source/coronaviruse/situati...

It's more patients vs resources than developed vs undeveloped, it's much easier to handle 91 cases at once than 900 or 9000. Italy is also a developed country and the mortality is around ~3% so far in Italy, Japan is developed and it's at ~1.5%. At the moment it seems like Korea is the biggest outlier with mortality under 1% but also with a lot of cases.

I'd rather be in a developed nation, but you're cherry picking your example.

Mortality rate is much, much lower than 3% if you exclude the elderly and the critically ill. Based on the Chinese CCDC's report released last week, covering 72,314 confirmed infections, mortality rate for a young healthy person is an order of magnitude lower.

80+ years old: 14.8%

70-79 years old: 8.0%

60-69 years old: 3.6%

50-59 years old: 1.3%

40-49 years old: 0.4%

30-39 years old: 0.2%

20-29 years old: 0.2%

10-19 years old: 0.2%

0-9 years old: no fatalities

I think most people are blowing it out of proportion. It's just a bad flu, and it seems to be responding well to remdesivir treatments in early trials. I also think that, with the news we're getting now about asymptomatic/minor cases, it's highly likely the real number of infections is much larger, which would mean the CFR numbers are way too high.

I think your post is intended to be reassuring to the presumable majority younger demographic on HN.

But it strikes me as very insensitive. Ignoring the validity of the source of those numbers, even if they hold up. We all know people over 40. We all know people with existing illness that put them in higher risk brackets. And we all know people in other regions of the world with poorer healthcare systems where the fatality rate may go up.

The parent’s fatality rate was broadly correct. You can’t just select out a big portion of the population to give a better survival rate. Then compare your hacked fatality rate to the unhacked seasonal flu rate. “Just a bad flu” that is up to 20x worse than the regular flu is a big big deal. The near wartime like response in China is not “just a bad flu.” This is affecting a lot of people in very serious and real world ways.

To me, his last paragraph was the most important part. We've seen that the case rate increases dramatically as more tests are administered.

This is implying a very strong selection bias in the past testing towards the most severe cases. The mortality rates are certainly much lower than these estimates.

>We've seen that the case rate increases dramatically as more tests are administered

Is that really true though?

The UK has administered some of the most tests at over 7,119 but only 13 confirmed cases. US has done a measly 445 tests, yet 60 cases? Might be things are just starting to ramp up, and yet we're behind on testing...

>US has done a measly 445 tests, yet 60 cases?

I think the US tests were much more targeted to those most likely to have contracted it. Like US citizens returning from Wuhan/Hubei.

Yeah that makes sense, although that would be a selection bias towards most likely cases, not necessarily most severe. Although China's data is most likely to be biased towards most severe...

Its very possible that I was incorrect. I now realize that my statement was based on something someone told me, and I can't find original sources.

I do continue to believe that the real cases are much higher than the reported cases, though, and many of these unreported cases are the ones with the mildest symptoms. This will continue to skew statistics for some period of time.

There's also reports that if somebody is dead then they won't treat them due to limited resources, they'll just ship them to the crematorium. Because of this I'd expect the 3% rate to hold at the very least, if not slightly increase.

> it strikes me as very insensitive

It's a physical fact that older people's bodies aren't as able to fight off diseases, and they are likelier to have mortality rates.

If no one's allowed to actually explain the physical facts of how the physical universe works, because it's "insensitive" to mention that older people are more likely to die when they're sick, how is our society supposed to come to correct [1] answers when it comes to medical science and health policy?

[1] By "correct," I mean minimizing the incidence and severity of illness within relevant resource constraints.

Agreed. Statements in contexts like this should be evaluated as either correct or incorrect. Worrying about sensitivities in science is typically counterproductive.

I think most people are blowing it out of proportion. It's just a bad flu

This isn’t science, it’s someone’s opinion on how concerned other people should be based on their interpretation of some very incomplete data.

> I think most people are blowing it out of proportion. It's just a bad flu

It isn’t just conveying information, its downplaying the severity of the situation.

It's not that those people groups don't matter, it's just that if it wasn't COVID-19, it could have been one of a dozen of other common ailments with elevated CFRs for those populations. My Nana beat cancer multiple times, but ultimately succumbed to a minor cough that turned into pneumonia.

Eventually time gets the better of us all. I don't think that's an insensitive statement, just pragmatic/realistic.

So what in your opinion would present a public health danger worth worrying about? A 6 percent mortality rate? 10 percent? Or is there no mortality rate worth fretting about if it only affects people 50-years-and-older?

> But it strikes me as very insensitive.

And here I am pondering if the generational time bomb's been defused. Parent's downright warm'n'fuzzy.

29 to 40 year old Chinese doctors have already died. If it was really only 0.2/0.4% for them, those events would have been extremely unlikely.

Also, in Iran it is concerning how much higher the CFR is there; which are leading some to think the virus may be exhibiting different behavior there.

and yes, the severe/critical case rate is somewhere around 14%-18%, which means you need hospitalization.

if you don't trust the chinese data, the data in south korea and italy is very trustworthy, and looks just as bad on those too.

Do you mean the Chinese doctors who were in prison for revealing the coronavirus to the public?

I imagine their health was terrible just from a few weeks living in a Chinese prison.

Nearly a quarter of the United States population is 60 and older.

https://censusreporter.org/profiles/01000US-united-states/

Are those numbers supposed to be comforting?

> 80+ years old: 14.8%, 70-79 years old: 8.0%

That makes it extremely likely that at least one friend or family member while die of this.

> 30-39 years old: 0.2%

To put things in perspective (https://www.statista.com/statistics/241572/death-rate-by-age...), that doubles this risk of death this year for that age bracket.

> 10-19 years old: 0.2%

That roughly quadruples the rate of death for that bracket.

And besides that, it's not the mortality that's the biggest worry but the amount that need critical care (10-20%). In a large outbreak there will simply be more critical cases than there are ventilators to put them on.

Edit - Does anyone know what sort of triage China put in place? Ie, when they don't have enough capacity are they favoring treating young people and giving up on the 80+?

These numbers are bullshit and I'm tired of having to repeat this everywhere. The Chinese government is lying about infections and there are legal and infrastructural issues which prevent the vast majority of cases from being diagnosed and reported, even if there wasn't an ongoing official attempt at minimization.

The virus has a ≈15% critical illness rate[1]. There are less than 100k ICU beds in the entire United States. They will very quickly be overwhelmed, exactly as they are in China, SK, Iran, Italy...the list is getting too long to keep up with. At that point the mortality rate approaches 15%. This is not the god damn flu and people who dismiss it as such are being hugely irresponsible.

1. I believe this article is referencing the published paper in Lancet that I was referring to https://coercioncode.com/2020/01/27/breaking-news-coronaviru...

Also bear in mind that the study is out of China, so it is most likely extremely conservative because the government has been publicly cracking down on anyone "spreading rumors" about the virus.

The peoples' collective, deliberate obviousness at this point is unbelievable.

Edit: wait a minute, I don't think I've even seen the paper I linked yet. This one reports 32% ICU admission rate and 15% death with lasting damage in recovered patients. Some 2 dozen Chinese authors to boot. Get your fucking heads out of the sand!

https://www.thelancet.com/journals/lancet/article/PIIS0140-6...

"The virus has a ≈15% critical illness rate[1]. There are less than 100k ICU beds in the entire United States. They will very quickly be overwhelmed, exactly as they are in China, SK, Iran, Italy...the list is getting too long to keep up with. At that point the mortality rate approaches 15%"

It's not just an ICU severely ill people need, but artificial respirators. I haven't seen numbers on how many the US has, but it may not be enough.

In addition, this illness has the potential to take a significant amount of medical personnel out of commission, either directly by making them sick or killing them, through attrition as they overwork themselves and collapse (we've been seen all of the above happen in China).

So even if you get in to an ICU you could still be screwed.

That said, the numbers coming out of China are just not trustworthy right now. We'll have to wait and see how bad it is outside of China (assuming enough people get tested and the tests actually work, neither of which are a certainty yet).

Thank you for this post and pointing out the fact that we will be out of ICU beds very quickly.

The complacency of ignorance on this really beggars belief.

> The virus has a ≈15% critical illness rate[1].

This is completely implausible.

Why do you say that? That’s what all the reporting says. That’s why the health care systems were overwhelmed.

We are seeing a lot of cases that are only being discovered after the patient has had symptoms for some period of time. It seems like a number of these cases have been mild enough that a lot of people would not have even sought care.

The idea that 1/6 cases are "critical" just doesn't align with this reality.

A 15% critical rate is very plausible. A widely shared report [1] on Chinese media described how an entire family died to virus. A man, his sister, mother, and father all died to the coronavirus. His wife is in serious condition.

Using the reported death rates posted above, the odds of this happening are 1.3% * 1.3% * 8% * 8% = 1/924556. That's nearly 1 in a million. I concede that the medical system in Wuhan was overwhelmed, and these 5 patients did not have access to medical care.

[1] https://www.caixinglobal.com/2020-02-17/four-deaths-in-one-f...

No it is not, Diamond Princess is the only controlled environment we have and it didn't have even close to 15% critical rate.

Exactly... also see this report: https://twitter.com/balajis/status/1232797151941611520

The situation is horrible, but look at the hospital workers rate of infection vs mortality.

Even horrible things can still be made to look more horrible.

Not at all. The number of serious/critical outside of China are about 3% and mostly elderly people.

The most significant experience is that efforts should be made before medical services were overwhelmed by flood of patients. Mortality would be higher if patients were not properly looked after.

Another bad flu is a terrifying prospect. If this coronavirus becomes cyclical like the flu, it will be a gigantic driver of global mortality.

Flu is also heavily dependent on age —

The mortality rate for seasonal flu for 18-49 year olds is .02% — so by your own numbers, this is 10 times worse.

You’re citing numbers from the Chinese govt. You can’t be serious.

Say this holds out, but say it's fast evolving. A woman in Japan for instance recovered from the virus 2 weeks ago, and is in the hospital w/ it for the 2nd time.

If you don't die the first time, what's to say you don't get this 3-4 times in the next year and the 4th time gets you?

remdesivir treatments

The synthesis route from Wikipedia does not look encouraging. Eight steps, some conducted at low temperature - good luck with scale-up! Production scale at dry ice temperature requires a dedicated facility, your garden-variety contract shop doesn't have the equipment.

> 80+ years old: 14.8%

Great. My grandparents have a more than insignificant chance of dying if they (probably) catch it. This really assuages my fears.

> Mortality rate is much, much lower than 3% if you exclude the elderly and the critically ill.

Why would we do this?

Not trying to be mean, but stopped reading after "based on the Chinese CCDC's report".

Can you explain those percentages? Why do they add up to ~30% and not 100%?

Death rate by age bucket

> Mortality rate is much, much lower than 3% if you exclude the elderly and the critically ill

Yeah, fuck them, right?

Do you get as upset when those same figures are given for the flu?

I do when people downplay the flu and say it's just a bad cold.

What same figures? The CDC says the high end of flu deaths is 40,000 [0]. You think that number compares with a 3 percent mortality rate for the 70 million Americans who are 60+ years old?

[0] https://www.cdc.gov/flu/about/burden/preliminary-in-season-e...

Why would you exclude people over 50, transplant patients, people with lung ailments or cardiovascular disease, people being treated for auto-immune disease, etc?

This reaction that it's OK as long as it doesn't kill young healthy people is sociopathic.

By the way, your "blown out of proportion" comment means you are fine, just fine, with an expected death toll of 4 million Americans, if it were to go epidemic and infect 40% of the US.

As a programmer, there are two options:

1. quarantine. 2. vaccine.

Given the fact that people will not self-quarantine mainly due to economic reason, the hope is to rush the vaccine as fast as possible through the FDA, and then to mass-produce it.

As laid out by the actual expert at Trump's coronavirus press conference tonight (not Trump), American regulations on safety and efficacy mean that "rushing" a vaccine will take about one year.

Quarantine isn't the only way to slow down the spread of a disease, and it's not unheard of for large changes in behavior to happen in response to large risks. Hand washing or even mask wearing could become more popular. E.g. https://zoomingjapan.com/life-in-japan/why-japanese-wear-sur...

From what I’ve read “rushing” for vaccines means it will be ready in 2 years

Since this is the first time that the world is facing such a pandemic, this must and I assume will be changed.

It will, thus bringing the ETA to 1 year or so.

I'm not sure that it is possible to do so.

Vaccines take a long time to develop; just throwing more money at it isn't a solution. You need to understand the pathogen, which just takes time. Getting the IGg out of horses or whatever takes time. Making sure you can use horse IGg versus rabbit IGg takes time. Trying it out on volunteers takes time. Mass producing it takes time. The officials aren't joking when they say 2 years.

SARS still really doesn't have a vaccine, and that effort started in 2003. Heck, Malaria took ~50 years for a kinda-okay-ish vaccine.

Fact is, there is a very low chance of a vaccine for this one.

Wash your hands, don't touch your face, stay calm.

First there is a small safety trial of several months, followed by an efficacy trial of I think six months. We could possibly skip the efficacy trial at the expense of potentially loading up people with multiple not-very-effective vaccines. And getting through safety trials doesn't mean there are no side effects, so we'd have to deal with those compounded as well.

Nah, the mortality rate is higher, around 9%, similar to SARS. Dividing total deaths by total cases is wrong. You need to divide by the sum of deaths plus total recovered cases. With that in mind it's already way higher.

That doesn't work either because the definition of 'recovered' is something like 'symptom free for 40 days' and most people dying are doing so within 1-2 weeks. What you really need to do is divide the total deaths by the total known cases two weeks ago (roughly when the most recent deaths would have been infected).

Last time I did this it looked like the overall death rate would come out at ~4.5%, reporting methods have changed since then and now it's coming out as more like 4%.

Even worse, we don't know how many people have actually been infected. The reported death rates aren't reliable either.

All the estimates we're seeing are almost entirely from unreliable data coming out of China.

So I'd take these estimates with a huge grain of salt for now. We should be in a much better position to have reliable estimates in a month or two.

True. On the plus side, the higher the actual number of infections, the lower the death rates that we should be expecting. There's a growing consensus that the virus can spread asymptomatically which (while not good for containment) means we're only seeing a small fraction of cases and the death rate is going to be much more like that of the flu (which isn't great but also hasn't destroyed civilization as we know it).

While there may be asymptomatic spread (it's too early to be certain of that yet either), "asymptomatic transmission likely plays a minor role in the epidemic overall, WHO says. People who cough or sneeze are more likely to spread the virus, the agency wrote"[1]

[1] - https://www.sciencemag.org/news/2020/02/paper-non-symptomati...

Counterpoint:

> US traders were particularly rattled by the Centres for Disease Control and Prevention confirming a COVID-19 infection in California in a person who apparently had no relevant travel history or exposure to another known patient. [1]

One asymptomatic carrier may not infect many people but if they infect even one new person in an as-yet-uninfected population center they'll create a new outbreak. The only defense against this is universal testing, otherwise by the time you start seeing symptoms it's too late.

[1] https://www.abc.net.au/news/2020-02-28/asx-carried-away-in-w...

Yes this is what people miss. There is a delay between known cases and deaths.

Reply on news.ycombinator.com