Hacker News Reader

Comment by slowmovintarget | Hacker News Reader

According to the estimates in the article -- predicted infections: 60% of 7.8 billion, current known fatality rate: 2.3% -- this virus is on track to kill roughly 108 million people. That's more than the Spanish flu of 1918 (high estimates put that at ~50 million).

View on HN

That's inaccurate. The known case fatality rate is 2.3%, but that's of cases which presented to a hospital and were diagnosed.

It's likely that 90% of the infections are asymptomatic, or close enough to be mistaken for a bad cold. That's bad, because such people are hard to quarantine, but it's good, because it means the true fatality rate may just be 0.2%.

Not that 0.2% of five billion wouldn't still be a horrible tragedy.

I agree.. we really really don't know much of what's going to happen here. We definitely need to stay on our toes and not get complacent here.

Also, world population in 1918 was about a quarter of what it is now. 50 million dead hit harder then than 100 million would hit now.

It's not always clear when to use absolute and when to use relative units.

I would suggest that each dead person is mourned by approximately the same number of people, and therefore absolute units are more appropriate.

Also, the world of 1918 was much less interconnected. It's not clear if that makes losing n% of the population better or worse, but I would bet on worse.

Unfortunately, those errors exist on both the numerator and the denominator.

Interviews of crematoriums in China claim that two thirds of the bodies they are picking up are from homes, not hospitals.

Many cases--perhaps 80-95%--in folks under 50 are hard to distinguish from cold, flu, or gastroenteritis. 2.3% rate is for those who develop more serious symptoms. See http://weekly.chinacdc.cn/en/article/id/e53946e2-c6c4-41e9-9... or https://news.ycombinator.com/item?id=22372003 for a summary table of serious cases.

2.3% is still based on official Chinese data, which is likely totally bogus.

Real rates are probably more nuanced with many under 50 at very low risk and those above 60 at higher risk. A serious case can present as a viral pneumonia, if there is a break down in health care system and those who need oxygen on a temporary basis cannot get it or if they develop a secondary bacterial infection and antibiotics are not available then mortality will go higher among those cases.

I think this year will be remembered by many as "the year my grandfather died of pneumonia."

According to the latest WHO sitrep [1], there are 3 deaths and 924 confirmed cases. That's a fatality rate of 0.325%.

China is an outlier because their health system is overworked.

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

LOTS of countries don't have a health system as advanced as China, this is where the risk is. You can't dismiss the risk based on the fact that China is overworked because it is likely that other countries could end up in that situation with sufficient amount of cases.

Wouldn't all health care systems be overworked with this high of an infection rate?

If this gets to where most of the people in the world are, ie: India/Bangladesh/Pakistan/Africa, do you think their health care system will be better than China currently or worse?

Dividing current deaths by current cases is just about the worst way of estimating CFR of a developing epidemic. Try current deaths vs cases from 5, 7, 10 days ago or deaths vs (deaths + recoveries).

The death rate doesn't mean anything unless we are looking at the "infected and recovered" number... so the death rate should only go up from there.

8 out of 1073 = 1.34125% as of 20.02.2019 [1], looks like it rises 1% per day;p but we both know that it's false assumption. And numbers we have right now doesn't matter really because it is too early to tell. Half of the world doesn't even have a capacity to test for COVID-19. And impact on developing countries will be on completely different level [2] (or maybe not, because population is much younger there)

[1] https://www.who.int/docs/default-source/coronaviruse/situati... [2] https://www.who.int/news-room/fact-sheets/detail/the-top-10-...

The article says

> COVID-19 could infect 60 percent of the globe if left unchecked

Clearly the outbreak is not unchecked. Many measures have been put in place already. I am pretty sure the global community isn't going to just standby and allow the virus to communicate freely.

I certainly hope not, though that is what Japan and Singapore seem to have settled on.

Singapore has mounted an extremely effective response.

They've been vigilant in their testing and transparent in their results, and it's making their numbers look bad. It doesn't look like any other country will follow their lead.

At least they are doing something (or probably the right thing to say is that they are the only country with competent measures) and because of transparency you can see what's going on almost in real time. I think their numbers look bad now, but the measures look solid and to be honest Singapore is probably the only country with adequate response. So if things go bad there, i'm pretty sure it will go bad everywhere.

[1] https://www.channelnewsasia.com/image/12436712/0x0/3000/5883...

[2] https://www.channelnewsasia.com/image/12455138/0x0/2316/3209...

[3] https://coconuts.co/singapore/news/covid-19-heres-every-nove...

[4] https://www.google.com/maps/d/u/0/viewer?mid=14y9TsLoO2Y6bLj...

A lot of practical advice and experience in this Feb 20 article on lessons learned (so far) from Singapore's experience: https://jamanetwork.com/journals/jama/fullarticle/2761890

> At least they are doing something (or probably the right thing to say is that they are the only country with competent measures) and because of transparency you can see what's going on almost in real time.

Oh absolutely. In these crucial moments seeing every other country failing in its own way, Singapore's response impresses me. I think this crisis is acting as a child/adult/senile test. Some countries are not developed enough to handle this; they are denying anything is wrong and trying to delay the economic impact as long as possible. Some countries have been stable long enough to ossify; each is failing due to its own unique type of entrenched-power disease. Singapore is grown up enough to have the right resources to fight, and young enough to have control of its faculties.

> So if things go bad there, i'm pretty sure it will go bad everywhere.

I'm not sure about this. Singapore is one of the most polluted cities in the world, due to Indonesia being chronically on fire. There's some research that suggests that air pollution upregulates ACE2 receptors, which would be expected to increase transmissibility for a virus known to use those receptors to enter cells. We don't know much yet about its spread in populations with low smoking rates and cleaner air.

CDC: The containment phase is really to give us more time. This virus will become a community virus at some point in time, this year or next year. [...] Right now we're in an aggressive containment mode, we don't know a lot about this virus, this virus is probably with us beyond this season, beyond this year, and I think eventually the virus will find a foothold and we will get community-based transmission.

Add asymptomatic transmission in an incubation period of ~14 days and there is not much to check. Also, since China waited months, the virus was already left unchecked, resulting in an estimation ~1 million infected: https://www.medrxiv.org/content/10.1101/2020.02.12.20022434v...

It's interesting that in your other comment you referenced another paper for a higher IFR although in this reference the model suggested 0.07%. So what do you believe? A higher total infection with a much lower IFR, or a higher IFR? Or you just pick papers that support your viewpoint for the sake of argument?

> Or you just pick papers that support your viewpoint for the sake of argument?

This. (Actually I believe both numbers are optimistically correct: 1 million infected and 50k death, but I also realize that's crazy talk and panic time. If that is what is really happening in Hubei I would quarantine 10% of the world's population).

I don't think CCP can cover 50k death with ease. Unlike many western people assumed, CCP is not omnipotent in China. There are checks and balances of power. Not distributed democratically, but still distributed in some opaque ways. 50k death can lead to some pretty serious power dynamics change if true.

Btw it's not too far off that right now 10% of world's population is more or less under some level of quarantine.

Yes, however because it spreads so easily (even in the period without symptoms!) it is likely to spread outside of China. Singapore has done well to contain it, but not all countries have great health and control systems and it will eventually be just spreading everywhere, it's almost impossible for it to be entirely contained based on the pure fact of how large China is and how hard it is to detect.

Yeah just like it is in Iran, where they diagnosed a case posthumously in somebody who hasn’t travelled abroad, which means it’s been in the country for at least three weeks and they just learned about it last night.

It's interesting the cases in Iran are so high. Lots of movement between China and Iran I assume.

They are really expecting it to reach over half the population?

I believe there was a quote from the Twitter account of an epidemiologist in Hong Kong to the effect that if the disease was left unchecked it would reach that many people.

Under what scenario would this disease be left unchecked? Seems like a false hypothetical.

It’s a baseline to urge people to act.

Reply on news.ycombinator.com