Hacker News Reader

Comment by Barrin92 | Hacker News Reader

>COVID-19 makes ~20% seriously ill and 10% dead

COVID-19 death rates in China are about 2-3% and significantly lower outside of China. (less than one percent). Probably due to the healthcare systems in the Hubei region being pushed to its limits.

Bad flu seasons in the US generally have a fatality rate of about 0.2% and infect about 30 million people annually, killing about 60000.

So even if COVID-19 cannot be contained and will linger around it will likely not be as deadly as influenza.

View on HN

>Probably due to the healthcare systems in the Hubei region being pushed to its limits.

What happens when the Italian healthcare system gets pushed to its limits? With a conservative estimate of 15% hospitalization (which I can't vouch for, but offer for the sake of discussion), there may be no medical system in the world which is prepared for this.

I notice a lot of downvoting without responses here. It smacks of denial and motivated reasoning. Usually on HN, when you get down-voted, you get a counter-argument. I think it speaks volumes when there's none.

If the original commenter here is suggesting that 15% of Italy's population will be hospitalized...that is its own counter-argument. That's pretty clearly not going to happen.

> That's pretty clearly not going to happen.

Sure, if you mean that the capacity isn't there.

And sure, if you mean it won't be 15% of the total, but rather 15% of the 70% of the population who catch it, then ok.

If you mean that that level of care isn't required for the 15% of severe cases which was seen in the studies, then you'll have to refute the studies of people who know more about this than you, with some better arguments than that.

http://weekly.chinacdc.cn/en/article/id/e53946e2-c6c4-41e9-9...

>of the 70% of the population who catch it,

Sorry, but where in gods name is this number coming from? There are currently a few hundred cases documented in Italy,rate of new coronavirus cases worldwide has slowed down to about 500-1000 per day.

For reference, the US has about 1 million hospital visits for pneumonia every year. I couldn't find numbers for Italy but I assume it's in the ballpark of 200k relatively speaking. So a few hundred extra cases or even a few thousand aren't going to push the system anywhere.

Not sure about the 70% but 60%-80% was talked about a few days ago: https://www.theguardian.com/world/2020/feb/11/coronavirus-ex...

> Usually on HN, when you get down-voted, you get a counter-argument.

Not really.

When you say "15% hospitalization" are you suggesting that 15% of the population will be hospitalized?

There are some estimates that 15% of conavirus cases end up needing to go to the hospital. So, 15% of however many Italians catch it might end up needing medical services, plus the standard baseline that already need it. That's going to be pretty tough for the system to handle.

Doesn't that depend entirely on how many Italians end up catching it?

It does, but in an average flu season, how many escape unscathed? Now factor in that the transmission rates may be significantly higher for this virus (R0 below 2 versus anywhere from 1.4-6.6).

10% is dead / recovered. Don’t look at dead / infected, that’s skewed by an exponential infection rate.

Dead/recovered is also skewed because if someone is going to die, they will die before the illness has "run its course," whereas someone who is going to recover will have it for longer before they count for their statistic. As a result it is also skewed by the exponential growth, albeit in the opposite direction. I really feel like none of the statistics I am seeing are satisfying...

That's not clear. Many people are getting antivirals and clearing the disease in a few days.

Presumably there are also people who die abnormally early. Unfortunately I have not seen what we really need to see: a breakdown of dead/recovered by day since infection. Without it there is no way to accurately assess the deathrate in an environment where new people are quickly falling ill.

There is no evidence to date that anti-retrovirals are successful at treating this infection.

I'm sorry, but that's just plain wrong. The first US case at Providence Everett got well in a few days thanks to antivirals.

Citation?

https://www.nejm.org/doi/10.1056/NEJMoa2001191

Many people in China have been treated. There is an ongoing trial that is supposed to have results in a couple of weeks. Patients in the U.S., Korea, and Thailand have all been treated with antivirals and dramatically improved within a day or two.

This was released Thursday: https://www.thelancet.com/pb-assets/Lancet/pdfs/S22132600203...

This Chinese study used 28-day mortality rate.

https://www.thelancet.com/journals/lanres/article/PIIS2213-2...

The best measure will be something of this sort, with a sufficient duration to catch most of the fatalities.

It assumes we know the amount of infected people.

We have very accurate numbers for surprisingly large group due to the cruse ship. As of 20 February 634 infected including 328 asymptomatic cases. https://en.wikipedia.org/wiki/2020_coronavirus_outbreak_in_c...

Adjusting for the delay between infection and first symptoms a relatively small percentage of people avoided symptoms.

That's a sample population biased toward the elderly, though.

80% were age 60 or over. Normal is 11%.

People stuck on the same ship 24/7 is not very representative of actual daily life.

How are you doing this adjustment?

Actually what the data says is the older you get the more deadly it is and below certain ages it's less than 2.

There are actually 5 episodes on "this week in virology". These people are virology professors in the US.

This episode an interesting look at the percentages.

http://www.microbe.tv/twiv/twiv-588/

EDIT: two other interesting aspects, over 80% of the cases were mild and over 50% of chinese males smoke.

I'm not understanding your conclusion at all. In an environment not taking precautions above normal flu season, the r0 for covid-19 is significantly higher than flu (2.6 vs 1.6) - in part because we have partially working flu vaccines. If it can't be contained, it'll be far worse.

Now I suppose measures China style will drop its transmission rate below flu, but at the minimum that's a heavy economic cost to be concerned with.

What you're missing is the multiplier. It seems to be substantially more contagious.

edit: To put a finer point on it, the estimates of the R nought range from the same as the Spanish Flu, to double the Spanish Flu.

I’ve seen C19 rates of 20% seriously ill (hospitalization) and 5% critical care (ICU). Almost no country has the capability for this, even in large cities.

There are vaccines for flu; none for C19 for 18 months at the earliest. Also, C19 is contagious when victims are asymptomatic. It aerosolizes so it can spread across rooms and potentially through ventilation systems.

Many C19 healthcare workers are becoming infected, some dying. This is unheard of with the flu.

And C19 is uncontained, unlike SARS. It meets the definition of pandemic now (2+ continents and community spread). The WHO will finally recognize this any day now.

To equate C19 with the flu is inaccurate and dangerous.

Watch this video from CNBC last Friday.[1] It’s excellent, and features experts who do not mince words.

[1] https://youtu.be/8A9BT3lAGHU

Curious why you say 18 months at the earliest for a vaccine. Is this a standard vaccine timeline? Are we guaranteed to be able to make a vaccine for this virus? I am not at all educated on how vaccines are developed.

Reply on news.ycombinator.com