back
6 comments
"Even if death doesn’t occur, some patients survive with permanent lung damage." It's not just about dying.
Beside the discussable mortality rate among healthy adults, and beside surving with permanent lung damage, many people who are 30-40 have kids who are 0-10. Things take a whole new dimension there.
Coronavirus isn't like the flu in this regard, and is actually less severe the younger you are, including children.

No one under 10 has died from it, and only one person under 20. The mortality rate for people under 50 is about .2% (compared to 2.3% for all cases).

So far it's basically a logarithmic function of age.

The lung damage piece is obviously frightening, and we obviously still have a lot to learn about it as we're not sure why children are able to fend it off so easily yet.

The point is that potentially quite a few kids are about to lose one or both parents this year, much more than baseline.
The mortality rate of kids 0-10 has been extremely low.

As for permanent lung damage, it's hard to say. From what I've read about ARDS, as expected, if you're young, lung function and general physical function can improve back close to baseline over time (e.g. 5 years). So it's not like you're going to be confined to a rocking chair but you won't be winning any marathons either.

Prognosis for permanent lung damage is you have five years to live.

EDIT: sourced: https://en.wikipedia.org/wiki/Pulmonary_fibrosis which we saw in SARS coronavirus-induced pulmonary fibrosis.

> There is no known cure. [...] Life expectancy is generally less than five years.

Now stratify by age and cause. In the case of SARS.
ARDS is not permanent lung damage. You may recover lung function after ARDS. You will not recover from permanent lung damage. It will even deteriorate and your prognosis is less than 5 years to live. Stratification is a red herring, but feel free to do so.
Not everyone is 30-40?
1/500 needs a citation. Because the WHO and CDC as of a few days ago said 1/50. But even 1/500 is an incredibly high mortality rate for something as infectious as the flu, and to make light of it is bizarre. Even your number would make catching it about 1000x more dangerous than skydiving.
So 1/500 sounds like a correct ballpark. A few days ago there was some data released by China on some 40,000 cases. When you age adjust the mortality stats by: - Assuming uniform age distribution - Accounting for the age distribution of the population.

You get something like .4% for the 30-40 age range. This was a crude estimate... but that's why I say ballpark. It also could go up/down as we get more data.

I agree the risk here is considerably high. People just don't understand scale well. Assuming the R0 is high, it could eventually sweep through the majority of the population. IIRC, the 1918 flu pandemic is estimated to have infected 80% of the population. So in the US alone that's 6,000,000 deaths, or one Sept 11. like attack every day for 200 days.

Deaths are a lagging indicator. 2119 deaths as of yesterday / 0.004 = 530,000 cases that are ~2+ weeks old. In other words 530,000 infections around February 5th. That seems extremely unlikely based on the data we have.

For a reasonably accurate estimate you need to look at percentage of total infections detected, and the lag between detection and death. Using say 50% detection rate and 7 days between detection and death you get 2119 / (63,851 / 0.5) = 1.7% fatality rate. That’s about as optimistic as I think is reasonable.

PS: A 50% detection rate may be high or low, but that’s the real unknown.

I don't disagree that the data has very large error bars. I'm simply saying that, using the current best guess we can do the following: - Stratify relative risk based on population distribution. - Guage the impact of the expected bad case of ~2% mortality.

Will mortality rate come down? Very likely. But it may not... And if not, what is the outcome at scale?

What bothers me is the difficulty in interpreting such numbers because they’re not normalized to health.

E.g. dropping sperm counts in men. Is that dropping in all men? Or is it a side effect of rising obesity in enough men to dent the average?

Is the mortality 1/500 here because a good health 30 year old has a 1/500 chance of dying? Or because the 1/100 30 year olds with poor health have a 1/5 chance of dying?

The report says 2.3% killed, 2 out of every 100 is 2%.

11.5 is 2.3% of 500, thus 23 out of every 1000 would die at that rate if accurate.

1/500 * 6,000,000,000 is a lot of people