Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

How to fight the coronavirus SARS-CoV-2 and its disease, Covid-19 [pdf]

drive.google.com

Post reply on HN
Print view View original

71–77 of 77 posts

Re: How to fight the coronavirus SARS-CoV-2 and its disease, Covid-19 [pdf]

#71

Post by devy »

natalyarostova wrote: post #28
devy wrote: post #26

It's disingenuous to compare COVID-19 death with traffic death in Dr. Lin's presentation. Here are the facts: - The 30k U.S. traffic death count is for the ENTIRE YEAR. The 3k death in Wuhun happened in a matter of weeks(6-8 weeks). - Given that U.S.'s inaction and squandered 2 months of lead time, we could see death count to be in the order of hundreds of thousands in U.S. alone.[1] This is NOT an estimate based on…

I didn’t get the impression he was comparing them. Just providing reference points. He’s pretty clear that this is bad.

The closest outbreak to COVID-19 is probably the 1918 Spanish flu (forward-looking statement). And we are going to have 100% likelihood that this year's COVID-19 death count is going to surpass 30k people globally.

Re: How to fight the coronavirus SARS-CoV-2 and its disease, Covid-19 [pdf]

#72

Post by tunesmith »

raphlinus wrote: post #70
tunesmith wrote: post #12

Earlier quoted context omitted.

The doubling time of deaths appears to be much slower than the doubling time of confirmed cases, and is probably a better comparison. It could be slightly faster than reports, if deaths are getting missed, or misattributed to other illnesses. But it still seems more accurate than judging the doubling time of confirmed cases, which is more just a measure of how quickly we are able to increase testing when there's a ba…

Where do you get that the doubling time of deaths is much slower than that of confirmed cases? I looked for data, and this graphic is not reassuring on the question: https://commons.wikimedia.org/wiki/File:Log-linear_plot_of_c...

I was eyeballing various country deaths at worldometers.info, and they looked more like 8-11 days to me. Then again, the early South Korea stats looked steeper.

But yeah, I recognize the larger point is we want to know doubling rate of "actual cases", and since that's currently unknowable, identifying the most reliable proxy. It just seemed to me that our testing rate occludes "confirmed cases" more than death rate occludes "actual cases" - at least with death rates we can identify the time points when mitigation rates are introduced, when hospitals get overwhelmed, etc.

Re: How to fight the coronavirus SARS-CoV-2 and its disease, Covid-19 [pdf]

#73

Post by sn9 »

This analysis seems largely correct but for one thing, the author assumes that people who contract the virus will develop immunity that lasts long enough to bend the exponential growth curve as herd immunity develops.

(The curve will slow for other reasons as a sufficient number of people get infected such that there just aren't enough people who haven't been infected to be newly infected.)

I didn't see a source for that --- it seemed like an assumption --- and everything I've read suggests that this is a major unknown at the moment.

It's possible the author has a source that was not included. Either way, the author should either include the source or more explicitly take describe the epistemic status of that hypothesis and work through the consequences of it not being true.

Re: How to fight the coronavirus SARS-CoV-2 and its disease, Covid-19 [pdf]

#75

Post by chme »

sn9 wrote: post #73

This analysis seems largely correct but for one thing, the author assumes that people who contract the virus will develop immunity that lasts long enough to bend the exponential growth curve as herd immunity develops. (The curve will slow for other reasons as a sufficient number of people get infected such that there just aren't enough people who haven't been infected to be newly infected.) I didn't see a source for…

> This analysis seems largely correct but for one thing, the author assumes that people who contract the virus will develop immunity that lasts long enough to bend the exponential growth curve as herd immunity develops.

I am not a medical professional, but I would have assumed that the body develops a immunity because otherwise why would they search for a vaccine?

Re: How to fight the coronavirus SARS-CoV-2 and its disease, Covid-19 [pdf]

#77

Post by sn9 »

chme wrote: post #75
sn9 wrote: post #73

This analysis seems largely correct but for one thing, the author assumes that people who contract the virus will develop immunity that lasts long enough to bend the exponential growth curve as herd immunity develops. (The curve will slow for other reasons as a sufficient number of people get infected such that there just aren't enough people who haven't been infected to be newly infected.) I didn't see a source for…

> This analysis seems largely correct but for one thing, the author assumes that people who contract the virus will develop immunity that lasts long enough to bend the exponential growth curve as herd immunity develops. I am not a medical professional, but I would have assumed that the body develops a immunity because otherwise why would they search for a vaccine?

I guess I was being overly broad. I'm not an immunologist or anything.

I was thinking more of the kind of immunity one gets for something like chicken pox, where one childhood infection basically protects you for life, vs something like the seasonal flu which you can get again and again.

Now I don't know if the reason you keep getting the flu is because it mutates more rapidly than the chicken pox virus and you still remain immune to the particular strain you got infected with or if that initial immunity weakens over time.

Maybe someone better informed can chime in.

Post reply on HN