6 ms·

" ... I'm still slightly bearish on the whole situation due to the combination of the virus' absurdly high infection rate[0] and its ridiculous ~25-30 day incub

by rsync 7y ago

" ... I'm still slightly bearish on the whole situation due to the combination of the virus' absurdly high infection rate[0] and its ridiculous ~25-30 day incubation period ..." You're right to highlight those aspects of the virus and I find them noteworthy as well. However, the statistics I am most interested in is mortality rate and rate of asymptomatic infections. I note with interest that among the 700+ infected on the cruise ship, there were 6 deaths as of yesterday - and this is among a relatively geriatric population. So, perhaps we see a mortality rate of <1% among a greater risk population.

fauigerzigerk 7y ago
What I'm most interested in is the hospitalisation rate and if we can slow down infections enough for hospitals to be able to cope. I think the death rate will ultimately depend on whether everyone who needs hospital treatment can get it. And that's why I think the Diamond Princess may not be representative of what happens in the real world.
- learc83 7y ago
  On the people on the Diamond Princess there are 36 critical cases + 6 dead. So 42/705 infected, so 6% likely needed to be hospitalized. But, the demographics skewed older than the general population, and there were 3200 people on board in basically a worst case scenario for spread. So 42/3200 is ~1% of a much older population.
  - gdubs 7y ago
    Aren’t the hospitalization rates something like ~20% according to the main study that came out of China? Also, the cruise ship cases aren’t all resolved outcomes yet. Edit: link to China study summary below, but including it top-level here as well. http://www.cidrap.umn.edu/news-perspective/2020/02/study-72000-covid-19-patients-finds-23-death-rate http://www.cidrap.umn.edu/news-perspective/2020/02/study-720...
    - learc83 7y ago
      We don't have any idea what the hospitalization rates are. We only know what the hospitalization rates are for people who have a confirmed diagnosis, which skews towards people with severe symptoms. >Also, the cruise ship cases aren’t all resolved outcomes yet. No they aren't, but the vast majority have been symptomatic longer than the average time till hospitalization and the number in critical condition hasn't increased in days. It's very unlikely they'll get to anywhere near 20%. And this is in a more geriatric population.
      - gdubs 7y ago
        Except we do have data, that’s my point. China’s study is being cited by most experts as this situation unfolds. It doesn’t mean these numbers will hold — it’s possible that what we see out of South Korea over the coming month might be more representative — but it’s what we have, and it’s not nothing: http://www.cidrap.umn.edu/news-perspective/2020/02/study-72000-covid-19-patients-finds-23-death-rate http://www.cidrap.umn.edu/news-perspective/2020/02/study-720...
        treis 7y ago
        Lots of people on the ship that tested positive showed little to no symptoms. It's very likely that there are a large number of infections that have gone undetected. Until we test everyone we won't know the true number.
        learc83 7y ago
        The data can't support your 20% hospitalization claim though, and no one in public health is using that study to make that claim. The study itself makes no such claims that this study is in any way applicable to the general population. About 40% of the patients in that study weren't even tested. They were included based on symptoms that include severe pneumonia. The ones who were tested, sought testing because they displayed symptoms severe enough to seek treatment. This is the very definition of selection bias, and there is absolutely no reason to pretend that a 20% hospitalization rate is a realistic outcome.
    - 7y ago
  - fauigerzigerk 7y ago
    I would like to know how many of the 705 have recovered, how many are still ill, and how many were admitted to hospital. If 42 were critical, that doesn't necessarily mean they were the only ones who needed hospital treatment.
    - learc83 7y ago
      You can look at the counts by date. The majority are past the mean time till hospitalization, and the the number in critical condition hasn't increased in days. >If 42 were critical, that doesn't necessarily mean they were the only ones who needed hospital treatment. Percent hospitalized isn't the thing to look at. It's percentage of people who need intensive care or a ventilator that matters. Many people were hospitalized for observation. Some people received IV fluids. How many of those people were saved by IV fluids vs. just being a nice to have is unknown. But that kind of care can be done in mass tent hospitals or even at home.
      - fauigerzigerk 7y ago
        >You can look at the counts by date. Are these numbers still updated now that many (or all?) have left the ship? Where can I see those numbers? >Many people were hospitalized for observation Yes, but if fewer people had been under close observation in a hospital more might have died. Admitting only precisely those who turn critical is not realistic in my (lay person's) opinion. I agree with you about the age distribution on the cruise ship though. This is very important to note. Still, I have to wonder what share of the population can get hospital/ICU treatment at any given time. I don't know about other places, but here in the UK hospital bed occupancy rates are dangerously high at the best of times. I don't know how many beds/ICUs can be freed up in an emergency. But I doubt it's anywhere near enough if everyone gets sick at the same time. I hope someone is doing some modelling to find out how aggressively we have to slow down the rate of infection so that hospitals are not overwhelmed.
Retric 7y ago
The cruse ship had 135 infected on February 10th, jumping to 634 on February 20th. So, most of those 700 are recent infections. The actual long term death total for the cruse ship is unknown, but should include people who died in their home country after returning. The good news is assuming you trust China’s numbers new infections are down massively from a February 2nd peak. Deaths in China are also down significantly but less so, again assuming you trust their numbers.
- danenania 7y ago
  I feel like the slowing infection rate in China (if it's actually true) is both very encouraging in the sense that it could show the virus actually is possible to contain, and also pretty scary in that so far the only measures that seem to be working at all are extremely widespread and draconian quarantines.
  - NicoJuicy 7y ago
    China was handling it after 2 months of spreading ( hurray censorship). This isn't happening anywhere else in the world.
    - nostrademons 7y ago
      Hard to tell if this is sarcasm, given that researchers just announced that coronavirus has been spreading undetected for 6 weeks in the Seattle area: https://www.msn.com/en-us/news/us/coronavirus-may-have-spread-undetected-for-weeks-in-washington-state-which-reported-first-two-deaths-in-us/ar-BB10BIr0 https://www.msn.com/en-us/news/us/coronavirus-may-have-sprea...
      - NicoJuicy 7y ago
        Not exactly sarcasm, but we will find out if bullying out high-ranking members of the CDC by the Trump administration was a bad move :)
- learc83 7y ago
airstrike 7y ago
According to my fancy math based on numbers from Johns Hopkins COVID-19 dashboard[0][1], we're closer to a 7% mortality rate [0] direct link: https://www.arcgis.com/apps/opsdashboard/index.html#/bda7594740fd40299423467b48e9ecf6 https://www.arcgis.com/apps/opsdashboard/index.html#/bda7594... [1] posted to HN: https://news.ycombinator.com/item?id=22475060 https://news.ycombinator.com/item?id=22475060
- carlsborg 7y ago
  Same data, nicer dashboard and breaking news: https://coronastats.net https://coronastats.net
- vonmoltke 7y ago
  First, you are conflating mortality rate with case fatality rate (CFR). Mortality rate is deaths with respect to a total population; CFR is deaths with respect to diagnosed cases. Second, since this is an ongoing outbreak the CFR is a moving target. Your calculation assumes the ratio of deaths to recoveries remains what it is now for the ~40k active cases. However, the CFR by that calculation has been steadily falling. Additionally, since CFR is only based on diagnosed cases the struggles with testing capacity, particularly in the US, will distort it by biasing diagnoses towards more severe cases.
  - airstrike 7y ago
    Thanks for the clarification!
bryanrasmussen 7y ago
there's still 36 serious/critical. I have a hard time believing all of them will be fine. https://www.worldometers.info/coronavirus/#countries https://www.worldometers.info/coronavirus/#countries
lancyH 7y ago
> among the 700+ infected on the cruise ship, there were 6 deaths Your death count is incorrect. Several deaths occurred after cruise members returned to their home country, and are currently counted as deaths in their home country instead of the cruise ship. Take for example: 78-year-old man in Perth - got infected on the cruise ship - but counted as a death in Australia.
- bryanrasmussen 7y ago
  So I guess that means the 36 serious/critical I pointed out are not necessarily serious/critical at the moment either. Is there an actual place that counts that?
- samvher 7y ago
  This seems like a very important point, thanks. Many people are pointing at differences between nations and comparisons with Diamond Princess to derive different fatality rates than are commonly mentioned in the media and to point at an "iceberg" of unsymptomatic (and uncounted) cases. It's good to be aware that some of these discrepancies might just be artifacts from the way the data is registered.
istorical 7y ago
Most deaths occur in week 3 or week 4 if they make it that far.

by rsync 7y ago

fauigerzigerk 7y ago

What I'm most interested in is the hospitalisation rate and if we can slow down infections enough for hospitals to be able to cope. I think the death rate will ultimately depend on whether everyone who needs hospital treatment can get it. And that's why I think the Diamond Princess may not be representative of what happens in the real world.

learc83 7y ago
On the people on the Diamond Princess there are 36 critical cases + 6 dead. So 42/705 infected, so 6% likely needed to be hospitalized. But, the demographics skewed older than the general population, and there were 3200 people on board in basically a worst case scenario for spread. So 42/3200 is ~1% of a much older population.
- gdubs 7y ago
  Aren’t the hospitalization rates something like ~20% according to the main study that came out of China? Also, the cruise ship cases aren’t all resolved outcomes yet. Edit: link to China study summary below, but including it top-level here as well. http://www.cidrap.umn.edu/news-perspective/2020/02/study-72000-covid-19-patients-finds-23-death-rate http://www.cidrap.umn.edu/news-perspective/2020/02/study-720...
  - learc83 7y ago
    We don't have any idea what the hospitalization rates are. We only know what the hospitalization rates are for people who have a confirmed diagnosis, which skews towards people with severe symptoms. >Also, the cruise ship cases aren’t all resolved outcomes yet. No they aren't, but the vast majority have been symptomatic longer than the average time till hospitalization and the number in critical condition hasn't increased in days. It's very unlikely they'll get to anywhere near 20%. And this is in a more geriatric population.
    - gdubs 7y ago
      Except we do have data, that’s my point. China’s study is being cited by most experts as this situation unfolds. It doesn’t mean these numbers will hold — it’s possible that what we see out of South Korea over the coming month might be more representative — but it’s what we have, and it’s not nothing: http://www.cidrap.umn.edu/news-perspective/2020/02/study-72000-covid-19-patients-finds-23-death-rate http://www.cidrap.umn.edu/news-perspective/2020/02/study-720...
      - treis 7y ago
        Lots of people on the ship that tested positive showed little to no symptoms. It's very likely that there are a large number of infections that have gone undetected. Until we test everyone we won't know the true number.
      - learc83 7y ago
        The data can't support your 20% hospitalization claim though, and no one in public health is using that study to make that claim. The study itself makes no such claims that this study is in any way applicable to the general population. About 40% of the patients in that study weren't even tested. They were included based on symptoms that include severe pneumonia. The ones who were tested, sought testing because they displayed symptoms severe enough to seek treatment. This is the very definition of selection bias, and there is absolutely no reason to pretend that a 20% hospitalization rate is a realistic outcome.
  - 7y ago
- fauigerzigerk 7y ago
  I would like to know how many of the 705 have recovered, how many are still ill, and how many were admitted to hospital. If 42 were critical, that doesn't necessarily mean they were the only ones who needed hospital treatment.
  - learc83 7y ago
    You can look at the counts by date. The majority are past the mean time till hospitalization, and the the number in critical condition hasn't increased in days. >If 42 were critical, that doesn't necessarily mean they were the only ones who needed hospital treatment. Percent hospitalized isn't the thing to look at. It's percentage of people who need intensive care or a ventilator that matters. Many people were hospitalized for observation. Some people received IV fluids. How many of those people were saved by IV fluids vs. just being a nice to have is unknown. But that kind of care can be done in mass tent hospitals or even at home.
    - fauigerzigerk 7y ago
      >You can look at the counts by date. Are these numbers still updated now that many (or all?) have left the ship? Where can I see those numbers? >Many people were hospitalized for observation Yes, but if fewer people had been under close observation in a hospital more might have died. Admitting only precisely those who turn critical is not realistic in my (lay person's) opinion. I agree with you about the age distribution on the cruise ship though. This is very important to note. Still, I have to wonder what share of the population can get hospital/ICU treatment at any given time. I don't know about other places, but here in the UK hospital bed occupancy rates are dangerously high at the best of times. I don't know how many beds/ICUs can be freed up in an emergency. But I doubt it's anywhere near enough if everyone gets sick at the same time. I hope someone is doing some modelling to find out how aggressively we have to slow down the rate of infection so that hospitals are not overwhelmed.

Retric 7y ago

The cruse ship had 135 infected on February 10th, jumping to 634 on February 20th. So, most of those 700 are recent infections. The actual long term death total for the cruse ship is unknown, but should include people who died in their home country after returning. The good news is assuming you trust China’s numbers new infections are down massively from a February 2nd peak. Deaths in China are also down significantly but less so, again assuming you trust their numbers.

danenania 7y ago
I feel like the slowing infection rate in China (if it's actually true) is both very encouraging in the sense that it could show the virus actually is possible to contain, and also pretty scary in that so far the only measures that seem to be working at all are extremely widespread and draconian quarantines.
- NicoJuicy 7y ago
  China was handling it after 2 months of spreading ( hurray censorship). This isn't happening anywhere else in the world.
  - nostrademons 7y ago
    Hard to tell if this is sarcasm, given that researchers just announced that coronavirus has been spreading undetected for 6 weeks in the Seattle area: https://www.msn.com/en-us/news/us/coronavirus-may-have-spread-undetected-for-weeks-in-washington-state-which-reported-first-two-deaths-in-us/ar-BB10BIr0 https://www.msn.com/en-us/news/us/coronavirus-may-have-sprea...
    - NicoJuicy 7y ago
      Not exactly sarcasm, but we will find out if bullying out high-ranking members of the CDC by the Trump administration was a bad move :)
learc83 7y ago

airstrike 7y ago

According to my fancy math based on numbers from Johns Hopkins COVID-19 dashboard[0][1], we're closer to a 7% mortality rate [0] direct link: https://www.arcgis.com/apps/opsdashboard/index.html#/bda7594740fd40299423467b48e9ecf6 https://www.arcgis.com/apps/opsdashboard/index.html#/bda7594... [1] posted to HN: https://news.ycombinator.com/item?id=22475060 https://news.ycombinator.com/item?id=22475060

carlsborg 7y ago
Same data, nicer dashboard and breaking news: https://coronastats.net https://coronastats.net
vonmoltke 7y ago
First, you are conflating mortality rate with case fatality rate (CFR). Mortality rate is deaths with respect to a total population; CFR is deaths with respect to diagnosed cases. Second, since this is an ongoing outbreak the CFR is a moving target. Your calculation assumes the ratio of deaths to recoveries remains what it is now for the ~40k active cases. However, the CFR by that calculation has been steadily falling. Additionally, since CFR is only based on diagnosed cases the struggles with testing capacity, particularly in the US, will distort it by biasing diagnoses towards more severe cases.
- airstrike 7y ago
  Thanks for the clarification!

bryanrasmussen 7y ago

there's still 36 serious/critical. I have a hard time believing all of them will be fine. https://www.worldometers.info/coronavirus/#countries https://www.worldometers.info/coronavirus/#countries

lancyH 7y ago

> among the 700+ infected on the cruise ship, there were 6 deaths Your death count is incorrect. Several deaths occurred after cruise members returned to their home country, and are currently counted as deaths in their home country instead of the cruise ship. Take for example: 78-year-old man in Perth - got infected on the cruise ship - but counted as a death in Australia.

bryanrasmussen 7y ago
So I guess that means the 36 serious/critical I pointed out are not necessarily serious/critical at the moment either. Is there an actual place that counts that?
samvher 7y ago
This seems like a very important point, thanks. Many people are pointing at differences between nations and comparisons with Diamond Princess to derive different fatality rates than are commonly mentioned in the media and to point at an "iceberg" of unsymptomatic (and uncounted) cases. It's good to be aware that some of these discrepancies might just be artifacts from the way the data is registered.

istorical 7y ago

Most deaths occur in week 3 or week 4 if they make it that far.