4 ms·

This is what I fundamentally don't understand. The condition can ONLY improve in terms of new cases, not deaths, 1) if there is a vaccine, 2) 40-50% people get

by yumraj 7y ago

This is what I fundamentally don't understand. The condition can ONLY improve in terms of new cases, not deaths, 1) if there is a vaccine, 2) 40-50% people get infected and develop immunity, or 3) people start living in bubbles, including masks and gloves and stay away from each other, that there spread is controlled. We know 1 and 2 have not happened. So either China is still in lockdown and this lockdown is how the life will be till there is a vaccine, or the data is not correct.

postalrat 7y ago
How do you know 2 hasn't happened?
- yumraj 7y ago
  Based on the published data. Also, 40-50% of China's population would be 600-800 million, or so. With that even at 1% 6-8 million would have died. Even with filtered data, it is doubtful that they would have been able to hide so many dead.
  - abhorrence 7y ago
    That assumes the death rate is actually 1%. In the hypothetical where 40%+ have been infected, I think it’s be safe to assume the death rate might be lower than we suspect it is.
    - zaroth 7y ago
      It was the same during the H1N1 hysteria. CFRs being quoted of 2-3% and it turned out much less deadly than an average flu. A few weeks ago we didn’t even know there were asymptomatic carriers. My prediction is that when all is said and done and the IFR is closer to 0.01% that this will all look very stupid. Experts in Ohio say they suspect that 1% of people are infected. OK, so where are the hospital cases?
      - Spare_account 7y ago
        At 0.01%, Italy would expect to see around 6,000 deaths. They're up to 1200 as of today and don't appear to have be nearing an inflection point. We should know soon I guess...
      - Symmetry 7y ago
        South Korea is doing some random testing to figure out how many cases only looking at symptomatic patients is missing. With that it looks like the total fatality ratio is just .6% for them. But of course they haven't had hospitals overwhelmed with too many patients at once and I expect that Italy will see a much higher fatality ratio.
        zaroth 7y ago
        I look forward to statistically valid serological (antibody) testing, but this usually takes a year or two to be released.
      - Scarblac 7y ago
        The hospital cases come like two weeks after the infections.
        zaroth 7y ago
        We're talking about a number on the order of 2^17. Doubling every 6 days that's 100 days. Doubling every 3 days that's still 51 days. You cannot have 100,000 cases with even a 1% hospitalization rate without at least... on the order of 250 hospitalizations.
  - rumanator 7y ago
    > With that even at 1% 6-8 million would have died. That number assumes a homogeneous distribution. As deaths are observed mostly in the elderly and China's demographics tapers off rapidly beyond 60yo, the incidence would be far lower than 1% of the total population. https://www.indexmundi.com/graphs/population-pyramids/china-population-pyramid-2018.jpg https://www.indexmundi.com/graphs/population-pyramids/china-...
SpicyLemonZest 7y ago
It could be that measures less severe than "masks and gloves and stay away from each other" are enough to bring R0 below 1. Maybe just not having people go out when sick with coronavirus is enough, and all the other measures are just a matter of bringing outbreaks to an end faster with a smaller peak. South Korea has definitively shown that the coronavirus can be controlled, even if you don't trust China.
- yumraj 7y ago
  > It could be that measures less severe than "masks and gloves and stay away from each other" are enough to bring R0 below 1. That is a very good point. I think you're right, focusing on R0 is the right thing and that might just be the key since estimated R0 was ~1.x-3.x, so perhaps masks, gloves and some common sense can indeed bring it down to <1.

by yumraj 7y ago

postalrat 7y ago

How do you know 2 hasn't happened?

yumraj 7y ago
Based on the published data. Also, 40-50% of China's population would be 600-800 million, or so. With that even at 1% 6-8 million would have died. Even with filtered data, it is doubtful that they would have been able to hide so many dead.
- abhorrence 7y ago
  That assumes the death rate is actually 1%. In the hypothetical where 40%+ have been infected, I think it’s be safe to assume the death rate might be lower than we suspect it is.
  - zaroth 7y ago
    It was the same during the H1N1 hysteria. CFRs being quoted of 2-3% and it turned out much less deadly than an average flu. A few weeks ago we didn’t even know there were asymptomatic carriers. My prediction is that when all is said and done and the IFR is closer to 0.01% that this will all look very stupid. Experts in Ohio say they suspect that 1% of people are infected. OK, so where are the hospital cases?
    - Spare_account 7y ago
      At 0.01%, Italy would expect to see around 6,000 deaths. They're up to 1200 as of today and don't appear to have be nearing an inflection point. We should know soon I guess...
    - Symmetry 7y ago
      South Korea is doing some random testing to figure out how many cases only looking at symptomatic patients is missing. With that it looks like the total fatality ratio is just .6% for them. But of course they haven't had hospitals overwhelmed with too many patients at once and I expect that Italy will see a much higher fatality ratio.
      - zaroth 7y ago
        I look forward to statistically valid serological (antibody) testing, but this usually takes a year or two to be released.
    - Scarblac 7y ago
      The hospital cases come like two weeks after the infections.
      - zaroth 7y ago
        We're talking about a number on the order of 2^17. Doubling every 6 days that's 100 days. Doubling every 3 days that's still 51 days. You cannot have 100,000 cases with even a 1% hospitalization rate without at least... on the order of 250 hospitalizations.
- rumanator 7y ago
  > With that even at 1% 6-8 million would have died. That number assumes a homogeneous distribution. As deaths are observed mostly in the elderly and China's demographics tapers off rapidly beyond 60yo, the incidence would be far lower than 1% of the total population. https://www.indexmundi.com/graphs/population-pyramids/china-population-pyramid-2018.jpg https://www.indexmundi.com/graphs/population-pyramids/china-...

SpicyLemonZest 7y ago

It could be that measures less severe than "masks and gloves and stay away from each other" are enough to bring R0 below 1. Maybe just not having people go out when sick with coronavirus is enough, and all the other measures are just a matter of bringing outbreaks to an end faster with a smaller peak. South Korea has definitively shown that the coronavirus can be controlled, even if you don't trust China.

yumraj 7y ago
> It could be that measures less severe than "masks and gloves and stay away from each other" are enough to bring R0 below 1. That is a very good point. I think you're right, focusing on R0 is the right thing and that might just be the key since estimated R0 was ~1.x-3.x, so perhaps masks, gloves and some common sense can indeed bring it down to <1.