3 ms·

I don't have the background to answer your questions, but even if I steelman and assume the worst outcome here (issues with the test, issues with epidemiology,

I don't have the background to answer your questions, but even if I steelman and assume the worst outcome here (issues with the test, issues with epidemiology, that we're crowding out hospitals, that we're infecting via hospitals), I would argue that we are still not overreacting: In the original outbreak of wuhan, there could not have been overpreparation because there was no preparation, the outbreak simply happened. So, we should look at wuhan pre-citywide-quarantine to get a feel for what it would look like if we chose to not "overreact", as originally there was no real "test" being administered, cases were diagnosed by CT scans of lungs for pneumonia, and there were not extreme imposed social distancing or population-wide quarantine. So, any issues we might have in the west, should not apply to the original outbreak in wuhan. And in wuhan the casualties of the outbreak were on the order of 1-5% (i think estimates varied a lot for some of the reasons you list). In any case, it overloaded hospitals and a lot of people died. If the infection reaches a broader population (the 'underreaction' scenario), it suffices to say the death rate will be a MINIMUM of 1-5%, as obviously the situation would be even worse with a large percentage of the population sick. This would continue until the population reaches herd immunity, estimated 30-60% of population infected. Thus, we can conclude 1-5% of that 30-60% of the infected population would die, which in america ranges from about 1 million to nearly 10 million deaths. This is far far far greater than the yearly flu outbreak. And that's assuming that the rapid spread of the disease doesn't cause secondary effects through deteriorating supply chains (who staffs grocery stores if 60% of the population is home? who drives trucks to keep supply chains going? who works the amazon fulfillment center?) Thus, I would conclude that the penalties for unabated spread of the disease are so horrific that the only reasonable response is something that looks like an overreaction: extreme social distancing, shutting down cities aside from vital functions, etc. Because atleast our groceries stores operate and our supply chains are running.

thu2111 7y ago
1% to 5% of what? This is the kind of statement that raises suspicions of over-reaction. https://www.nytimes.com/2020/03/19/health/wuhan-coronavirus-deaths.html https://www.nytimes.com/2020/03/19/health/wuhan-coronavirus-... As of yesterday the best data suggests a 1% fatality rate amongst confirmed cases in Wuhan. Wuhan has 2169 confirmed deaths in a population of 11 million, so if we speak of the casuality rate for the population it would be 0.019%. Is that a lot? Well, it's the sort of level that's lost in the noise of regular influenza. Thus, we can conclude 1-5% of that 30-60% of the infected population would die No, we can't conclude that. You're taking a death rate of a very selectively chosen sample and then projecting it onto the entire population, which isn't statistically valid. People keep making basic logic errors like this one, which is what triggers questions like the OPs about possible over-reactions. All the data we have is useless. Not a single stat has a level of accuracy that would normally be accepted for scientific use. We don't know what the infection rate is, which means we don't know how many people get sick and die, we don't know even what the mortality rate is because in fact we don't even have a globally agreed definition of COVID-19 mortality. Italy examined their own death records and discovered that of all deaths marked as caused by CV, only 2 appeared to have no pre-existing or underlying health conditions. If that's true then SARS-CoV-2 is sort of like HIV. That makes it very hard to establish precise definitions. If someone has been battling a severe health problem for months, is very likely to die anyway, they die, and then their blood showed they got SARS-CoV-2 two days before dying - what was the cause of death? Health systems don't agree on how to answer this question which makes comparing numbers between locations very tricky. This is being floated as a possible explanation for the divergent death rates between Germany and Italy. who staffs grocery stores if 60% of the population is home? Here you're not using temporal reasoning. Why would 60% of the population be at home simultaneously except via forced government fiat? Even if you assume everyone has to stay at home until they get the disease (worst case scenario, not implemented anywhere) people don't all get sick instantaneously, and once recovered people can go back to work as they aren't a threat to anyone. Most people recover within a week. So, to ask a question about how many people will be out of the workforce requires discussion of timeframes and curve slopes. Something to be aware of is that the "60-70% of the population will get it" figures are based on a form of modelling that assumes everyone in the population interacts with everyone else. That simplifies the maths but clearly isn't close to being true. Thus, I would conclude that the penalties for unabated spread of the disease are so horrific that the only reasonable response is something that looks like an overreaction And yet your post contains statistical and logical errors, which is why the number of people questioning that sort of conclusion is going to be non-trivial for the foreseeable future.

thu2111 7y ago

1% to 5% of what? This is the kind of statement that raises suspicions of over-reaction. https://www.nytimes.com/2020/03/19/health/wuhan-coronavirus-deaths.html https://www.nytimes.com/2020/03/19/health/wuhan-coronavirus-... As of yesterday the best data suggests a 1% fatality rate amongst confirmed cases in Wuhan. Wuhan has 2169 confirmed deaths in a population of 11 million, so if we speak of the casuality rate for the population it would be 0.019%. Is that a lot? Well, it's the sort of level that's lost in the noise of regular influenza. Thus, we can conclude 1-5% of that 30-60% of the infected population would die No, we can't conclude that. You're taking a death rate of a very selectively chosen sample and then projecting it onto the entire population, which isn't statistically valid. People keep making basic logic errors like this one, which is what triggers questions like the OPs about possible over-reactions. All the data we have is useless. Not a single stat has a level of accuracy that would normally be accepted for scientific use. We don't know what the infection rate is, which means we don't know how many people get sick and die, we don't know even what the mortality rate is because in fact we don't even have a globally agreed definition of COVID-19 mortality. Italy examined their own death records and discovered that of all deaths marked as caused by CV, only 2 appeared to have no pre-existing or underlying health conditions. If that's true then SARS-CoV-2 is sort of like HIV. That makes it very hard to establish precise definitions. If someone has been battling a severe health problem for months, is very likely to die anyway, they die, and then their blood showed they got SARS-CoV-2 two days before dying - what was the cause of death? Health systems don't agree on how to answer this question which makes comparing numbers between locations very tricky. This is being floated as a possible explanation for the divergent death rates between Germany and Italy. who staffs grocery stores if 60% of the population is home? Here you're not using temporal reasoning. Why would 60% of the population be at home simultaneously except via forced government fiat? Even if you assume everyone has to stay at home until they get the disease (worst case scenario, not implemented anywhere) people don't all get sick instantaneously, and once recovered people can go back to work as they aren't a threat to anyone. Most people recover within a week. So, to ask a question about how many people will be out of the workforce requires discussion of timeframes and curve slopes. Something to be aware of is that the "60-70% of the population will get it" figures are based on a form of modelling that assumes everyone in the population interacts with everyone else. That simplifies the maths but clearly isn't close to being true. Thus, I would conclude that the penalties for unabated spread of the disease are so horrific that the only reasonable response is something that looks like an overreaction And yet your post contains statistical and logical errors, which is why the number of people questioning that sort of conclusion is going to be non-trivial for the foreseeable future.