5 ms·

Swag maths on the business-as-usual scenario incoming, bound to be very wrong: 3.4% mortality rate among those presenting symptoms. (via WHO) As low as 17.9%

Swag maths on the business-as-usual scenario incoming, bound to be very wrong: 3.4% mortality rate among those presenting symptoms. (via WHO) As low as 17.9% of cases asymptomatic: https://www.forbes.com/sites/brucelee/2020/03/18/what-percentage-have-covid-19-coronavirus-but-do-not-know-it/#42991ddf7e90 https://www.forbes.com/sites/brucelee/2020/03/18/what-percen... Of the symptomatic cases, 80.9% are mild (https://www.sciencealert.com/large-chinese-study-finds-most-coronavirus-infections-are-mild https://www.sciencealert.com/large-chinese-study-finds-most-...) and we're assuming only the ones that present risks are being tested. So if the 25.5m figure stands correct, we're looking at what, a range of 195,000 to 695,000 possibly passed on in california before June depending on whether you apply it to not-mild (19.1%) cases v. all symptomatic (82.1%) cases? I can't find any concrete sources on how anyone is calculating current mortality rates for the bug, but: 1. that swing is unfathomable to me 2. those numbers are unfathomable to me. Especially once extrapolated to the whole country assuming 56% of the nation contracts it.

wy3yebdudu 7y ago
Not terrible considering state population is roughly 39,560,000. It'll be interesting to see how this goes the illegal alien communities.
op03 7y ago
Do the math for your neighborhood or county and you will sleep better. State numbers and national numbers for anything can freak anyone out if they dont deal with large numbers day in day out. Big numbers hide small numbers distributed all over the map. Plug in numbers for India or China and it's not complicated to produce helplessness and heatattacks.
- jader201 7y ago
  According to these numbers, all you have to do is look at the percentage: 56%. That’s a big number in any scale. That’s over half of your family and friends. Not arguing these numbers are accurate. Just saying that they’ll still freak you out if you believe they’re accurate.
NotAnEconomist 7y ago
My back of the napkin math indicates that roughly 2x as many people will die in the US this year. Normal deaths are 2.8M/year. There's 330M people in the US. At 70% infection, that's 230M people. At 1% mortality, that's 2.3M excess deaths. That's 80% of the normal deaths for a year; and since other causes of death haven't stopped... we get 180% for the year, or "twice as many".
- jameslevy 7y ago
  Perhaps within the next five years we will see the number of deaths decline, as a pandemic results in people who would die in the next few years instead die earlier.
- Drunk_Engineer 7y ago
  Most coronavirus victims were already in poor health. So your analysis is double-counting.
  - beaunative 7y ago
    TBH, it's impossible to 'not double-counting'. We don't live in math.
njarboe 7y ago
This article[1] states that our best natural experiment for COVID-19 is the Diamond Princess cruise ship and its quarantine passengers. From that he estimates the the mortality rate is between 0.05% and 1%. That is, we don't really know what the rate is. The WHO giving a number with two significant digits implying some reasonable certainty about the number, is very deceiving and very likely just plain wrong. [1]https://www.statnews.com/2020/03/17/a-fiasco-in-the-making-as-the-coronavirus-pandemic-takes-hold-we-are-making-decisions-without-reliable-data/ https://www.statnews.com/2020/03/17/a-fiasco-in-the-making-a... By:John P.A. Ioannidis professor of medicine, of epidemiology and population health, of biomedical data science, and of statistics at Stanford University and co-director of Stanford’s Meta-Research Innovation Center.
- epistasis 7y ago
  I don't know a single scientist that thinks that Ioannidis is on the right track there. He's known for being a contrarian, which can be a useful thing. But it certainly doesn't make him right, in fact he's quite likely wrong. And he's absolutely wrong, as he's not managing risk, but looking to make the minimum deviance from normal activity. Ie a very foolish take. It's sad that Stat News publishes that in this time. But at least they published this the next day: https://www.statnews.com/2020/03/18/we-know-enough-now-to-act-decisively-against-covid-19/ https://www.statnews.com/2020/03/18/we-know-enough-now-to-ac...
  - wahern 7y ago
    He can't be wrong because he deftly avoids making any conclusions. All he's expressing is doubt and uncertainty, which is 2/3rds of FUD. I think it goes without saying that the numbers we have (whichever ones you choose) are wrong, even to an order of magnitude or more. But the responsible, constructive way during an emergency to express a contrarian opinion is by identifying facts and crafting reasoned conclusions. That includes doing enough due diligence to have some reasonable confidence that you understand the premises behind the public authorities' responses. For example, the clean data he wants to see might not even exist. Telling a blind man that he's blind isn't helpful; he knows that and is presumably (hopefully) making decisions accordingly, such as keeping a wide berth to any potential hazards. Here's a small handful of what we do know, independent of the cold, hard, vanilla statistics on SARS-CoV-2 that we'd all like to have: Italy is experiencing a public health crisis, whether justified by the number of deaths or not; two national lawmakers in Iran have died (how many die annually of seasonal flu?); Wuhan was no joke; initial outbreak in the U.S. so far seems to align with its evolution elsewhere. All these things are also evidence that can and should be part of drafting early responses. These aren't population mortality rates, r0, infection fatality rates, or the like, but evidence nonetheless; evidence that should be heeded in the absence of more concrete quantitative evidence.
realusername 7y ago
Those mortality percentages are even likely to be even worse in the US since the US has one the highest rate of diabetes of the developed world and for some reason that seems to affect a lot the virus. I really hope that the worst case scenario can be avoided.
all_blue_chucks 7y ago
When we run out of ICU capacity the fatality rate goes up a lot.
- timdorr 7y ago
  And when we deploy effective treatment plans, the fatility rate goes down. Or the virus mutates with deletions, the fatility rate goes down. Or when testing capacity finally reaches demand and lets us find infected folks sooner, the fatility rate goes down. I'm not trying to be mean to you, just pointing out that there are potential positive outcomes ahead. We honestly don't know what's going to happen next, but there are plenty of reasons to have hope.
  - threeseed 7y ago
    > We honestly don't know what's going to happen next Actually you sort of do. Look at Italy, China, South Korea all of whom have had the virus for a longer period than the US. And given that US has done too little, too late it's looking a lot more like Italy than China which can't give anybody much reason to hope.

wy3yebdudu 7y ago

Not terrible considering state population is roughly 39,560,000. It'll be interesting to see how this goes the illegal alien communities.

op03 7y ago

Do the math for your neighborhood or county and you will sleep better. State numbers and national numbers for anything can freak anyone out if they dont deal with large numbers day in day out. Big numbers hide small numbers distributed all over the map. Plug in numbers for India or China and it's not complicated to produce helplessness and heatattacks.

jader201 7y ago
According to these numbers, all you have to do is look at the percentage: 56%. That’s a big number in any scale. That’s over half of your family and friends. Not arguing these numbers are accurate. Just saying that they’ll still freak you out if you believe they’re accurate.

NotAnEconomist 7y ago

My back of the napkin math indicates that roughly 2x as many people will die in the US this year. Normal deaths are 2.8M/year. There's 330M people in the US. At 70% infection, that's 230M people. At 1% mortality, that's 2.3M excess deaths. That's 80% of the normal deaths for a year; and since other causes of death haven't stopped... we get 180% for the year, or "twice as many".

jameslevy 7y ago
Perhaps within the next five years we will see the number of deaths decline, as a pandemic results in people who would die in the next few years instead die earlier.
Drunk_Engineer 7y ago
Most coronavirus victims were already in poor health. So your analysis is double-counting.
- beaunative 7y ago
  TBH, it's impossible to 'not double-counting'. We don't live in math.

njarboe 7y ago

This article[1] states that our best natural experiment for COVID-19 is the Diamond Princess cruise ship and its quarantine passengers. From that he estimates the the mortality rate is between 0.05% and 1%. That is, we don't really know what the rate is. The WHO giving a number with two significant digits implying some reasonable certainty about the number, is very deceiving and very likely just plain wrong. [1]https://www.statnews.com/2020/03/17/a-fiasco-in-the-making-as-the-coronavirus-pandemic-takes-hold-we-are-making-decisions-without-reliable-data/ https://www.statnews.com/2020/03/17/a-fiasco-in-the-making-a... By:John P.A. Ioannidis professor of medicine, of epidemiology and population health, of biomedical data science, and of statistics at Stanford University and co-director of Stanford’s Meta-Research Innovation Center.

epistasis 7y ago
I don't know a single scientist that thinks that Ioannidis is on the right track there. He's known for being a contrarian, which can be a useful thing. But it certainly doesn't make him right, in fact he's quite likely wrong. And he's absolutely wrong, as he's not managing risk, but looking to make the minimum deviance from normal activity. Ie a very foolish take. It's sad that Stat News publishes that in this time. But at least they published this the next day: https://www.statnews.com/2020/03/18/we-know-enough-now-to-act-decisively-against-covid-19/ https://www.statnews.com/2020/03/18/we-know-enough-now-to-ac...
- wahern 7y ago
  He can't be wrong because he deftly avoids making any conclusions. All he's expressing is doubt and uncertainty, which is 2/3rds of FUD. I think it goes without saying that the numbers we have (whichever ones you choose) are wrong, even to an order of magnitude or more. But the responsible, constructive way during an emergency to express a contrarian opinion is by identifying facts and crafting reasoned conclusions. That includes doing enough due diligence to have some reasonable confidence that you understand the premises behind the public authorities' responses. For example, the clean data he wants to see might not even exist. Telling a blind man that he's blind isn't helpful; he knows that and is presumably (hopefully) making decisions accordingly, such as keeping a wide berth to any potential hazards. Here's a small handful of what we do know, independent of the cold, hard, vanilla statistics on SARS-CoV-2 that we'd all like to have: Italy is experiencing a public health crisis, whether justified by the number of deaths or not; two national lawmakers in Iran have died (how many die annually of seasonal flu?); Wuhan was no joke; initial outbreak in the U.S. so far seems to align with its evolution elsewhere. All these things are also evidence that can and should be part of drafting early responses. These aren't population mortality rates, r0, infection fatality rates, or the like, but evidence nonetheless; evidence that should be heeded in the absence of more concrete quantitative evidence.

realusername 7y ago

Those mortality percentages are even likely to be even worse in the US since the US has one the highest rate of diabetes of the developed world and for some reason that seems to affect a lot the virus. I really hope that the worst case scenario can be avoided.

all_blue_chucks 7y ago

When we run out of ICU capacity the fatality rate goes up a lot.

timdorr 7y ago
And when we deploy effective treatment plans, the fatility rate goes down. Or the virus mutates with deletions, the fatility rate goes down. Or when testing capacity finally reaches demand and lets us find infected folks sooner, the fatility rate goes down. I'm not trying to be mean to you, just pointing out that there are potential positive outcomes ahead. We honestly don't know what's going to happen next, but there are plenty of reasons to have hope.
- threeseed 7y ago
  > We honestly don't know what's going to happen next Actually you sort of do. Look at Italy, China, South Korea all of whom have had the virus for a longer period than the US. And given that US has done too little, too late it's looking a lot more like Italy than China which can't give anybody much reason to hope.