-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Coronavirus
- Thread starter Wenche
- Start date
this is classic missing the forest for the trees right here.
Clearly the chart is showing that people in Lombardy are dying at a much higher rate than the rest of Italy. This is useful to show, combined with on the ground reports, that Lombardy was not able to flatten the curve in time and it has led to a higher than necessary fatality rate. Yet you get hung up on the fact that Italy has more undiagnosed cases and that neither rate is fully accurate missing the whole point of the comparison.
I’ve made this point before in this thread but the people advising us that the US is NOT taking it seriously are doing so as professionals. They are the ones sounding the alarm. Yet you and Elon are here to tell us the panic is dumb when the professionals, who are accounting for the things you and Elon claim to be experts on, are saying the exact opposite.
EVNow
Well-Known Member
Yes, that gives low numbers in a rapidly increasing outbreak. Is that what you are saying
Here is a good link, someone posted above.
Why this epidemiologist is more worried about coronavirus than he was a month ago
First, some definitions from Steven Riley at Imperial College. The infection fatality rate (IFR) gives the probability of dying for an infected person. The case fatality rate (CFR) gives the probability of dying for an infected person who is sick enough to report to a hospital or clinic. CFR is larger than IFR, because individuals who report to hospitals are typically more severely ill.
….
Scientists working at the London School of Hygiene and Tropical Medicine, Imperial College London and the Institute for Disease Modeling have used these approaches to estimate the infection fatality rate. Currently, these estimates range from 0.5% to 0.94% indicating that COVID-19 is about 10 to 20 times as deadly as seasonal influenza. Evidence coming in from genomics and large-scale testing of fevers is consistent with these conclusions. The only potentially good news is that the epidemic in Korea may ultimately show a lower CFR than the epidemic in China.
….
Scientists working at the London School of Hygiene and Tropical Medicine, Imperial College London and the Institute for Disease Modeling have used these approaches to estimate the infection fatality rate. Currently, these estimates range from 0.5% to 0.94% indicating that COVID-19 is about 10 to 20 times as deadly as seasonal influenza. Evidence coming in from genomics and large-scale testing of fevers is consistent with these conclusions. The only potentially good news is that the epidemic in Korea may ultimately show a lower CFR than the epidemic in China.
It does no such thing. What you're doing is equivalent to me surveying people who are on their deathbeds to see if they currently have an ingrown toenail, then declaring that 95% of people with an ingrown toenail die within 24 hours.
Neither the numerator nor the denominator are what you're using them as.
- The numerator is neither A) the number of people who are dying of the disease, nor is it B) the number of people who have the disease and will die of it. It's the number of people who died who also happened to have a positive confirmation. This is relatively close to (A), but it's actually (B) you want.
- The denominator is not the number of people who got infected, and nor is it anywhere close to the number of people who got infected, and indeed is likely at least an order of magnitude off from the number of people who got infected. Your denominator is the number of people who tested positive. Which requires actually being tested. Most people who get the disease aren't tested. But most people who get very sick are.
I strongly advise people to stop playing amateur epidemiologist and calculating their own "IFRs", and to instead use the actual IFRs in peer-reviewed papers cited by WHO and the CDC.
Yes, that gives low numbers in a rapidly increasing outbreak. Is that what you are saying
High numbers, not low numbers. Because of the far higher rate of undiagnosed cases, since testing is biased toward severe cases when you're far from a universal-testing scenario (and we are very far from a universal-testing scenario).
No dispute - CFR and IFR are higher than a typical influenza, which is why this disease should not be ignored**. Although the last line should not just be a little one-liner. It's not just "a lower CFR"; it's an order of magnitude lower outside of Wuhan's early days. And indeed, Wuhan at present is itself an order of magnitude lower than Wuhan in its early days.Those early Wuhan numbers have biased up the global numbers ever since.
** - "Not being ignored" != "panicking, hoarding, and bad math", of course
Karen we are going in circles so respectfully I disagree and combined with the on the ground reports I do believe the numbers are meaningful. Honestly no interest in a back and forth further, you’re hung up on the fact that many cases go undetected and straight missing the forest for the trees. It is a fact Lombardy hospitals are struggling to treat everyone, we know with a more spaced out curve there more people would have survived.
Jason Van Schoor on Twitter
Jason Van Schoor on Twitter
Point from Wuhan and Lombardy. The hospital staff matter
Once the hospital staff get sick, there is a self reinforcing downward spiral.
If a region can avoid that tipping point, then this covid19 is over hyped.
But if that tipping point is exceeded - ouch. because Lombardy hospitals were really really good.
what about the non Covid 19 fatalites due to delayed hospital services? missing stat.
Once the hospital staff get sick, there is a self reinforcing downward spiral.
If a region can avoid that tipping point, then this covid19 is over hyped.
But if that tipping point is exceeded - ouch. because Lombardy hospitals were really really good.
what about the non Covid 19 fatalites due to delayed hospital services? missing stat.
It's not just "a lower CFR"; it's an order of magnitude lower outside of Wuhan's early days. And indeed, Wuhan at present is itself an order of magnitude lower than Wuhan in its early days.Those early Wuhan numbers have biased up the global numbers ever since.
** - "Not being ignored" != "panicking, hoarding, and bad math", of course
And herein lies the problem with trying to apply a statistic from one country to another. So many factors unique to each country that can affect the numbers. IFR, CFR, etc is more accurate when used within borders.
Here is the latest update: Coronavirus Update (Live): 114,422 Cases and 4,027 Deaths from COVID-19 Wuhan China Virus Outbreak - Worldometer
The picture is the update with the new reportings in the last 24 hours Europe time.
The picture is the update with the new reportings in the last 24 hours Europe time.
Or to put it in perspective, Italy is probably having more deaths from Covid19 per day then car accidents now.
It's actually 1/30th of that total that is the US amount, not 1/3rd. Roughly 100 per day. Just a bit different!
Hmmm, 0.3% (lowest CFR I've seen) x 50% infection rate x 7.8 B world population = 1,170,000.
I can see these numbers tweaked upwards or downwards but the impact on our economy is real.
Every TSLA stockholder realizes our values are based on future earnings. This works both ways. If stocks sense a lower future earnings...
I haven’t seen much explanation for why the countries where Coronavirus first hit are seeing fewer and fewer cases now. China is reporting less than a hundred new cases a day, down from thousands a day. This is a country of 1.4 billion people. 80,000 Chinese have tested positive for corona virus. But that means only one out of approximately 20,000 Chinese have tested positive. Why not more?
Deaths peak about 10d-2wks after cases peak, but then show similar decline.
Similar findings in South Korea. Fewer new cases every day.
If other countries follow this model (approximately 3 weeks from first death reported to peak deaths), then Italy could get worse for another week, US for 2 more weeks, before seeing a reduction.
Coronavirus Update (Live): 114,422 Cases and 4,027 Deaths from COVID-19 Wuhan China Virus Outbreak - Worldometer
Deaths peak about 10d-2wks after cases peak, but then show similar decline.
Similar findings in South Korea. Fewer new cases every day.
If other countries follow this model (approximately 3 weeks from first death reported to peak deaths), then Italy could get worse for another week, US for 2 more weeks, before seeing a reduction.
Coronavirus Update (Live): 114,422 Cases and 4,027 Deaths from COVID-19 Wuhan China Virus Outbreak - Worldometer
The answer to that question that you pose is widely available for example in the WHO report on coronavirus in China. The answer is that China implemented a rigid program of contact tracing, quarantine, testing, and forced isolation. At scale, and with no squish room. It isn't because the virus magically tapered off.
Although they blew the initial response, once it became clear that they had an epidemic on their hands they listened to the epidemiologists and ID specialists. Please read the WHO report. No other country has implemented anything like what the Chinese did. That's why all the other countries are still in the so-called exponential phase of transmission, with the possible exception of South Korea. Chinese non-pharmacological methods of epidemiologically informed mitigation, isolation and quarantine are the only proven techniques shown to break the human transmission chains. Read up and learn here. See detailed discussion here
Last edited:
So can we talk about the coronavirus in the thread that was specifically carved out and walled off from the general thread in order to not offend the delicate sensibilities of posters in the general thread?
Here, I'll go first.
Seattle area official outlines potential next steps in coronavirus response, including cancellation of events
Looks like they are getting ready to raise the response level in Seattle and possibly elsewhere in Washington state. Why we aren't at Level 5 right now, before we become the next Italy, is somewhat baffling to me. There will be thousands of cases next week and the local authorities are still dragging their heels in the emergency response here.
Here, I'll go first.
Seattle area official outlines potential next steps in coronavirus response, including cancellation of events
Looks like they are getting ready to raise the response level in Seattle and possibly elsewhere in Washington state. Why we aren't at Level 5 right now, before we become the next Italy, is somewhat baffling to me. There will be thousands of cases next week and the local authorities are still dragging their heels in the emergency response here.
All of Italy is now quarantined. Yes, the entire country is now under quarantine.
All of Italy placed under coronavirus lockdown
All of Italy placed under coronavirus lockdown
Doggydogworld
Active Member
While your points are basically valid, you make strong statements which are untrue. Crude deaths/cases math absolutely can underestimate CFR in the early days of an outbreak. To take an extreme case, I hope it's obvious that crude fatality rate for a 100% fatal disease will always be less than 100% until the epidemic is over.
You can create simple models with different CFRs, doubling rates, mild case ratios and incubation/progression times and come up with all kinds of scenarios in which "single point in time" death/case math underestimates CFR. The 0.7 (formerly 0.6) crude fatality rate quoted for Korea, where testing is extensive but many infections are quite recent, is very likely to rise over time.
You've also argued strongly against pre-symptomatic transmission. This paper tested samples from day 1 of symptom onset for 8 mild-to-moderate cases. Among the conclusions:
Critically, the majority of patients in the present study seemed to be already beyond their shedding peak in upper respiratory tract samples when first tested.
Note that these patients were being monitored due to a known exposure. That's why they were tested at the first sign of symptoms. It seems likely they were already contagious by then. Furthermore, initial symptoms are often mild. These Day 1 infectees in "Panic is Dumb" regions will often continue to circulate.
Similar threads
