Is Coronavirus less fatal than early predictions suggested? Two Stanford medical professors suggest that current mortality estimates are way too high.

American Thinker ^ | 03/26/2020 | Andrea Widburg

[hal ogen]

So maybe our society was punked?

[Tolerance Sucks Rocks]

Good. At the risk of elating the Branch Covidian FluBros, I say, LET’S CELEBRATE THIS NEWS!!!

[elbook]

All depends which of the ‘predictions’ you give any credence to.

We have plenty of attention whores competing as chicken littles who the leftist media are very happy to amplify these days.

[GOPJ]

Think of the virus as an inferior Chinese product... lots of great intention and BS - not much compared to American, Russian or German bio-weapons...

[GOPJ]

Think of the virus as an inferior Chinese product... lots of great intention and BS - not much compared to American, Russian or German bio-weapons...

[SeekAndFind]

RE: All depends which of the ‘predictions’ you give any credence to.

This is not a very helpful statement. I want to give credence to predictions that are closest to what is occurring in the real world. The question is this -— which of these predictions most closely tracks the real world?

[Svartalfiar]

Here's putting your link in context.

US total cases: 85,594
US deaths: 1300
US recoveries: 1868

This gives the following stats:
Current mortality rate: 41%
Best-case CFR: 1.52%

If we take 20% as the number of undiscovered infections, that brings the numbers down to:
Current mortality: 34.2%
Best-case CFR: 1.27%

[Svartalfiar]

Over 4 months in, with over 2 months of not much intervention at the get-go, worldwide cases are being reported as under 600K....have to multiply those world numbers by 50 to approach the number of Flu cases during an average/mild Flu season....in America alone.

You do realize the flu gets even more wild-ass guessing than the Wuhan virus? So far this year, there has been about 250M flu cases in these US. Which CDC then inflates to 7MM. Nearly a 30x increase from the actual reported numbers.

[kabar]

The 85,594 are confirmed cases. We don't have enough data to make an informed estimate of the number of asymptomatic and unreported cases.

How did you arrive at a 41% mortality rate?

At present, it is tempting to estimate the case fatality rate by dividing the number of known deaths by the number of confirmed cases. The resulting number, however, does not represent the true case fatality rate and might be off by orders of magnitude [...]

A precise estimate of the case fatality rate is therefore impossible at present.

[trebb]

And?

With the Flu, they look at Flu-like symptoms and many cases go unreported so the actual numbers of those infected with the Flu are logically higher than reported - but you're nuts if you think we've had 250M cases of the Flu in the US this year (and your math doesn't add up so you may have put in an extra zero) so I guess should stop here.....but

If you watch the Task Force briefings, you get some good info...right now, the original, really bad predictive models/numbers aren't adding up....by a long shot. The only way for some numbers to even approach some of the models (one model predicted 500K deaths in USA alone and had to be modified to a possible 20K) is if there are massive numbers of truly asymptomatic cases running around and now they are searching to see if those cases even exist and, if so, what the actual numbers may be....because none of the predictive models was worth beans and only served to cause panic...now, as true data comes in, the virus can still be deadly, but the picture is way rosier than it used to seem.

[Svartalfiar]

The 85,594 are confirmed cases. We don't have enough data to make an informed estimate of the number of asymptomatic and unreported cases.

True, which is why I included the second set of numbers accounting for 20% of cases going undiscovered. The China study can't be trusted, but even that revised their estimate from 86% down to 35%. A study on the cruise ship puts the number at 17.9%. Another study I saw said I think 14-16%. So I figured 20% is a decent, somewhat accurate number to use.


How did you arrive at a 41% mortality rate?

Math. 1300 deaths and 1868 recoveries mean you have 3168 cases with a known outcome. Out of those, 1300 deaths is 41%. That number will come down as more cases resolve, since recoveries lag deaths, and it's pretty obvious based on other countries farther along that the ultimate CFR is much lower, somewhere between 1% and 5%. Your other method of estimating CFR is to take current deaths and compare that to total cases at a given time ago equal to the length of the infection resolution time, but we don't know what that is, so it's hard to be accurate. I've seen medical estimates anywhere from ten days to nearly a month (28 days) for full case resolution. Obviously those two numbers will give a HUGE range of possible results.

[Svartalfiar]

And?
With the Flu, they look at Flu-like symptoms and many cases go unreported so the actual numbers of those infected with the Flu are logically higher than reported - but you're nuts if you think we've had 250M cases of the Flu in the US this year (and your math doesn't add up so you may have put in an extra zero) so I guess should stop here.....but

And what?
Yes, of course the numbers are higher, but not 30x higher than confirmed numbers. Flu-like symptoms can be a range of not-flu causes, including the common cold, respiratory infections, mono, meningitis, bronchitis, strep, RSV, etc etc. But many times, docs just assume it's the flu because many of these get the same treatment, so why bother testing for influenza? Real flu numbers are certainly higher than the confirmed cases, but nowhere near as high as the CDC claims they are. Same thing with "flu" deaths - even the confirmed ones are often flu+pneumonia, so the patient likely dies from the pneumonia cause, not from the flu. So not only do you have the same random inflation as above, but even the confirmed number is likely lower than it's claimed. Actual deaths than can be definitively determined to be caused by influenza are a couple hundred to a few thousand each year.

If you follow the link in this other thread, you'll see that the two big influenza reporters (clinical labs and public health labs) report a total of 280M positive specimens for influenzas. (No extra zeros!) Out of a total of 1.3MM total tested. So for every case thought to be flu and tested for it, only 21.5% of those were even positive. So yes I do agree with you that there's been more than 250M flu cases this year, that's just the bottom number based on confirmed tests. But the real number is certainly not anywhere close to the 39MM that CDC is claiming. That is a 140x guesstimate of the provable numbers!


If you watch the Task Force briefings, you get some good info...right now, the original, really bad predictive models/numbers aren't adding up....by a long shot. The only way for some numbers to even approach some of the models (one model predicted 500K deaths in USA alone and had to be modified to a possible 20K) is if there are massive numbers of truly asymptomatic cases running around and now they are searching to see if those cases even exist and, if so, what the actual numbers may be....because none of the predictive models was worth beans and only served to cause panic...now, as true data comes in, the virus can still be deadly, but the picture is way rosier than it used to seem.

A couple studies I've seen place the number of hidden infections at less than 20% of the total case number. The Chinese study that claimed 86% was (besides being unreliable coming from China) based on China going full-deny/coverup mode. Once they switched to quarantine/incinerate mode, that study's authors dropped their estimate to 35%. The Diamond Princess study estimated unknown infections to be 17.9%. Another study I've seen drops it to 14-16%. So 20% is a higher-end, but likely close, estimate for unknown infections; which is good and bad. (While high unknown infections mean more people with it and more getting exposed, it also means the CFR is much lower than confirmed cases will calculate to. A low percentage running around means less spread/better containment, but the CFR estimates of 1-3% are more accurate, and that's a lot of people at the rate this thing spreads.)

Hopefully though, between the quinine treatments and some of the other stuff being tried, we've at least found an effective treatment that will bring the mortality rate to a crashing halt. All the higher estimates of 1-5% are based on nothing but ventilators with almost no drug support. Once effective drugs are added in, we should see 50%, 80%, maybe even 95% of 'death' cases recover instead.

[Svartalfiar]

I forgot to mention, just to be thorough, that you can’t take deaths divided by total cases unless your outbreak is completely over. I mean, you can, but that only gives you an absolute best-case CFR, because it assumes that every single currently sick person will recover, and your deaths will stay at the number they’re at. Especially when your caseload is rising quickly, that will inflate your death rate downward, because the most recent cases are the group with the most deaths yet to come which that formula is counting as a recovery. While groups further along in the disease will have already had most of the deaths occurred and accounted for (so most still sick will recover).

[Yaelle]

It took me a while but your stats are very interesting. Certainly this virus is more contagious and more virulent than a typical flu. To believe otherwise is to basically distrust everyone in the world. We all (especially our educated) can tell this is different.

It won’t always be this way. AIDS killed savagely but we know how to stop that now. This will also fade into the many viruses we now can treat easily. And the already sick and infirm will always be “hardest hit” including by a typical light cold.

[kabar]

True, which is why I included the second set of numbers accounting for 20% of cases going undiscovered. The China study can't be trusted, but even that revised their estimate from 86% down to 35%. A study on the cruise ship puts the number at 17.9%. Another study I saw said I think 14-16%. So I figured 20% is a decent, somewhat accurate number to use.

Disagree. When you look at the CDC numbers for the generic flu, they use a sample of 1.4 million tests combining data from public health and clinical labs with about 20% testing positive similar to the 14% we are finding with the Wuhan virus. Based on the data for the generic flu, CDC estimates that an additional 38 million are positive but unreported thus arriving at their .1% mortality rate.

20% is low. Based on the latest data from NY 44,635 tested positive and 101,118 negative, which is an infection rate of 31%. The unreported cases must easily be much greater.

[utahb52]

“What we have is proof that some kind of militarized CDC is going to be necessary going forward, and that measurement methods & counter techniques are going to have to be essentially ‘invented’ in order to provide reliable information/action to a command & control organization on a real-time basis.”

https://en.wikipedia.org/w/index.php?title=USAMRIID&redirect=no

[Svartalfiar]

Disagree. When you look at the CDC numbers for the generic flu, they use a sample of 1.4 million tests combining data from public health and clinical labs with about 20% testing positive similar to the 14% we are finding with the Wuhan virus. Based on the data for the generic flu, CDC estimates that an additional 38 million are positive but unreported thus arriving at their .1% mortality rate.

But that math isn't looking at quite the same thing. The 20% influenza-positive is from cases that exhibited flu-like symptoms, and were tested. There's large numbers of people with flu-like symptoms that aren't tested. Nor people that don't exhibit symptoms. All that math says is that of people who thought they had flu and were actually tested, 80% are NOT the flu. So really, if you extrapolate that out, if 39MM people think they have the flu, then it's probably only 8MM that are actual influenza cases. Likewise, flu deaths are greatly inflated as well, with only a couple hundred to a few thousand being confirmed to have influenza, and even then, those are "flu-related", not "flu-caused". Most of the deaths are due to pneumonia, which can have a huge variety of causes, and isn't necessarily the flu.

Positives out of suspected cases is different than positives out of unsuspected cases.


20% is low. Based on the latest data from NY 44,635 tested positive and 101,118 negative, which is an infection rate of 31%. The unreported cases must easily be much greater.

Except here we're looking at numbers as the caseload is on the upswing. There's definitely a lot of untested positives out there, but many of those simply haven't hit the symptom stage yet. They will present, at which time they won't count as an undiscovered carrier. The undetected infection rate will drop quite a bit as positives have symptoms and get tested. But of those tested in NY, are those random people, or are they people exhibiting symptoms that would lead medical staff to think they have the Wuhan virus? The issue with this virus that makes things difficult is the asymptomatic spread. The only difference between these hidden carriers and someone who ends up in the hospital is just time. Assuming those tests were all of sick people, that NY math just tells us the same as above: that of people with the symptoms, 70% do NOT have the virus.

Again, there is a difference between undiscovered carriers in general, and positives for people exhibiting Wuhan-like symptoms.

[kabar]

if 39MM people think they have the flu, then it's probably only 8MM that are actual influenza cases.

The 39 million number comes from the CDC. This is their estimate of those who they think had the flu.