"... at the time the samples were taken. 'Our findings suggest that there is somewhere between 50- and 80-fold more infections in our county than what’s known by the number of cases than are reported by our department of public health,' Dr. Eran Bendavid, the associate professor of medicine at Stanford University who led the study.... The initial data is the first to provide greater clarity about where a community is in the pandemic."
From "Antibody research indicates coronavirus may be far more widespread than known/Of 3,300 people in California county up to 4% found to have been infected" (ABC News).
याची सदस्यत्व घ्या:
टिप्पणी पोस्ट करा (Atom)
पोस्ट्स
१२३ टिप्पण्या:
We have been consistently told by the President that there are plenty of tests.
Yet this makes pretty clear the extent to where February's denials have left us without the extensive testing that would give us good data.
Now. after being reassured the Feds are doing all they can, this week placed all responsibility on Governors.
What leadership!
Are they saying Californians all eat bats at wet markets. One more way to attract tourists to San Francisco.
this report is quite the ink blot
When you read, that cases are EIGHTY FIVE TIMES HIGHER THAN PREVIOUSLY THOUGHT...
Do that make you think
A) WOW! This shows we Need More Shutdown!!
or
B) WOW! This shows we Didn't Need a Shutdown AT ALL
????
Twice as many tests have been performed in the US as in any other country. Our testing rate is in line with S. Korea's on a per million basis. Do you think Trump personally should be out sticking swaps in people's noses across the land?
I'm pretty tired of the complaints about 'leadership.' Could Hillary Clinton have conjured 330 million COVID-19 tests out of thin air in February? She couldn't respond to an emergency involving four Americans in Benghazi. How about Biden? Did he have some magic pipeline of testing kits available to him?
The pandemic is not about Trump. It's not even about Cuomo or DeBlasio.
Ha. I’ve seen both of those reactions in print.
The initial data is the first to provide greater clarity about where a community is in the pandemic.
Maybe San Miguel county was first.
"2,583 tests have been processed (population about 8,000)
17 were positive
43 were borderline
2,523 were negative"
The first 800 or so test were definitely not random, since the were done on "first responders" and their families living in Telluride.
Not sure about the rest of the test, but apparently they're done by request of the testee, so not random either; I didn't see the word "random" in the abcfakenews article.
The San Miguel county tests covered a far higher percentage of the population than the Santa Clara county tests: >25% vs .17% in SC county.
4.2 % of the population my have immunity. That’s not very many people, not nearly enough for herd immunity. At this point, the antibody tests are more valuable than the antigen tests. When will widespread antibody testing be made available to Americans? Wouldn’t that be considered vital to the reopening of the economy?
She blinded me with science! Science!
Pop Quiz... Do You Get FLU or is your immune system healthy?
Blogger Mark said...
We have been consistently told by the President that there are plenty of tests.
Yet this makes pretty clear the extent to where February's denials have left us without the extensive testing that would give us good data.
Now. after being reassured the Feds are doing all they can, this week placed all responsibility on Governors.
What leadership!
In an alternate universe, President Hillary would have swilled Chardonnay and shit 320 million test kits right out her ass at the first sign of a novel virus. Orange Man Bad!
I think cases are grossly undercounted, but it has to be pointed out that this study has a lot of uncertainty. The false positive rate on their test has enough error to account for many of their results. Also, their sample was not random, it was people recruited on the internet, many of whom might have sought testing because they thought they were infected but couldn't be tested any other way.
@ Inga - I think all Democrats should stay home until they are personally tested. Don't you agree?
Pop Quiz time... Do you get FLU or is your immune system up to, (pardon the expression), snuff.
this report is quite the ink blot
Ceteris paribus, what difference at this point does it make. The real sickos can be tested, the rest will never be sure, and it looks like acquired immunity isn't a sure thing anyway.
Onward.
A case of the mild flu, usually produces these kinds of results. Nothing new here.
When will widespread antibody testing be made available to Americans?
I read yesterday:
"Abbott is significantly scaling up its manufacturing for antibody testing and is expecting to immediately ship close to 1 million tests this week to U.S. customers, and will ship a total of 4 million tests in total for April. The company is ramping up to 20 million tests in the U.S. in June and beyond as it expands the tests to run on its new Alinity™ i system, Abbott said in its statement."
"The company described the test as “a lab-based serology blood test for the detection of the antibody, IgG, that identifies if a person has had the novel coronavirus (COVID-19).”"
Also the SM county tests are for active infections, rather than previous infection - I think.
“Inga - I think all Democrats should stay home until they are personally tested. Don't you agree?”
No. I think if people are so inclined they should go out as they see fit. Take your chances. Hopefully there won’t be such a huge spike as to overwhelm the healthcare system. If you’re inclined to go out and protest in large groups, make sure you yell really loudly. It would be interesting to see what happens, a good experiment.
Our 4 most populous states:
1.California: Population - 40 Million
No. of Corona deaths: 1,100
2:Texas Population: - 29 Million
No. of Corona deaths: 400
3. Florida Population 22 Million
No. of Corona deaths: 726
4. New York .Population: 19 Million
No. of Corona deaths: 17,000
Anyone can see that Texas & Florida have extremely LOW numbers. Either the virus didn't spread there or it did spread and most symptoms were mild and folks got over it quickly. They should reopen immediately.
The most interesting comparison is Cali & NY. The Empire State has about half as many people as the Golden State, yet 17X the death counts. That disparity means 2 separate things are happening in the two states.
The Santa Clara study suggests California got it early and most folks had mild cases. That simple. No need for "herd immunity". Just mild cases.
As for NY, they are greatly inflating their counts. If you are hospitalized on death's door with, say, diabetes (83,000 deaths/ year) and either get the virus or are "presumed" to have the virus, they will count it as a Covid19 death to juice the numbers.
In other words NY is not distinguishing between folks who die WITH the virus as opposed to folks who die FROM the virus.
The term is Co-morbidities. So, if you dont have a comorbidity, i.e., you are young and healthy, well, you're risk is very low, and probably should return to work.
The great Revelation when all this is over is that the asymptomatic were highly communicable. That is what allowed this virus to spread like wildfire.
As for NY, they are greatly inflating their counts
It would be fun to see a list of all deaths for each state and compare those to last year
Of the 17,000 "covid-19 deaths" in New York how many of those people were going to die from something anyway?
Am i saying those deaths don't count? I'm saying why are you counting them HERE?
The Empire State has about half as many people as the Golden State, yet 17X the death counts.
A swollen head may prove to be another co-morbidity.
21st Century "Missile Crisis" if you are reading this... you have survived.
Many of these comments seem to be missing the point of these results. If (mind you, I said if) these results hold up, this is huge. Lethality 50 times lower means that the absolute worst possible case for the US: everyone (everyone!) gets the disease (this won't happen) and we don't find any better treatments or vaccines ever (also really unlikely), and that's 360,000 dead. That may sound like a lot, but Althouse did the same worst case calculation a couple weeks ago that I just did here and got 11 million dead. If this result holds up, we're done with this disease, it really is just a bad flu.
The data that we have has been pretty consistent since day 1. This whole dem panic is pretty much a hoax driven by the media and the “scientists” working at NIH and CDC. All we need to know are the results from the Grand Princess and the Diamond Princess. Everything else is hyperbole and bad models.
I posted a skeptical analysis of this last night. It can be found at MarginalRevolution.
If true this is good news because it means the IFR is a lot lower than previous estimates.
It might not be true, there are problems with the study, and it’s only one study.
Blogger gilbar said...
Of the 17,000 "covid-19 deaths" in New York how many of those people were going to die from something anyway?
Call me a pessismist, but I going with 100% mortality in the long run.
What this pandemic has taught me, is that a lot of people's brains die before their physical body expires.
If he is right and there are, say 50X as many cases as we know about, that means the case fatality rate and hospitalization rate are 1/50th of what we have been telling each other.
That would be GOOD news, that the viral infection is so rarely acute.
A gentle reminder for those concerned for our sisters and brothers to the North:
Canada Population: 38 Million
No. of Covid-19 deaths: 1,300 over 3 months.
This means the "curve" started low and flat, stayed low and flat in the middle, and ended low and flat.
Have a good weekend!
Testing can never tell us the economic and public health costs of a continued shutdown. It can help to isolate those who test positive. If you have a truly random sample of the specific population you want to learn something about, then testing can give you a snapshot of the state of things on the day you did the test. I don’t think governments have specific plans about how their actions might vary based on random test results.
So until you are at the stage of reopening, when you want to keep the actively infected from coming in to work, going to bars, etc., what’s the point of testing, except that “everyone” says it’s important?
Lethality 50 times lower means that the absolute worst possible case
Look, aside from other problems with this study, the idea of applying the “50 times” conclusion broadly across the US should be very clear. The only reason this county’s results for antibodies came in at such a high multiple of the reported cases is that they initially only tested low numbers. This is also true of many other locations in the US but it isn’t uniformly true. In order for the results based on the multiple to be applicable to the entire US would be if there had been a standard percentage of the population being tested everywhere.
One example makes this point very obvious. If NYC had 50 x the number of infections that they’ve identified that would be 50 x 127,000 or 6.35 million infected- that’s more than 75% of the population. And if you use the SC high end numbers you end up with more people infected in NYC than the entire population.
The point is simple- the “50 x” is specific to that county even if it is accurate, because it is dependent on the proportion of the population that had already been tested to identify cases of Covid-19. If you believe the study has useful conclusions then at least understand that only the numbers relating to the percentage of people testing positive (2.5-4%) can be applied to other communities.
Clusterfuck. The good news is we won't get close to the Althouse concern that the US could suffer 11 million deaths from the bat flu.
Yet this makes pretty clear the extent to where February's denials have left us without the extensive testing that would give us good data.
Are there two "Marks" commenting ? One sounds reasonable and the other sounds like a lefty nut. I've noticed this before.
BAG, the NY results are worse than you say. They are counting"probable" cases with NO testing as WuFlu deaths. About 4500 of them, I believe.
A gentle reminder for those concerned about world wide epidemics like the Black Plague and the Spanish Flu of 1918:
World Population: 8 Billion people
No. of world-wide Covid -19 deaths: 155,000
So, roughly, out of the 8 Billion people on Planet Earth, 60 Million people die each year from something. (A buncha cute babies are born too!)
So, the expected number of deaths over 3 months is about: 15 Million people
So, during the 3 month world-wide Corona pandemic of 2020:
1. 15 Million were expected to die
2. And 15,155,000 actually died (if one graciously accepts, that no double counting had occurred)
Looks pretty close to a 1% world wide excess mortality rate for the Kung Flu.
Not to be snarky, but during the Black Plague, did 99% of folks die from something else?
Yesterday I read about an interesting study. The CDC tested everyone at a Boston area homeless shelter. Results:
Number tested: 397
Number positive for COVID-19: 146 (36.8%)
Number with COVID-19 symptoms: Zero
Since homeless people are scarcely representative of the general population it is by no means clear what we can draw from this study. Should we sell our homes and live on the street? Abuse alcohol? Abuse meth? Maybe if you've lived for a while on the streets and you aren't dead yet your body has an unbelievably strong immune system? But the study does support the Santa Clara study's finding that there are a lot of people who have the disease but who are asymptomatic out there.
“ Not to be snarky, but during the Black Plague, did 99% of folks die from something else?”
In Europe the overall deaths from the plague were about 35%. The exact number is hard to ascertain. So, I doubt 99% of those who died died of something else. That would give an implausibly high death rate: it's rare for each person to die 30 to 40 times in a two year period.
If you’re inclined to go out and protest in large groups, make sure you yell really loudly. It would be interesting to see what happens, a good experiment.
Inga, I haven't abused you for your IMHO over-caution, and you haven't commented on my prediction of 15K-60K deaths, but when you say this above, what are you hoping will happen? That such people will be shot by fascists, or swept up in butterfly nets and whisked off to isolation units, or beaten off with firehoses of disinfectant? That some bad thing will happen to enemies of the people who dare to raise their voices?
Your side has never ever minded making noise, indeed "let's make some noise until they hear us" seems to be a motto, and to be hinted at that you shouldn't would seem the depths of hellish repression to you, your cause, your freedoms.
There is a meme going around with a picture of the Founding Fathers debating the Constitution, with the caption added:
Of course you realize that none of this applies in the event of a virus
Is that how you see it? Inga, step back and look at this. Try to be objective.
So more cases, but less deaths is a good thing, right?
Hmmm - sounds suspiciously like what a number of us having been saying all along.
Yes, there are two Mark's commenting here — and have been for some time. Good that you finally noticed.
It's both good and bad. Good that a lot of people probably have an innate immunity to it. Bad because it is therefore highly infectious it's still rises to the level of pandemic. This might be the perfect warning signal for us because the next one may be just as infectious and we could be less resistant.
Bay Area Guy said...
Our 4 most populous states:
1.California: Population - 40 Million
No. of Corona deaths: 1,100
2:Texas Population: - 29 Million
No. of Corona deaths: 400
3. Florida Population 22 Million
No. of Corona deaths: 726
4. New York .Population: 19 Million
No. of Corona deaths: 17,000
Anyone can see that Texas & Florida have extremely LOW numbers. Either the virus didn't spread there or it did spread and most symptoms were mild and folks got over it quickly. They should reopen immediately.
I'm not sure if I've raised this before, and if I have, perhaps not in this forum, but try this one on.
As infection is passed from host to host, isn't it vitiated? That is, A infects B infects C infects D, but each one in turn is less infectious? Maybe D can't infect E successfully or symptomatically?
If so, perhaps Alice was patient zero in New York and Bob or Carol or Doug was patient zero in California. Could that explain the difference in apparent virulence and lethality?
Mikko Paunio M.D., M.H.S., an adjunct professor in general epidemiology at the University of Helsinki, Department of Public Health, wrote,
"The World Health Organization has spread fear of Covid-19 without knowing the actual circulation rate of the virus. I calculated the SARS-COV-2 infection fatality rate (IFR) from antibody prevalence blood samples taken from donors in the capital region of Denmark in early April. According to my calculations the IFR is 0.13%, making the virus approximately as dangerous as the seasonal flu."
See his paper which was rejected for publication by MedRxiv and PLOS.
Hi Doc K,
"BAG, the NY results are worse than you say. They are counting"probable" cases with NO testing as WuFlu deaths. About 4500 of them, I believe."
Yeah, the NY numbers are atrociously counted. Nearly 80-90% of Covid-19 deaths in NYC have serious comorbidities.
Those comorbidities are. Diabetes, Lung Disease, Cancer, Immunodeficiency, Heart Disease, Hypertension, Asthma, Kidney Disease, and GI/Liver Disease. Source: ftnote 1 above
@CStanley I'm only talking about lethality, you're talking about transmission statistics. We know how many people died in Santa Clara county. IF the Stanford study is right, we now know what the case fatality rate is and it's a lot lower than the reported rates.
We discussed this research in the café thread. It has its limitations, which the authors claim to have adjusted for, but even with those limitations it seems more likely that the prevalence rate should be at least the 2.49%–4.16% they estimate, rather than lower, especially if the virus is anything like the super contagious monstrosity we’ve been led to believe. Moreover, community spread was confirmed in Santa Clara County by Feb. 28, 2 weeks before any shelter-in-place order, and long before testing had ramped up.
I’m going to do some reckless generalizing here, so caveat lector: I’m not a scientist, etc. I’m drawing on stuff I’ve previously written, but with some adjustments to the math to account for changes in available data. Also adding in a special feature about NYC, for the hell of it.
Now, Santa Clara County reported 5 new deaths yesterday, bringing the total count up to 74. Let’s say it’ll be 100 by the end of next week, to better account for the lag between infection and death, since this research is estimating prevalence in early April. That would suggest a death rate of 0.12%–0.2%, which is of on par with seasonal flu, if perhaps in a slightly meaner mood than average.
Let’s say 2.49%–4.16% of the US population was infected by early April, as the study concludes is likely for Santa Clara County. (Not enough for any potential herd immunity, unfortunately.) That would mean 8.2–13.8 million Americans were infected by early April. The total US death toll was 37,154 as of yesterday. To better account for the lag between infection and death, let’s push that number up to a generous 50,000 deaths by the end of next week (not an exponential increase, since we’re not experiencing that at present), accounting for deaths resulting from infections acquired in early April. That would be a death rate of 0.36%–0.61%. That’s bad, but it ain’t exactly the Spanish flu.
OK, so let’s extrapolate from the implied 0.12%–0.2% death rate for Santa Clara County. A US death toll of 50,000 by the end of next week would indicate 25–42.7 million Americans were infected by early April, which is around 7.6%–12.9% of the population. (Still too low for any potential herd immunity nationally, alas.)
Are we scared yet?
NYC has ~12,200 deaths (well, they may be fudging things upward, but accepting for the sake of argument), so unless every New Yorker has been infected, they have a higher real fatality rate than 0.12%. If the fatality rate were 0.2%, about 6,000,000 were infected by early April, 68.4% of the population. Maybe not impossible, but it does seem a tad high.
NYC’s true fatality rate could, perhaps more reasonably, fall in the 0.36%–0.61% range. With a population of 8.77 million, that would mean total infections ranging from to 2 to 3.4 million, or 22.8%–39% of the city’s population. That might be enough for herd immunity, if herd immunity is possible.
In any case, the virus would seem to be both less contagious and less deadly than initially feared. That’s good news, if true.
I expect that this entire episode will end up being seen as an unscientific overreaction, driven by fear, bad data, and bad models. Yes, people have and will really die, but the shutdown will not turn out to have been justified by good science. We will come to find that it was mis-classifications and that data fed into bad models that led to bad policy. We will also learn that we were not prepared for even a mild pandemic if it did happen, and that we ignored counter evidence and alternative voices too much, and were too quick and single-minded in adopting just one theory and response.
Kyjo. Nice post. You make some very good points and reasonable conclusions. Maybe what's going on in NY and Italy is the viral loading is the limiting factor on creating infections that show symptoms and eventually death for a high percentage of those with symptoms.
If that is true that means everyone having masks and washable gloves could possibly stop this thing and its tracks while the country goes back to business.
1000 words https://twitter.com/EWErickson/status/1251113801384185858/photo/1
Oso Negro said...
Call me a pessismist, but I going with 100% mortality in the long run.
sorry, black bear; thought it was OBVIOUS i meant would have died BY NOW, anyway
I keep forgetting that you're NOT smarter than the average bear
Bagoh2o: I agree 100% with your last sentence.
"Antibody research indicates coronavirus may be far more widespread than known/Of 3,300 people in California county up to 4% found to have been infected" (ABC News)."
But of course, it was always clear that what was "known" was bound to be very wrong. Therefore, what was "known," including the obvious risk profile everywhere Chinese Lung AIDS spread, was no cause for panic.
Well, I can admit when I was wrong, and I was wrong about this. I have thought (and said here a couple of times I think) that I thought this was going to be the equivalent of a severe flu. I was wrong; it's not even that.
Most of my skepticism arose from being in public/global health and epidemiology for the last 30 years. At best, epi is a very weak science, and it is politicized, regularly. They ("we", I guess, but I'll admit it) simply don't know nearly as much as people think they/we do. Relying on them to determine whether or not to shut down a quarter of the economy was ludicrous from the start.
That said, I wonder how this will play out in the future. A lot of people are figuring out that the "experts" were seriously wrong. I expect a myth will develop and be promulgated in the media/elitist circles that they really were right and all the shutdowns and social distancing and heavy-handed government was the only thing that kept it from being as bad as predicted. One wonders if we as a whole will actually learn anything from it, or just have it all go right in one ear and out the ass as usual.
Politics and a very poor media will probably prevent the general public from learning the real lessons from this. Highly complicit in that will be "experts" and their academic and bureaucratic colleagues more interested in politics or covering their ass than getting out the truth. We pay an enormous price for letting one political persuasion take over academia and the government bureaucracy. They just spent 3 years trying to overthrow an election, and they will never stop trying to control every single thing.
Whoopsie Woo:
IInfluential Covid-19 model uses flawed methods and shouldn’t guide U.S. policies, critics say
“It’s not a model that most of us in the infectious disease epidemiology field think is well suited” to projecting Covid-19 deaths, epidemiologist Marc Lipsitch of the Harvard T.H. Chan School of Public Health told reporters this week, referring to projections by the Institute for Health Metrics and Evaluation at the University of Washington.
Others experts, including some colleagues of the model-makers, are even harsher. “That the IHME model keeps changing is evidence of its lack of reliability as a predictive tool,” said epidemiologist Ruth Etzioni of the Fred Hutchinson Cancer Center, who has served on a search committee for IHME. “That it is being used for policy decisions and its results interpreted wrongly is a travesty unfolding before our eyes.”
I was actually going to work for IHME a few years ago but decided the commute was too long.
@Anthony, I’ve honestly been a little shocked to realize how many epidemiologists are really activists with credentials in epidemiology. When they start running their mouths off about the need for Medicare for All, it becomes obvious. It didn’t matter that the US had more critical care capacity per capita than anyone else, it didn’t matter that we had among the lowest average hospital occupancy rates, and it didn’t matter that we had more ventilators. Italy, with its high WHO ranking, was collapsing, and without universal healthcare, the US was likely to be in even worse shape. I guess I shouldn’t have been surprised
Kyjo: "That would suggest a death rate of 0.12%–0.2%"
Not arguing with you, but: a better measure of WuFlu impact would be QALYs lost, as argued by yours truly from the outset.
When life expectancy and QALYs are factored in, death rate among young people should be considered about ten times as bad on average compared to seniors -- assuming a loss of 50 QALYs for kids under 18 dying, and (generously) a loss of five for obese, sick, old Wuhan victims over 75. So a QALY-adjusted death rate for WuFlu would look lower relative to death rates for flu and pneumonia, which kill more young people. Rough example: 600 kids dead from flu in 2017-8 means, 600 x 50 = 30K QALYs lost, or about the current equivalent of 6K, 6000 x 5 = 30K, seniors dying from WuFlu now.
The rate of false positives might match the deduced infection level.
Look at this:
https://www.bostonglobe.com/2020/04/17/business/nearly-third-200-blood-samples-taken-chelsea-show-exposure-coronavirus/
The epidemic is raging in Boston. But where are the bodies? Why is Boston so different from New York City? Are New York hospitals killing people by the hundreds or even thousands as result of malpractice and unsanitary conditions. According to a BMJ article from 2016, "roughly 251,000 deaths occur every year as a result of medical error. If morbidity is factored in, this number skyrockets." The article can be summarized as: "Every physician will kill a patient every year." Is New York's healthcare system killing thousands of people?
@ Sebastian at 10:22 AM Please define term QALY means.
as with most of these events, 'they print the legend' as we now know, hi had a particular vector of infection, and the public health authorities in san Francisco, were adamant in adopting proper containment, except for one lone commissioner, that randy schilts noted,
of course the public health commissioner there, failed upwards to head the major lobby that deals with the disease, David Horowitz was one of the others who noted this failure, Michael Fumento, saw how this early sjw busywork was promulgated by fauci, specially after it got into the blood supply,
IMHE should get to work on a model that can somewhat predict what happened last week.
@Sebastian, I don’t necessarily disagree, but I’m not confident concerning the validity of adjusting for quality of life. I do think life expectancy is a valid consideration, though, and for that reason it seems rather obvious that a disease which favors killing old, sick people is not as bad as a disease which kills young, healthy people at a substantial rate, other things being equal.
Anthony at 10:05 AM. Great comment.
"They ("we", I guess, but I'll admit it) simply don't know nearly as much as people think they/we do." I believe this applies well beyond epidemiology. Probably to almost everything.
There will be histories written of this episode. I suspect it will be impossible to know which ones really tell the true story and which are bent through the lens of bias. Much like historical fiction about the Civil War "if Lee had won at Gettysburg".
The critiques of this study about the false-positives are solid. You can address this in one specific way- you retest the positives (ideally, you retest the entire sample) 2 or times. True false positives will turn up negative on a retest most of the time, and all but certain to turn up negative on two retests. This, of course, assumes false positives are truly random.
One caveat to this is that the unknown false positives (someone testing positive for COVID-19 antibodies who was never exposed to the virus) just have antibodies against some other virus that triggers a positive result. However, the Stanford study did address this in one way- they had truly 371 truly negative samples, and 369 of those did come back negative. This suggest that the population of people with such non-specific results to the test is already very, very low and not really a problem if the sample size is expanded, and if the sample is more likely to contain true positives.
In othe words, take this test and method to NYC is the next step for the Stanford team. There will definitely get a population sample that starts to overcome the low incidence problem that makes false positives an issue in the first place.
This study, though, doesn't stand all by itself as an outlier- all of the evidence suggests that the true fatality rate for this disease is under 1%, and all but certain to be under 0.5%.
There will be a huge battle to keep studies like this from getting published and disseminated because there is an enormous amount of political capital now invested in justifying the actions that have already cost at least $10 trillion dollars in the US alone. If the fatality rate comes to be known to be less than 0.5%, there is no justification whatsoever for the actions of the government at all levels- literally none.
(I should also apply the caveat that most of the researchers I work with -- as are, I think, most -- are conscientious and really do good work within their narrow purview, and acknowledge the limitations, biases, etc. Most of the heavy politicking comes at higher levels where funding and policy decisions are made)
(I also don't fault IHME all that much; I'd hate for my little model of temperature-related excess deaths to be the basis for far-reaching public policy)
And when I say low incidence problem- this is what I mean:
Suppose you have a test that gives 1% false positives as a known error rate. For every 100 truly negative people you test, you get 1 positive result. If the thing you are testing for really only has 2 out of 100 to be found, then half of your positive results could be the result of the random error. This is less of a problem if the thing you are looking for as an incidence of 10% or more. In the case of the Stanford study, the thing they are looking for reads back at about 3%, but if the error rate is in the range of 3%, then they have a problem. It is a smaller problem if the result they find comes back at 15%, and smaller yet if 30%. Additionally, though, the test, being new, can have its error rates more closely measured (something they did try to do and can claim some success).
Big Mike at 9:11 said
Yesterday I read about an interesting study. The CDC tested everyone at a Boston area homeless shelter. Results:
Number tested: 397
Number positive for COVID-19: 146 (36.8%)
Number with COVID-19 symptoms: Zero
By itself it really doesn't mean much but when added to the Santa Clara study and bits of information coming in we are getting a better picture. The Santa Clara study gives two ways to look at the numbers. One as a percentage if the population (2.5 to 4.2%) and the other as a multiplying factor i.e. total cases being (50 to 85 fold) the number of confirmed cases. In most places that I have heard of, to be tested you have to have been exposed or running a fever. If that criteria is reasonable consistent. Then you might calculate the total cases by a factor of 50, just to use the lower end of the Santa Clara numbers. If that is the case over 10 million New Yorkers have been infected and since most of those cases would be in NYC metro the disease may be about to burn itself out. If we see a sudden drop in new cases there then we are in the ballpark.
Which is why we need more studies.
"define term QALY means"
Quality-Adjusted Life Year.
I didn't invent it. I don't think it's the be all, end all. I do think All Lives Matter. But it is part of standard thinking about standard medical decision-making under (financial, resource) constraint. It's one sensible, not-extraordinary way to gauge how and where you should use your resources. It forces you to compare cases and think about trade-offs. It's the kind of thing we do, explicitly or implicitly, all the time. The fact that we, that is, the alarmists and alarmist public officials, gave up on normal thinking about risk and limits and trade-offs was one major feature of The Panic of 2020, with disastrous effect.
The Daily Star reports on study out of Cambridge suggest "the Rona" started in southern China as early as September.
@Sebastian Thank you
Oh and 10 million is a ridiculous figure now that I look at it again. Duh
"Mark" - We have been consistently told by the President that there are plenty of tests. Yet this makes pretty clear the extent to where February's denials have left us without the extensive testing that would give us good data.
Some people think that Trump should just pull things out of his ass with the same ease that they pull their partisan nonsense out of theirs.
Never mind that such things did not exist in large numbers anywhere on the planet and that there was no ability to do widespread administration of testing even if there were, much less getting lab results.
Are there two "Marks" commenting?
Michael K -- there is one "Mark" commenting and one Mark commenting. Only one deserves the quote marks.
There is Lefty Mark, and Star Trek Mark. Unsurprisingly, the latter is the sane one.
Just curious, for those who are highly skeptical about the NYC mortality rate reporting, what else could be contributing to the excess deaths there?
"Of the 17,000 "covid-19 deaths" in New York how many of those people were going to die from something anyway?"
Of the 17,000 "COVID-19" deaths, how many were actually COVID-19 at all?
We don't know. We'll likely never know. The books on this pandemic have been so thoroughly cooked that only widespread antibody testing will ever tell us even part of an accurate story.
"what else could be contributing to the excess deaths there?"
Do we know for a fact the number of "excess deaths"? I'd venture the improper use of ventilators may be hastening the demise for some folks.
Two more acts of malfeasance to lay at the feet of that damn Trump -
Washington Post headlines --
Contamination at CDC lab delayed virus tests’ rollout
and
Hundreds of nursing homes with virus cases have violated infection-control rules
Everything is Trump's fault.
The Santa Clara study projects death rates from corona virus infection
to range between 0.086- 0.4%. This is more evidence that the Great 2020 Pandemic is bit of a medical dud. Meanwhile, the over-reaction to covid-19 is an economic nuke.
The beneficiary of all this should be President Trump, who has proven an exceptional leader. His early touting of hydroxy-chloroquine (despite massive scorn and obloquy from the media) was brilliant and will be seen as one of many things he did (travel bans, etc.) to impede the epidemic in the US. The crowning success of his efforts is getting Fauci and Birx on-board with his early reopening of the economy.
“ Just curious, for those who are highly skeptical about the NYC mortality rate reporting, what else could be contributing to the excess deaths there?”
An excellent question, especially since at one point so many were in agreement that excess deaths is an important metric to measure what damage the epidemic has done.
what else could be contributing to the excess deaths there
What is the usual daily death rate in NY from years past? That would indicate if, in fact, there are "excess deaths."
But in any event, NY by its own admission is adding in presumed and possible COVID deaths without any delineation between the confirmed cases and the guessed cases.
And all the data up to April 14 was based one confirmed cases only.
When you change the rules in the middle of the game -- whatever the reason -- you are going to skew any determination of trends.
“But in any event, NY by its own admission is adding in presumed and possible COVID deaths without any delineation between the confirmed cases and the guessed cases.”
That’s not accurate, and it’s also not relevant to CStanley's question, which was about excess deaths.
The question is -- will the people who are wont to cook the numbers going to let us go over the hump?
"That’s not accurate, and it’s also not relevant to CStanley's question, which was about excess deaths."
Agree that it's not directly relevant to the excess deaths question, but it IS accurate. We know from news reports that NYC is assigning COVID as cause of death in suspected cases with no confirmed test result. We know that a person with severe comorbidities, for whom a case of the common cold would be lethal, are having their deaths pinned squarely on the WuhuFlu.
No, it’s not accurate. Look up the words “possible” and “probable”. It’s exactly the kind of sloppy deceptive shit that CNN does.
Is New York's healthcare system killing thousands of people??
And/or they are labeling unproven covid deaths as due to covid in order to get more fed money. Never let a crisis go to waste, especially when money is involved. You want more of something throw money at it. (I hate how cynical I've become...)
Looking bad in Japan right now. Hospitals are being overloaded.
One other small sample (from a Reuters article on 4/17) from the Theodore Roosevelt:
"Sweeping testing for the coronavirus among the entire crew of the Roosevelt has already yielded a curious result: The majority of the positive cases so far are among sailors who are asymptomatic".
I'm not sure why they label this "curious".
I estimate that about 7000 people per month die in the New York metropolitan area in a normal year (based on 2017 mortality data). There is definitely an excess over that in late March and April of this year, and I think it logical to assign most of that to COVID-19 in one form or another. What I would be interested in, though, is this- what happens to mortality over the next year? In other words, people who die of COVID-19 today, can't die of something else next Winter, next Spring, next Summer. I suspect what you are going to see is that COVID-19 increases mortality during the pandemic, but that most of the effect is going to be a shifting of mortality from the next year or so into mortality measured today.
It is actually possible to study this effect with actuarial tables- all you need is better information about the people who have died of COVID-19- enough data about their age, demographics, and co-morbidities can accurately tell us how many QALYs were lost to the virus.
"Looking bad in Japan right now. Hospitals are being overloaded."
And, yet, when I look at this page of data, Ken, I don't see how that statement is possible. Less than 100 people have died of COVID-19 in Japan in the last 10 days total, zero yesterday. Japan, as a whole, had fewer new cases yesterday than the state of Louisiana. Is this case of you lying again, mistaken again, or are you just believing some more media hype? Give us a link so we can assess the source of information for your statement since it doesn't really jibe with the information Japan itself is reporting about deaths and cases.
I estimate that about 7000 people per month die in the New York metropolitan area in a normal year (based on 2017 mortality data). There is definitely an excess over that in late March and April of this year, and I think it logical to assign most of that to COVID-19 in one form or another.
But you need to subtract out the confirmed cases from that usual death count.
The issue is the matter of questionable unconfirmed cases.
Explain how past deaths determine current admissions and we'll talk.
“ In March, there were 931 cases of ambulances getting rejected by more than five hospitals or driving around for 20 minutes or longer to reach an emergency room, up from 700 in March last year. In the first 11 days of April, that rose to 830, the Tokyo Fire Department said. Department official Hiroshi Tanoue said the number of cases surged largely because suspected coronavirus cases require isolation until test results arrive” AP
Ken B:
I found an article online from the BBC on the issues in Japan. It does look bad for Tokyo at the moment. Makes me think about the subway system and whether Japanese are among those that use masks frequently.
The rate of increase of new cases does not appear to be all that dramatic (worldometers) rising from ~5,000 to ~10,000 in about eight days.
Yancey Ward at 11:24am
There will be a huge battle to keep studies like this from getting published and disseminated because there is an enormous amount of political capital now invested in justifying the actions that have already cost at least $10 trillion dollars in the US alone. If the fatality rate comes to be known to be less than 0.5%, there is no justification whatsoever for the actions of the government at all levels- literally none.
I don't know, hindsight is 20 20 and all, but given that most policy makers went into this blind. Also the CCP lied which made the fog the government has acted in thicker than it needed to be. The ones getting killed will be those who delay opening things up as well as those who took things to far, like Governor Whitless in Michigan. A lot of authoritarians have been exposed, for what they are, which is a good thing.
So, hospitals were already "overwhelmed" in March of 2019- got it, Ken- you just listen to fear-mongering without any critical thought whatsoever.
Ken B. said,
"An excellent question, especially since at one point so many were in agreement that excess deaths is an important metric to measure what damage the epidemic has done."
Oh, I still it's an important metric in my eyes.
Unfortunately, it's not simple.
(1) Some deaths are being caused by the virus.
(2) Some deaths are being caused by the lockdown.
(3) Some lives are being saved by the lockdown (I'm talking about fewer traffic fatalities).
(4) Flu deaths and coronavirus deaths are closely related since most of the people being killed by one, have underlying health conditions, and they could have been killed by the other.
It's going to be difficult to accurately separate out the true reason that people died.
But if the overall death rate is about the same as last year, which seems to be what is happening, then I can't see how the lockdown was even remotely justified. It didn't really save lives.
The article that I read said that part of the problem in Japan is that the number of ICU beds is low compared to other countries.
AP apanese hospitals also lack ICUs, with only five per 100,000 people, compared to about 30 in Germany, 35 in the U.S. and 12 in Italy, said Osamu Nishida, head of the Japanese Society of Intensive Care Medicine.
also
Japan has been limiting testing for the coronavirus mainly because of rules requiring any patients to be hospitalized. Surging infections have prompted the Health Ministry to loosen those rules and move patients with milder symptoms to hotels to free up beds for those requiring more care.
emphasis added
That is just insane policy.
Seriously, Ken, did you not immediately go and see what Japan itself was reporting for deaths and new cases before making that statement? As for the story you gave us- did it ever occur to you that the hospitals are increasingly turning away ambulances to preserve capacity in anticipation of a flood of COVID-19 cases, rather than an actual present flood of them? I know my hospital just down the block from me cancelled all elective stuff a month ago in anticipation of a flood of cases that never arrived- the parking lot is ghost-town the last week as I suspect the nurses and staff are beginning to be laid off, and it wouldn't surprise me to find the hospitals here were trying to shift patients between one another in a kind of game to keep capacity open, though I suspect that has died down here as the flood never arrived in the first place.
Star Trek Mark knows the solution to the confusion, but he's too stubborn to do it.
Where did I say “overwhelmed”? I said overloaded. If you turn away patients because you lack capacity you are overloaded. That’s what the word means.
Retreat back to your fantasy world where 7500 is the death toll for the entire epidemic.
Mandrewa
But the question was the current excess death we are seeing in NY.
“ But if the overall death rate is about the same as last year, which seems to be what is happening, then I can't see how the lockdown was even remotely justified. It didn't really save lives.”
For what period, and only during lockdown? That would make no sense. There is a counterfactual here you need to consider. Here is a simple example. The town of X near Florence cut off all visits in 1348. When they reopened in 1351 it was found they had the same death rate as usual for that period. Do you conclude their precautions were for naught.
Overwhelmed/overloaded, yeah, Ken, that is a YUGE difference in meaning. You still didn't address the fact that the story admits the hospitals were "overloaded" in March of 2019, though, did you?
"Star Trek Mark knows the solution to the confusion, but he's too stubborn to do it."
I personally like the confusion. Some commenters are already on the same page as me - I see the name and I already know what to expect. And ones like ARM and Inga I also already know what to expect. But then I see "Mark" and have to wonder, is this going to be sane-Trekkie Mark or evil Mirror Universe Mark (with the goatee, natch).
Keeps me on my toes.
Not that the comments aren't valuable - even if you and I already agree on 80-90% of things, I still like reading other people's take on events and politics. But I don't expect to be shocked by them, because my take is probably pretty close (or diametrically opposed, depending on the commenter in question).
Ken B., I don't really think the Black Plague is relevant. In particular because I think we have so little information about the Black Plague, comparatively.
But to address excess deaths in New York:
If we look at the country as a whole, I don't think there are any excess deaths.
But if we look at the New York metro area, yes I think there are excess deaths. It seems certain that something unusual is going on in the New York metro area even if some of the deaths are being miscounted.
Unfortunately because we know deaths are being miscounted all we are going to be left with is this "excess mortality" in trying to figure out what happened later.
New York should be compared to Boston. It's a puzzle why the fatality rate is so low in Boston compared to New York, given that there are multiple lines of evidence now that a large part of the Boston population has already been infected with the virus: (a) the sewage study; (b) the study released the other day where 200 random people were tested and one third had antibodies to the virus; and (c) the other study released the other day where homeless people in Boston were tested and something like 40% had antibodies to the virus and yet no one from this population of homeless had apparently died.
So why are the consequences so different in Boston and New York?
Like many people I suspect that the New York subway system has played a major role, and unlike many people I have a hypothesis as to just why, see my speculation in an earlier thread about the initial dose and the CD147 receptor altering the odds of death.
Another comparative is the UK. The New York metro area seems to have been harder by the coronavirus than just about anyplace else in the world, but the UK isn't far behind.
Here's an article about excess deaths in the UK
Until recently the total deaths in the UK have run in parallel with previous years. It was only by weeks 13 and 14 of the epidemic that there were excess deaths.
But this doesn't mean they are going to end the year with excess deaths.
As the author points out:
"The average age of the people who’ve died from coronavirus in the UK so far is 79.5 and a majority of them have underlying health conditions."
And:
"Until coronavirus came along, the vast majority of respiratory deaths in the UK were recorded as due to bronchopneumonia, pneumonia, old age, etc. “We don’t really test for flu, or other seasonal infections,” he wrote. “If the patient has, say, cancer, motor neuron disease or another serious disease, this will be recorded as the cause of death, even if the final illness was a respiratory infection. This means UK certifications normally under-record deaths due to respiratory infections.” "
So in other words it is going to be difficult to compare this year with previous years as they aren't the same thing.
But we will still have excess deaths. Yet it is still possible that the UK will not end the year with excess deaths:
"This data, showing a sharp increase in the number of deaths in Week 14, strongly suggests that COVID-19 is more deadly than seasonal flu. The five-year average for Week 14 is 10,305, whereas the number of deaths in Week 14 in 2020 was 16,387, and will be larger still in Week 15. Against this, if Professor Spiegelhalter is right, many of the people who died of COVID-10 in Week 14 would have died later in 2020 anyway from another cause."
Mandrewa
I think NY shows what can happen if the virus spreads widely quickly. As does northern Italy. It’s incorrect to look at national numbers rather than regional ones, where the region is defined by the contagion. If New York State or California seceded from the Union the new national rates would tell us nothing. It’s not a pertinent calculation. Let’s annex India and bring our rates down?
The Black Plague example is a reductio of your argument. Unchanged death rates do not show precautions were unnecessary.
Of course any epidemic will crowd out other diseases. Will it do more than that? Excess deaths is how we know. So indeed we cannot know the full significance of the excess in NY yet. But neither are they meaningless.
It's all about viral load. Also in Boston you don't have to use public transportation if you don't want to good. Also Boston is where the smartest people in the country are and so everybody has adapted to the distancing very effectively and quickly. It must be that Republican Governor we have the Lord over us stupid over-educated expert elite libtards
Just like I said.
hey Ken. did you save my quote.
Also Boston is where the smartest people in the country are
I assume you refer to Harvard freshmen, as we know that they know more than Harvard seniors. I have suggested for years that Harvard collect four years' tuition and grant a diploma the day they are accepted.
The saying is that Harvard is extremely hard to get into and impossible not to graduate from. MIT has the smart undergrads. Both schools have top of the world level graduate programs. Boston is and Massachusetts has more colleges per capita than any other place in the world. Even the deplorables here are smart.
“ It's all about viral load.”
I do hope that is true, but it’s uncertain yet. However it is a good idea to assume it matters and try to reduce the viral load of each encounter. That means distance, and in some situations, masks.
Here's the current excess death data for the period of interest from the CDC. No need to argue in a vacuum. Look specifically at Table 5. I know what I think it implies, but draw your own conclusions.
Sorry, I stepped away after asking my question. On a side note, I can now attest that it’s worth the extra effort to melt and brown the butter before mixing it into chocolate chip cookie dough.
Regarding excess deaths in New York, there’s been a ten fold increase in the number of people dying at home. Maybe a fraction of those are people dying at home of heart attacks because they’re afraid to go to the hospital, but it stands to reason that most are Covid deaths, doesn’t it? When people first become ill and call their doctors, generally they’re told to stay home unless they experience severe breathing difficulty. They may or may not be told to get tested for flu, strep, and/or SARS2 during the time that they are fighting the illness at home. So if there’s a large number of poeple who are suddenly dying at home of pneumonia (the deaths at home are about 200 per day, I think) then it would be a serious undercount if these weren’t labeled as probable Covid deaths even though many of them didn’t have access to testing.
I think Yancy’s point about time shifting may be valid but we’ll have to wait and see if that proves true.
As for comorbidities, I feel like a lot of people are thinking that preexisting illness is necessarily extreme as in a person who’s already terminally ill. And I know there have been some anecdotal reports of hospice patients and such....but the reality is that the comorbities listed for Covid deaths are ridiculously common, and could range from someone who takes Losartan for high blood pressure all the way to someone who’s on a kidney transplant list. If you’re assuming it’s usually the latter group, think about the Diamond Princess. The population may have skewed older, but it’s a relatively healthy group of seniors who would have been taking a cruise. Have you ever heard of 13 people dropping dead from their underlying conditions all on one cruise ship? If we can accept that these people obviously died of Covid (even though their underlying health conditions may have predisposed them for having a more serious case than did other passengers) then why isn’t the same also true for the general population?
@ Ficta...that chart will be very useful but the data is not complete enough now.
The footnotes:
NOTE: Number of deaths reported in this table are the total number of deaths received and coded as of the date of analysis and do not represent all deaths that occurred in that period.
*Data during this period are incomplete because of the lag in time between when the death occurred and when the death certificate is completed, submitted to NCHS and processed for reporting purposes. This delay can range from 1 week to 8 weeks or more, depending on the jurisdiction, age, and cause of death.
And you can tell it’s lagging behind because the table only includes 13,130 Covid deaths.
CStanley: you make good points re: comorbidities. I'll point out another anecdotal case: David Lat. As with any virus, some folks may be more prone to getting severely ill due to their own unique biology.
Not that this is meaningful in any way, but when I lived in Alaska, the folks on cruise ships that came to visit were known colloquially as "nearly deads".
I live in Santa Clara County, also known as Silicon Valley. Compared to other places, we had more people traveling to China for business and family reasons and more people traveling everywhere, because they could afford it. So I'd expect we'd have a higher-than-average exposure rate. (On the flip side, few people use public transportation and density is low.)
Our hospitals canceled elective surgeries to prepare for a Covid surge that never happened. They've laid off nurses and other staff. I really think we need to let people get non-emergency medical care again. What counts is not diagnoses but people who require hospitalization. That number is nowhere near hospitals' capacity.
"the reality is that the comorbities listed for Covid deaths are ridiculously common"
Yeah, I think that some who are ranting about comorbities are relatively young people who think you're either vigorously healthy or on death's door.
About any "shortages" of equipment, beds and materiel, remember that a lot of this medical stuff is subject to certificate of need regulations imposed by the "experts," meaning that facilities were actually forbidden from having it without demonstrated need at that time and production was limited.
@Ficta,
Thanks for the prelim CDC numbers. Hard to digest from my phone, but will do when I can print them out.
This little nugget was interesting:
Deaths due to COVID-19 may be misclassified as pneumonia or influenza deaths in the absence of positive test results, and these conditions may appear on death certificates as a comorbid condition. Thus, increases in pneumonia or influenza deaths may be an indicator of excess COVID-19-related mortality.
Some geniuses on this blog claim that this is NOT the Flu -- with multiple exclamation points. But if one reads the early papers from China in Jan and Feb, they are describing the patients with flu-like symptoms suffering from pneumonia.
The test results may or may not differentiate between flu or Covid-19, but the symptoms clearly overlap. Without the test, you can't clinically tell the difference.
CStanley at 4:31 pm
but it stands to reason that most are Covid deaths, doesn’t it?
No it doesn't, your making all kinds of assumptions about how these poor souls are passing away without any data. I could just reasonably argue that they are all from cancer or suicide. Someone could suffocate their spouse with a pillow and then say. Hey they died of respiratory distress so it's covid. I know what, Jeffery Epstein didn't kill himself he died of Covid 19.
@Andy-But why would deaths from those other causes increase ten fold during this time period?
टिप्पणी पोस्ट करा