Undercounting COVID Deaths
The counting methodology is flawed and that's assuming honest reporting.
According to official measures, over 3 million people are infected and nearly 208,000 people are dead from COVID-19. There’s reason to believe the actual numbers are much higher.
FT (“Global coronavirus death toll could be 60% higher than reported“):
The death toll from coronavirus may be almost 60 per cent higher than reported in official counts, according to an FT analysis of overall fatalities during the pandemic in 14 countries.
Mortality statistics show 122,000 deaths in excess of normal levels across these locations, considerably higher than the 77,000 official Covid-19 deaths reported for the same places and time periods.
If the same level of under-reporting observed in these countries was happening worldwide, the global Covid-19 death toll would rise from the current official total of 201,000 to as high as 318,000.
To calculate excess deaths, the FT has compared deaths from all causes in the weeks of a location’s outbreak in March and April 2020 to the average for the same period between 2015 and 2019. The total of 122,000 amounts to a 50 per cent rise in overall mortality relative to the historical average for the locations studied.
In all the countries analysed except Denmark, excess deaths far outnumbered the official coronavirus death tolls. The accuracy of official death statistics from the virus is limited by how effectively a country is testing people to confirm cases. Some countries, including China, have retrospectively revised up their death tolls from the disease.
According to the FT analysis, overall deaths rose 60 per cent in Belgium, 51 per cent in Spain, 42 per cent in the Netherlands and 34 per cent in France during the pandemic compared with the same period in previous years.
Some of these deaths may be the result of causes other than Covid-19, as people avoid hospitals for other ailments. But excess mortality has risen most steeply in places suffering the worst Covid-19 outbreaks, suggesting most of these deaths are directly related to the virus rather than simply side-effects of lockdowns.
David Spiegelhalter, professor of the public understanding of risk at Cambridge university, said the daily counts in the UK, for instance, were “far too low” because they only accounted for hospital deaths.
“The only unbiased comparison you can make between different countries is by looking at all cause mortality . . . There are so many questions about the rise we’ve seen in death that have not got Covid on the death certificate, yet you feel are inevitably linked in some way to this epidemic.”
The extra deaths are most pronounced in urban areas with the worst virus outbreaks, and have completely overwhelmed reporting mechanisms in some. This is especially worrying for many emerging economies, where total excess mortality is orders of magnitude higher than official coronavirus fatalities.
Beyond all of this, there’s obviously very good reason to believe China, Iran, and other authoritarian states are deliberately underreporting their cases.
China, a country of over 1.4 billion people and the presumptive origin of the infection, claims a mere 82,830 cases and 4,633 deaths. Even if we believe the hype about how aggressively and quickly they began testing and extreme quarantining measures, it’s just about inconceivable that they’ve had half the cases and a third fewer deaths than Germany, population roughly 80 million which had a long window to prepare for the outbreak and has also been lauded for its response.
Was this post cut off in mid-sentence? [Just editing detritus. Fixed now. -jj]
It’s still mind-boggling how little we know. A regular commenter here has said repeatedly that the only sure statistic is the death count. No, even that number has been clearly suspect from the beginning. Deaths attributed to COVID-based complications, without identifying the virus as the underlying cause, is just one of many problems with the mortality numbers.
Meanwhile, some states “opening up,” based on gut feeling more than reliable data. We could have done all of this better. To our national shame, we did not.
I don’t think there’s common agreement about what the objectives should be. If the goal is to prevent the health care system from being overwhelmed, we’re succeeding pretty well. I suspect that would have been hard to sell but I see the “bending the curve” graphs with one axis labeled “Active Cases”, the other measuring elapse time, a horizontal line indicating system capacity as being sales tools.
If the objectives are to minimize total cases or deaths, we’ve flopped.
And ignore China. We don’t know whether they’ve had 83,000 cases or 830,000. We also don’t have a reliable definition of what their testing has consisted of. It could be as little as taking a temperature for all we know.
I have a lot more confidence in what South Korea has reported. If anyone has actually “bent the curve”, it has to be South Korea. Do we have the level of social cohesion and compliance as a country it would have taken taken to emulate what the South Koreans have done? Was it scaleable? I’m skeptical on all counts but we have to admit we haven’t covered ourselves in glory. Better leadership at the top would have helped.
It’s a minor nit but not sure why Iran keeps getting mentioned as a country that is underreporting its cases. As of today, it has report 91K cases and almost 6K deaths. Of its neighbors only Turkey reports similar numbers while the surrounding countries of Iraq, Afghanistan, Pakistan, Azerbaijan, and Turkmenistan are not in the same league as reporting. Turkmenistan apparently doesn’t report at all and has banned the word coronavirus.
It is almost as though someone once said it and it became a standard cant.
It is just a fundamental fact that due to the shortage of tests, we are under counting. One can speculate about the reasons for this (I pick, h) All of the Above) but there is no disputing that we don’t know the true #.
@Dave Schuler:
My understanding is that this has more or less been the tactical goal of “Flattening the Curve” in the US. And, while I haven’t kept up on the lit, this was how it was initially explained by a lot of the infectious disease experts.
You are correct that this is very different than “minimize total cases or deaths” and a much harder political sell (even if we had a competent chief messenger/executive).
James also wrote:
While the authoritarian states appear to be engaged in the most egregious under-reporting, I don’t think we can say they are the only ones doing this. Its pretty clear that there are lots of incentives for any government (local, state, or federal/central) to under-report.
The only question is how intentional said under-reporting is. I don’t think, even in the US, we have wide-spread agreement on which deaths count towards the tally and which ones don’t. That’s also before we get to testing limits.
@Dave Schuler:
Actually the gold standard for fighting the virus may no longer be SK, but Australia and New Zealand
And interestingly, one is ruled by a climate change denying conservative and the other a democratic socialist.
I think we’ll find out it all comes down to lack of testing. If you see the numbers for Mexico, the past four days read like: 2,100 new cases, zero new cases, 2,000 new cases, 835 new cases. That almost seems like a joke.
I’ve also been seeing reports of a high rate of false negatives in tests. That’s a very serious problem, but it wouldn’t account for undercounting COVID-19 deaths.
Many countries limit testing to those showing symptoms, too. This allows pre- and asymptomatic cases to spread the virus.
A cheap, reliable antibody test might help. If you’ve antibodies to the SARS-COV2, then you’ve been exposed. You then are tested for the virus and see what that shows.
@Sleeping Dog:
So here in America we’re basically fucked, then.
@Dave Schuler:
And this wins the award for “Understatement of The Year”
The current rightwing talking point is that we’re overcounting Covid deaths, calling everyone who dies with the diseases a Covid death, whether they died from a car crash or cancer.
This ignores, of course, what you highlighted in the post, about death rates being much higher than in previous years even above the official Covid death rates. The rightwingers are also claiming that any co-morbidities a patient had are the actual reason for death, even though many of those conditions, such as high blood pressure and diaabetes, are treatable and the individuals would likely have otherwise lived a long life were it not for Covid-19.
@Kingdaddy:
That would be me. And what we are doing here is exactly that: counting corpses. The excess deaths measurement is just a superior method for doing it. Number of cases is useless for comparison purposes, it rests on testing which is inconsistently applied and may be full of false negatives. Corpses are real. I try always to use the best data at hand and that was, and remains, deaths.
I’ve said from Day 1 the Chinese were lying about the death tolls and the case load as well. In fact I’ll repeat what I said then: I will not be surprised to learn of mass graves in Wuhan and elsewhere.
And by the way, God help India.
@Sleeping Dog:
Slightly off-topic but I’ve been fascinated by the simultaneous voter failures in three big Anglosphere governments, the US, the UK and Australia. We know the US and UK were subjected to attack by Russian intelligence, I don’t know if the same is true of Australia. And Murdoch’s big lie machine is strong in all three.
I’ve read one suggestion that it is occasioned by the end of the Cold War, which had supplied much of our national narrative since the 50’s. Americans need a narrative, a mythology. Our national narrative has been nibbled away at by people who preferred truth to myth, and I’m with them, but without wise Founders, bold pioneers, manifest destiny, (a side dish of Lost Cause), America as savior and America as the freest and richest place on earth, what exactly is our unifying narrative? We aren’t a race or an ethnic group or a religious group. So what about us makes us, us? If you want to steer a ship you need a course, otherwise you’re blown this way and that by wind and current.
@Michael Reynolds: we actually have some pretty good narratives, from the Declaration of Independence to Obama’s “we’re not red states or blue states, but the United States” speech. But we have a large group of people in this country, funded by some very wealthy backers who are skilled at propaganda, who have chipped away at those narratives for decades.
@Scott: I don’t trust Turkey, either. And the others mentioned are barely-functioning states. For whatever problems it has, Iran is a modern country with an anti-modern political system.
@mattbernius: I agree that there are incentives to under-report even in democratic countries. But there are systemic pressures in place to make that more challenging.
@Sleeping Dog: My Australian student thinks that, ironically, the massive wildfires plaguing his country likely went a long way in controlling the spread of the virus. Partly because the government was already on crisis footing and partly because people were already staying indoors at unusual rates.
@Monala:
“We hold these truths to be self-evident, that all men are created equal.”
Not women.
Not black people.
Certainly not Indians.
Also: not created.
And: not remotely equal in fact, though ideally before the law.
The Declaration is brilliant in parts and silly pettifogging in other parts. But women, blacks and Indians which, together in. 1776 were at a minimum 60% of the population of the colonies, were never included.
The not a red state, not a blue state thing is a reaction to partisanship, it’s not a call to arms, it doesn’t define us except as not being what we quite clearly are.
The doctor I saw in person last week did not give me the test. He thought it might give me a false negative, and in any case he didn’t see much point since it didn’t affect the treatments I was getting. One problem is that there’s no established treatment for Covid-19 yet. It’s just a matter of dealing with the symptoms, and people with non-extreme symptoms–who make up the vast majority–are given regular treatments for more common illnesses such as a cold. I told the doctor about my own personal history with asthma (my mom has it, and I had an isolated asthma attack as a small child) and my suspicion that I might have undiagnosed asthma now, though the asthma-like symptoms I’ve been experiencing are also very standard Covid-19 symptoms. He prescribed me an asthma inhaler for my current breathing problems and told me that after this is all over, I should get myself tested for asthma.
(I recently read an account of having Covid-19 while also having asthma from the singer Pink, who’s about my age.)
I hope to eventually take the antibody test when it becomes available to confirm once and for all whether I had Covid-19, mostly for these statistics (as it’s almost certain I have it).
I’m not trying to start a fight here. I’m pretty sure the last paragraph was meant to say 4,633 deaths in China, and a German population of 80,000,000. [Yes, fixed – jhj]
Michael – I’m still waiting for your apology. By your own standards: fool or liar.
@95 South:
Apologize for what? For accurately characterizing you as either a liar or a moron? I’d do that except it’s true. Liar, moron, your choice dude.
Now, when you outgrow Cult45 and confess your lies and apologize and do your best to make amends, I still won’t apologize because I’ll have been right. But I might start paying attention to what you have to say.
@Michael Reynolds: You should have paid attention before you accused me. Better people learn what they’re talking about first. Still time, though. Find where I said I supported Trump.
“Some of these deaths may be the result of causes other than Covid-19, as people avoid hospitals for other ailments. But excess mortality has risen most steeply in places suffering the worst Covid-19 outbreaks, suggesting most of these deaths are directly related to the virus rather than simply side-effects of lockdowns.”
I don’t understand. Wouldn’t more people be avoiding hospitals for unrelated conditions in the places with the worst breakouts? If I was in Idaho I might skip a tetanus shot, but if I was in New York City I’d need to show every symptom of a heart attack before I’d go to the hospital. Also, where the hospitals are overcrowded with coronavirus patients, the system is going to be more taxed. You should expect more unrelated deaths. So why assume they’re directly related to the virus?
Also, do we normally count the increase in deaths in the death toll of a disease?
From the Washington Post, about 2 hours ago:
https://www.washingtonpost.com/investigations/2020/04/27/covid-19-death-toll-undercounted/?arc404=true
Dying is not flipping a switch for the most part. People start declining around age 20, and by the time we’re at substantial risk of dying from this illness we have accumulated a lot of vulnerabilities. Our heart’s rate capacity to beat is roughly 220 minus your age. If your body demands a rate of 180 beats a minute to sustain your existence, then when you’re over 60 it is not capable to meet the demand. The same for renal function, respiratory capacity, or liver function. A small shove will cause your systems to fail. That is why the CDC has used the metric of excess deaths over baseline in previous flu epidemics.
“How We Die” by Sherwin Nuland, a Yale surgeon, has good details on the process. It’s been more than twenty years since I read the book, but I stick by my recommendation.
@95 South:
It’s saying that both are true — yes people are avoiding hospitals due to concerns about breakouts AND that based on modeling the steep rise in excess mortality suggests that *most* of those deaths are directly related to the virus.
I admit that *most* isn’t a super-specific term. But let’s take that as a “majority” instead.
This is where doing regression analysis on time- and location-based death data series from past years comes into play. Which is epidemiology’s bread and butter.
Depends on the conditions, including our ability to test. See for example: https://www.nature.com/articles/d41586-020-01008-1
I’d also suggest the CDC’s site on death counting – https://www.cdc.gov/surveillance/projects/understanding-death-data.html
But perhaps you’d like to advance an argument on why you don’t think we should be considering the outside the norm, excess death rates in calculating COVID-19’s public health impact.
@James Joyner:
Agreed. Unfortunately, there are also some systemic pressures (outside of direct government actors) that also are enabling/encouraging an undercount (in the name of “just questioning the data”). But things are still very different in more Democratic countries (with free press and more transparency).
That’s a really interesting and on its face, pretty defensible theory. I’m sure once we are passed all of this, there’s going to be a lot of research into that.
@Michael Reynolds:
I am with @Monala, “We hold these truths to be self-evident, that all men are created equal…” is the place to focus. The problem is that too many people don’t want to embrace the notion (or admit that it is an aspiration notion rather than a statement of completion).
Too many want it to be a “republic, not a democracy” because they don’t really believe in the equality of humankind.
@mattbernius: I brought up that old study that claimed 37,000 deaths per percentage point of unemployment. I explained why I thought it was high, and others have noted problems with it, but people also noted that loss of insurance or even housing would put people in danger. Moreso internationally. Would however many fatalities indirectly due to unemployment also be in the number?
“But perhaps you’d like to advance an argument on why you don’t think we should be considering the outside the norm, excess death rates in calculating COVID-19’s public health impact.”
I’m only interested in consistency.
I wonder how exact is the counting of COVID-19 related deaths. Are there standards on how co-morbidities taken into account? I read a local account about a nurse in her sixties who died with “underlying health conditions”. You got to wonder what that means. She was working as a nurse for crying out loud.
And is it treated or untreated underlying health conditions. For instances, I take a low dose of lisinipril for minor elevated blood pressure and another low dose of generic Lipitor. All bring vitals into normal range. Is that considered underlying health conditions? Or is the fact that I’m 66 the underlying health condition.
Seems a lot of leeway to code things.
@Michael Reynolds: And yes, that is the aspirational part.
Our hope at unity is largely creedal. Another place to look, the Gettysburg Address:
Of course, embracing these creeds requires acknowledgement that we have not lived up to them.
@95 South: The only good data I know of right now is for NYC. A month ago, during the worst days, we knew there was about 25 times the normal fatality rate. On one day in particular there were roughly 250 deaths attributed to COVID and 225 to other causes.
On a normal day there are about 25 deaths in NYC. Subtracting 25 from 225 leaves 200 unexpected deaths.
If all 200 of those other deaths were COVID, that would give us a multiplier of 1.8, i.e. there are 1.8 X as many COVID deaths as reported.
Things that might affect that 200 number:
– If there are a significant number of people who died because they couldn’t or didn’t go to the hospital, if would deduct from the 200
– If there were people who would have been killed in pedestrian accidents, car accidents, work related accidents, etc that didn’t because they were staying home, then it would add to that 200
Personally I find it hard to believe that either of those could affect the number by more than a few, but I admit that’s just a moderately educated guess.
@Scott:
Based on what I have read, this is true of flu death counts (and other reasons for death). It just isn’t as straightforward as we might like (or think that it usually is).
As I have argued elsewhere, if we try to make sense of all of this by comparing it to flu season, we need to realize that flue season counts are not based 100% on tests nor death solely due to influenza.
That is not to say what the right way of counting ought to be, but that it should inform how we understand the situation.
@95 South:
In the long run, I would agree. But the spikes in death rates started before the effects of unemployment could hit that level. Unemployment could certainly contribute to any number of problems, but not immediately.
And I agree that some of the death spikes are likely from people not seeking care for other ailments. No doubt someone will figure out some estimates of those–but the spikes are pretty pronounced in the context of a new virus without adequate testing, Logically some portion, likely a substantial portion, are directly linked to Covid-19.
@Michael Reynolds:
A bit OT, but a while back JKB here parroted a story that had been circulating in right-wing circles around the time, to much guffawing: that Facebook censored a post as containing “hate speech,” before it turned out to be from portions of the Declaration of Independence. The implication of this story as righties told it was that it showed how PC standards of “hate speech” were so extreme they’d even exclude parts of the founding documents.
I looked up the story, and as usual it didn’t fit the narrative right-wingers were imposing on it at all. The Facebook post in question did include portions of the Declaration–namely an excerpt that referred to “the merciless Indian savages.” An automated flagging software censored the post. After the users complained about the deletion, it was quickly reinstated.
In other words, it had nothing to do with Facebook standards of hate speech showing a contempt for the country’s founding ideals, as the right-wingers tried to imply. The fact is that, for better or worse, the Declaration includes some ugly language and ideas that Americans today should find offensive, and there should be nothing remarkable about the fact that a computer software flagged it, just as it was right for it to be reinstated once a human was able to look it over and consider the historical context. But I think it also speaks to the right-wing habit of viewing the founding documents as pure and perfect, and looking with suspicion upon anyone who criticizes parts of them.
A flip to side this: there has to be a national decline in automobile-related deaths simply because a lot less people have been driving. So while there are aspects of this situation that would contribute to more death (e.g., people not going to doctor for fear of getting Covid-19 and then dying of something else that would have been preventable) there are aspects (less driving) that would lead to less death than normal.
Staying at home likely has reduced any number of risky behaviors (DUI for example).
@James Joyner: That could be. A population density roughly 1/10 of that of the US doesn’t hurt either. (Although the US population density is not too bad overall. The US cognitive density is another story.)
The latest NYC numbers are interesting, and consistent with those from a couple of weeks ago. NY State has been doing a serology test to estimate the number of people who have been exposed to the virus. It has now reached 25% in NYC.
There are 8.4M residents of NYC, so a 25% infection rate is 2.1M.
There have been just under 13K deaths attributed to COVID in NYC, which gives a fatality rate of 0.6% if 100% of the COVID deaths have been identified as such.
If the multiplier from a few weeks ago is still correct, then the fatality rate is 1.8 x 0.6, or just over 1%
No, but when said death count increase is upwards of 100% over the 5 year average for the same time period, might it not be wise to speculate about whether a brand new, never before seen virus that frequently causes death has some role in the increase?
Asking for a friend. (And I would note that the first question may be part of the reason that some people who claim to be anti-Trump are seen as water carriers for him. Just sayin’…)
@Michael Reynolds: @Steven L. Taylor:
My problem with Michael’s statement is that it tells only part of the story of our founding. Yes, if you weren’t a white male landowner in 1789, you were not a full citizen. In the case of slaves and Native Americans, you were less than a person. However, many of the Framers expected that situation to change, at the very least in one major category. They expected slavery to die a slow, natural death. Of course, that calculation failed, but that was the assumption.
The Constitution was always a future-oriented document, written with the expectation that rights would be clarified or expanded. Certainly, the Framers had many racist and sexist attitudes that neatly coincided with the limitations in the unamended Constitution as written. However, for whom would they be imagining rights to be expanded, if not the people they didn’t currently trust or like?
@Kingdaddy:
It’s even worse than that. They’re “opening up” because they don’t want to pay UI claims. They know they’re going to end up killing millions of people. They don’t care. They think that, somehow, all of the dead will simply vanish into the cornfield, leaving The Economy standing strong.
The U.S. has devolved into a third world banana republic.
I don’t see this country lasting in its current form long enough for another election. States are going to be forced to secede for their own survival. The multi-state coordination conglomerates are the footprints for new countries. Those states might not even realize that right now; in fact, I’m sure they don’t. But as the death toll mounts and infrastructure breaks down in the red states, they’ll have two options: (1) Let their people die and their infrastructure be destroyed alongside the red states or (2) Secede.
I can’t get out of Florida fast enough.
@Scott:
In my case, yes, I go with option B.
@Monala:
And they’ve succeeded. There’s not a damn thing about this country that’s “united” anymore. We are red states and blue states, Branch Trumpidians and everyone else.
The U.S. is not going to survive COVID-19 in its current form. Secession won’t be a good thing, but it will be necessary.
The only “positive” will be that by the time it happens, the red states won’t be in any position to fight back. Half their populations are going to be dead, and their infrastructure will be in ruins. As the body count mounts, people are going to go nuts and start rioting, burning down cities. The feds will send in military to shoot all of them, but their bodily fluids will spread the disease even more quickly. If the feds decide to nuke those cities, well, you can expect secession of the blue states to happen pretty much two hours later.
Who’s going to be left to force them to stay?
The last half marathon I did was Atlanta, literally days before the apocalypse hit. I’m glad I got to run through that city before it was gone. God help anyone who lives there, and I have friends there. I just hope they get out before it’s too late.
@Sleeping Dog:
I’m kind of holding my water on Australia and New Zealand. As of this writing no country south of 23°N has had a major outbreak. Maybe that’s due to climate or temperature or humidity or the angle of the sun or some other factor that hasn’t been identified yet but when it’s true everywhere except for a handful of Caribbean islands that have had larger outbreaks undoubtedly due to tourists it’s hard to attribute Australia’s or New Zealand’s being spared from a major outbreak to anything they’re doing.
And, yes, that includes Brazil and the highly publicized Ecuador. The cases/deaths per million population there are 280/19 and 1,288/33 respectively. Ecuador is doing better than Germany.
“Also, do we normally count the increase in deaths in the death toll of a disease?”
If you are so uninformed, you should stop speaking.
@Dave Schuler: From everything I’ve seen, Ecuador’s numbers are virtually meaningless. The NY Times has people on the ground and a lot of access, and four days ago their analysis showed that Ecuador might have the worst fatality rate in the world.
@Dave Schuler:
In another article a few weeks ago, specifically on New Zealand, it was noted that NZ was among the first countries to test all incoming individuals and to require that arrivals quarantine for 14 days. This would make a large difference, but you are right, that the upcoming winter and their flu season will be the real test.
@Scott: That’s a good question and one that hasn’t been adequately answered.
On the blood pressure medications–Reuters reported last week that there’s some concern about the much-higher prevalence of high blood pressure as an indicator of severe covid-19 illnesses. They’re trying to untangle if it’s the high blood pressure or the medication that is making things worse. Since lisinipril is an ACE inhibitor, it might be worth watching where that research goes.
It’s one of the reasons I find the “most of the dead had underlying health conditions” so frustrating. An enormous number of Americans have what are classified as underlying conditions, with the big three being high blood pressure, obesity, and diabetes.
@95 South:
Wow, it took me a while to figure out what you were talking about. I’m guessing you are referring to the one John Crudele cited in this NY Post editorial (https://nypost.com/2020/04/20/explaining-the-link-between-unemployment-deaths-amid-coronavirus/) — “Corporate Flight: The Causes and Consequences of Economic Dislocation” by Barry Bluestone, Bennett Harrison and Lawrence Baker (1982).
As far as I can tell, this work is rarely referenced (Google turns up less than 100 references in the 38 years since its publication. Further, from what I can tell it wasn’t a peer-reviewed study and I can’t find anyone who reproduced their estimates in a peer-reviewed study.
FWIW, that number seems incredibly high to me. Since the unemployment rate has just jumped from ~3% to approximately 20% that would suggest that we’re going to see approximately 592,000 die from unemployment related causes.
Does that honestly make sense to you? Personally I’m not sure it makes sense to be hanging any “I’m just asking questions or quoting statistics” on a single study with that type of rate.
For the record there are definitely studies that find a correlation between long-term unemployment and increased mortality (see for example – https://www.theguardian.com/society/2015/feb/11/unemployment-causes-45000-suicides-a-year-worldwide-finds-study or https://www.ncbi.nlm.nih.gov/pubmed/24993734), but the claimed rates of death are, from what I can tell, nowhere near as spectacular as that 1982 study claims.
@Slugger: I admire the Nuland book. It was invaluable when I was researching my first novel. Since both my parents died of hypertension + congestive heart failure, it gave me a good idea of what I’m in for, one of these days.
@MarkedMan: Perfectly apropos of the discussion above, the Washington Post and Yale just looked at excess deaths across the nation and find a multiplier of 1.9, which is pretty close to that for NYC.
Note that in the article the NYC number is 2.5 but the analysis was done before the correction NYC made to include people diagnosed post mortem, so my back of the envelope calculation of 1.8 is probably not too far off.
Bottom line, it seems pretty clear that there are almost twice as many COVID fatalities as reported and that the overall fatality rate is 1% or so.
@Kylopod: remember a few years ago, when many rightwingers got upset about NPR’s annual reading/tweeting of the Declaration of Independence on July 4th?
https://www.usatoday.com/story/news/politics/onpolitics/2017/07/05/npr-tweeted-declaration-independence-some-people-got-angry/451112001/
@Steven L. Taylor: And auto insurers – you know, the folks who depend on actuarial data – know this. Most of them are offering premium discounts right now because of it.
@Jen: I’ve seen some of the reporting on the blood pressure medications, which is of great interest to me as my wife takes losartan. Losartan is an ARB, not an ACE2 inhibitor, and as the Reuters piece briefly highlights, ARBs may actually be of some benefit, or at least less harmful than the ACE2 inhibitors. I certainly hope that’s the case. The University of Minnesota is currently recruiting participants for randomized clinical trials of losartan for COVID-19 patients (clinicaltrials.gov identifiers NCT04311177 and NCT04312009) but it will be a while before they are complete.
Sadly, at this point, we don’t know much except that we don’t know much…
@mattbernius: I raised the same concern earlier. I don’t see a massive, short-term unemployment event having that kind of fatality rate. When I brought up the study the first time, I arbitrarily reduced it by 90%. But suicides, domestic abuse, and drug overdoses are all correlated to unemployment, and loss of housing, steady food, or insurance will ruin lives. I don’t know what number we should use, but it’s not zero.
Is there any silver lining here that we are closer to infection rates that approach herd immunity?
@Monala, @ Steven L. Taylor:..there has to be a national decline in automobile-related deaths simply because a lot less people have been driving.
Last week I received a check from my automobile insurance company equal to 30% of my monthly premium for April and May.
@Joe: We’re closer, but even at the highest estimates of infection rate we have five or six times more of this to go. Also, herd immunity isn’t universal, so any relatively untouched populations will still be at risk.
@95 South: I agree, with little effect in the early weeks and increasing as time goes on. Will it be offset by fewer work and car related accidents? I have no idea but it also should be factored in.
I’ve already seen a few municipalities reporting very significant increases in domestic abuse calls.
Tracking and tracing: there is a lot of talk about this. It’s in the air, it’s being looked at. But the big internet and social net corporations have used up their capital when it comes to privacy. Most people now do not trust Google, Facebook, ATT, Microsoft, and the government agency anymore than the local used car dealer, and probably a lot less. How many times have people been discussing some product and then they start getting phone calls, emails, and even texts about that product. The last few years are a littered landscape of bank account hacks, store account hacks, and distribution of millions of names and addresses.
The idea of using a tracking and tracing app is a non-starter. People will ditch their cell phones before their movements are monitored, and then controlled by some sort of health police. No American citizen wants a “yellow” star put on their door.
“I will track you down and then isolate you!” Mayor of LA, the man who wants you to snitch on your neighbor.
” Protests greeted Israeli Prime Minister Benjamin Netanyahu’s roll out this week of a surveillance program that uses the country’s domestic security agency to track the locations of people potentially infected with the virus. (ScienceMag) Israel, of all places?
@Joe: I think one problem is that the non-distancing value of Rzero (the average number of people that each carrier infects) is so high. In this chart posted on Kevin Drum’s blog it shows that most countries started at somewhere between 3 and 4 and even with the most extreme distancing regimes in place have only managed to get it just under 1, the point at which infection dies out. IANAS, but the 25% (presumed) immunity that NYC has might help under the distancing scenario but would probably not be nearly as effective if we go back to close quarters.
@MarkedMan: It’s always a problem for public policy. You want to take the second and third order effects into account, but they’re increasingly difficult to measure. In this case, it’s important to have consistent and clearly labelled data. A virologist doesn’t want the change in unrelated overdoses in his data, but a policy advisor needs those. If a mine closes, a municipality could see a decrease in the number of deaths, but if a county reports -15 coronavirus fatalities people aren’t getting the data they think they are.
@Dave Schuler: “And, yes, that includes Brazil and the highly publicized Ecuador. The cases/deaths per million population there are 280/19 and 1,288/33 respectively”
I wouldn’t take that Brazil number seriously. Bolsonaro is a mini-Trump, and he’s insisting the virus is just a cold.
@95 South:
For practical use you often just need to know whether they are small or big, and I have a feeling we will be able to get there. If they are small, it doesn’t really matter how small. If they are big they are obviously important, but are likely to be easier to measure.
This isn’t the scientific approach, because a scientist typically strives to quantify every variable.
@Nightcrawler:
I’m in North Florida. Internal job boards in my company are shut down right now and might reopen in May. When they do there are 10 stores in Portland Oregon. I’m going to apply to every damn opening at any of those stores. With any luck, in the next 2-3 months I’ll be going JAX to PDX. With any luck, the only confederate flags I’ll ever see again will be on TV.
@95 South:
I wouldn’t expect you to know what numbers we should use. That’s where public policy experts, epidemiologists (as much of what you are describing falls into public health concerns that have been studied for decades), and economists come in.
Those are the same folks who are also talking about the undercounting going on that you seemed to be calling into question up thread when you stated you didn’t understand how they are counting deaths and making statements like the majority of excess deaths can be attributed to the virus. I mean this entire thing is feeling like a bit of a Tom Nichol’s “Death of Expertise” loop where because you don’t understand how this disease modeling and death accounting works you are questioning the disease modeling and death accounting works. And then, to make matters more confusing, you start to grab random facts from partisan editorials and muddy the waters further.
Put a different way, the questions you are asking are not particularly new. They have been being researched for years. And yes, there are ways and models of measuring second and third order effects. I don’t know specifically what they are, in much the same way that I don’t specifically know how brain surgery works. But I know enough about the general fields involved to know they have been working on developing methodologies to deal with this sort of stuff for decades.
Again, I honestly have lost the thread — what is your underlying point in raising these questions? Is it to start a “what’s worse deaths from the disease versus deaths from the economy” conversation?
Or is it to suggest that you think COVID-19 related deaths are ultimately being over reported and economy based deaths are going to be under reported?
At some point there has to be a structured line of inquiry behind asking these questions (and calling specific evidence into play as part of your inquiry).
@mattbernius: Pulling out the reputable from the nonsense is like peeling an onion.
First off, discount everything you hear from TV and radio pundits. They are not experts on this or on anything and their “just speaking common sense” opinions aren’t worth much. This is non-partisan. No TV pundit’s opinion about anything but TV-punditry is worth much.
Secondly, if you are going to pay attention to the guests you have to be able to trust the pundits and their shows, not for the facts but for their commitment to screening. And I think almost all of them fail this. Certainly there doesn’t appear to be any show on Fox News that vets their experts in any way shape or form, but the same basically seems to apply to almost any other TV or Radio network with the exception of Public Radio and Public TV.
I could go on and on, but basically I trust the sources I do because they have a long history of doing the work, of putting boots on the ground, and of vetting the outside experts they use. Public Radio (NPR, PRI and a few others), NY Times, The Atlantic, The Washington Post, and a few others. I can’t speak to Public Television other than the last news show I watched was the MacNeil Lehrer News Hour and both of them were on it. And yes, it was very good.
As an example of the dangers of common sense: Kevin Drum, who I have a great deal of respect for, did a “It’s just common sense kind of post” that involved the way facemark filters worked on protecting from the virus. I happen to have been working on filter projects for more than a half year and now know more than I ever wanted to, and there are still reams of things I don’t know. Drum supposed that the way filters worked in blocking a virus was that the openings in the media were smaller than the virus and so they couldn’t get through. Good common sense. But not at all correct. The pathways through filters are typically much wider than the particles you want to block. The real mechanisms (there are three) are left as an exercise for the reader. Hint: they involve the strong (or weak? I get them mixed up) atomic force.
@Teve: They won’t be the last ones you see. Eastern Oregon and Washington both have a high count of “Ammon Bundy” types. Just stay on the coast! 😉
@Teve:
I have really bad news for you bud:
https://www.theatlantic.com/business/archive/2016/07/racist-history-portland/492035/
@MarkedMan:
Actually, would you share? This reader, for one, can’t figure out what those three mechanisms are. And I think it would be helpful to know (and share).
@Monala: replying so I can click the “notify me” button!
@Jen:
It’s a well known legal fact that you’re allowed to kill someone with underlying conditions and face no charges.
@Teve: Fun Fact*: Before it was a State, it was illegal for a black person to reside in the territory of Oregon.
It’s a more gentle form of racism there. You might not even notice it for a while. Seattle, a similar city, is Racist AF, but with less of the scornful, in your face racism.
I used to work for a well known tech company on a team with the black engineer, and people always assumed he was the cleaning staff. They would make a mess in the kitchen, and then see him and tell him that they were very sorry about the mess, but they have to go to a meeting, and would he please clean it up.
They were very polite, but it just seemed unthinkable that a black man could be a software engineer.
*: the fact is not actually fun.
@95 South: except that , as I posted in the other thread , we saw a *reduction* in death rates in the areas with higher unemployment during the Great Recession.
Theory isn’t reality.
@Joe: No, but some people in the health field seem to be speculating that this virus won’t be one for which having been exposed will offer any immunity. I only saw a headline on this, so I don’t know how to refer you to the article, but I still hope those speculations are wrong.
On the other hand, that how our most famous corona virus (aka “the common cold”) works.
@Monala: So particles of interest are typically 10 microns down to 0.01 microns. Anything bigger is probably just dust. Particles in that range end up stuck to the fibers in a filter by the (strong? Weak?) atomic force, but get there by three mechanisms. For the smallest, it’s Brownian motion. That’s basically what happens when really tiny particles are smacked around by gas molecules. Think of pachinko balls falling down the board. They have a an average downward velocity and on average are all headed in the same direction but the path of any ball is a crazy pattern of ricochets. In the case of our small particles one of these bounces inevitably smacks it into a filter fiber.
In the case of the medium size particles they aren’t much affected by Brownian motion but are very affected by the air so they are carried on the currents curving around the fibers. As long as there are enough fibers and the pattern is random the particles eventually brush up against a fiber and then, bam, atomic force.
The largest particles are the easiest to understand as their momentum causes them to be less affected by air currents so they eventually plow head on into a fiber.
Dust may get trapped on the surface of the filter and can be vacuumed off, but the atomic force holds the littler particles too hard to be vacuumed. It’s an odd thing about filters that the particles adhering to the fibers just give them more surface area. If you use a prefilter to catch the dust the expensive filter actually gets more efficient with time, although it gradually gets tougher to push the air through.
@Gustopher: For that matter, every one of the fatalities died of heart failure. Because the legal definition of death is that the heart stops beating.
(Actually, I think in most states brain death counts so that old saw is no longer valid….)
Uhhh… yeah, sure… that’ll happen.
@MarkedMan:
This is a good reminder that Dunning-Kruger isn’t restricted to a particular ideology or side.
And yeah, the more someone invokes “common sense” the faster they tend to head towards D-Kville.
@MarkedMan:
A character in one of my ebooks who sometimes worked as a talking head said to his wife. ‘The talking heads don’t know s@#$ most of the time and are making it up as they go along.”
@mattbernius: oh the whole state of Oregon was whites only back in the day. I’m aware of the huge amount of racism in the area. But from visiting and having several friends who live there, I’m expecting to have a totally different experience living within the city, sort of like how when I lived in Raleigh near NC State I forgot for long periods of time I was in North Carolina.
@Gustopher:
Fine. That’s better than the in-your-face racism you’ll find here. I have endured that for years. I’m done with it.
@Just nutha ignint cracker: the 6 months i lived in Olympia WA were vastly better than any 6 months i’ve ever spent in Florida, Georgia, or Texas. Not even close.
@MarkedMan:
Uh..wut? Are you talking about van der waals forces? How on earth would the strong or weak force be involved in that???
@Teve: Dude, I’ve lived in the PNW for my entire life except for the 8 years I was in Korea. I lived in Metro Portland for 5 years. It’s bound to be better in some ways than Florida (though cost of living won’t be one of them), but to expect that you are somehow going to completely–or even predominantly–escape redneck attitudes and lost causers? You’re kidding yourself.
@95 South:
Excellent point.
@MarkedMan:
The impact of the shelter-in-place orders is way higher than previous scenarios. I would expect that both positive and negative effects are going to be larger than expected.
Compared to something that kills 1% in good conditions (no overwhelmed health system) getting completely uncontained and having higher cases and death rate… the second order affects are nothing.
Compared to the death counts of the current slow moving pandemic, though, it’s probably noticeable. But that’s comparing against the wrong thing.
Now, if we had enough testing so we were able to say that the death rate was 0.1%… that would change everything. But we don’t have the testing, so we have to work off the best estimates. I give no credence to those who screw up the testing and then say “well, we don’t have the data, but I’ve got a hunch it will all just magically disappear.”
@Teve: You are absolutely right About string/weak force, so my bad, and according to this paper the mechanism is the Van Der Waals force, as you said.
@Michael Reynolds: @Kingdaddy: @Steven L. Taylor: I’ve long seen the Declaration, and even the Constitution, as aspirational. I don’t know that many saw women’s equality as an aspiration in 1776 or 1789 but certainly many opposed slavery at the founding. But, more importantly, founding ourselves on a creed that we were clearly not living up to created a forcing function. At points where the zeitgeist was more favorable, those words could be used to shame those on the fence into action.
@Steven L. Taylor:
Yup. We still have no idea what killed my first wife. The autopsy found nothing conclusive. The death certificate says it was complications from strep. That’s because she had gone to the hospital the previous day and been diagnosed with strep and prescribed antibiotics. When we investigated the records afterwards, we found that the test for strep came back negative.
So, the death certificate says strep. The only thing we can rule out: strep.
@Teve:
I’m in Jax too. We’re probably neighbors, heh.
My husband and I are extremely fortunate that we own property in Delaware, and that we were already planning to move back there pre-apocalypse.
It’s so sad what’s about to happen down here. I hate the government, and I hate my landlord, but I don’t hate Jacksonville or Florida. There are a lot of Confederate flags and MAGAts here, but there are also a lot of good people who didn’t vote for any of this and don’t support it.
@mattbernius: I’m not trying to play a game. I’m just participating in a conversation.
I’m moving to Delaware. There are Confederate flags and MAGAts there, especially below the D&C canal, in “Slower Delaware.” I’m not fooling myself into thinking it’s some sort of Utopia. But it’s a blue state, and the governor isn’t out to kill tens of thousands of people to avoid paying UI.
It’s safer there than it is in Florida, or any red state.
@Jen: There’s a bit of an attitude towards coronavirus victims: “oh, I guess if you have high blood pressure/obesity/other underlying health condition then it’s your own fault that you’re dying of coronavirus.”
Makes me want to bite someone’s leg off, especially since it’s our present style of work that causes most of us to not get enough exercise. I’d LOVE to be able to a) hold down a good paying job, and b) get in the amount of walking I used to get while working in Japan.
@95 South:
Got it.
Then, yeah, I guess the answer to the underlying initial question is that while journalists covering these stories are not always the best at explaining it, the various experts who are doing the underlying analysis have tools to account for exactly the questions you are raising.
My suggestion is, where possible, to go directly to those sources. There are a number of qualified expert epidemiologists, public health researchers, and others actively tweeting on these topics in relatively transparent language. Carl T Bergstrom, a University of Washington Biology Prof, is a good place to start as he aggregates a lot of studies: https://twitter.com/CT_Bergstrom, you might also want to look at more indepth science reporting sites like Scientific American, Nature, and the New Atlantis (https://www.thenewatlantis.com/).
@James Joyner:
The limitation of the Declaration as national narrative is that – forgive me indulging in fiction tropes – there’s no action. Nothing happens, no one is doing anything.
Pilgrims did stuff. Pioneers did stuff. Cowboys did stuff. Bootleggers, inventors, astronauts all did things. The Civil War had a lot of action. The Depression was all about struggle and survival. WW2 had lots of action and the whole ‘savior of the world’ thing. Civil Rights had marches and sit-ins. The Cold War had spies and mini-wars and the long, twilight struggle to preserve our way of life and keep fluoride out of the water supply.
The essence of the Declaration in the modern context is self-congratulation not for doing but for being. It’s essentially identity politics. Hi, we don’t like you, King, so we’re going to be a different country where we’re going to demonstrate moral superiority. We are free, all you other poor dumb bastards out there are not. That’s not a narrative it’s just introducing a character: the American.
Now we are the most powerful nation on earth, but that was never what we were supposed to be about. America was the scrappy upstart. America was revolutionary. Now? We are the status quo power and have been for 70 years. What do we do? Where’s the action? The propulsive force of the narrative? We stomp little countries till we get bored. We move money from Pocket A into Pocket B. We make memes and invent new ways of propagating memes.
We are not the freest country on earth. We are not the only democracy or the most democratic democracy. You know what we are as a nation? We’re old, rich and tired. We aren’t special, we aren’t going anywhere, we aren’t doing anything but making money and growing the world’s fattest asses.
The US of A is in decline, not because someone else ‘beat’ us but because we’ve run out of ideas beyond make money and buy stuff.
@grumpy realist:
Cancer patients experience this all the time. Lung cancer patients get it the worst, but we’ve all heard it, even though ~70% of cancers have no known etiology.
I wasn’t “supposed” to get breast cancer. I had no family history. I wasn’t obese. I didn’t smoke or drink. I was a marathon runner.
But in the U.S., illness is viewed as a punishment for wrongdoing, a character defect, instead of something that just fucking happens.
@James Joyner:
I am sorry about your wife.
Unfortunately, autopsies frequently don’t give families the answers they seek. As you said, it’s not that straightforward. It’s not like it is on TV.
@Nightcrawler: sadly I’m in Lake City. The reason this county only went 70% for Trump is that 20% of the residents are black.
@Just nutha ignint cracker: I think you’re interpreting a casual comment I made as if it were a binding legal contract. The six months I spent outside of Olympia Washington were so much better than life in the Deep South that it might as well have been Utopia. And I expect Portland will be just as good as Olympia was.
@Nightcrawler: according to Area Vibes, Jacksonville and Lake City are similar on several livability metrics, except LC has worse schools, housing, and employment.
Despite being a much bigger city than the little podunk town I live in, Portland has much better numbers on 9 out of 10 crime metrics.
@Teve:
I’ve lived all over the South – Virginia, North Carolina, Tennessee, the Florida panhandle as well as Orlando and Sarasota, and every part of the west coast is better.
@Teve: Nah, I just wanted to make sure that if you really believed that you were moving to Utopia, that someone warned you that it might be prudent to lower your expectations some. On the other hand, I live somewhere that went 68% for Trump and where my barber was asking me one day what business schools had admitting kids who didn’t already speak English (something I’ve also heard Portland residents opine). I suspect that except for the climate, I probably would be able to do okay in Lake City FL. By the same token, when I was a little younger than you are now I was one of those people, so it may be natural that it doesn’t trouble me as much. If you come, I hope it works for you.
@Michael Reynolds:
I think it’s more than that. We were making a fresh start and doing so on the principles of the Enlightenment. Sure, some of it was grandstanding. But we’ve returned to those principles time and again and they’ve made us steadily more enlightened. We’re much more “equal” now than we were in 1776. Or even 1976. Or, hell, 2006.
Making money and buying stuff is pretty much who most Americans were in 1776. Many of the grievances in the Declaration were, after all, economic ones.
There are other countries doing some things better than us, especially under our current leadership. We’re hampered by trying to operate under a Constitution written to solve the problems of 1787. But, aside from maybe some tiny, homogenous countries like the Nordics, I’m not sure anyone is doing it better.