Cause of Death
This seems to be OT’s week for thought experiments, so let’s do another one.
Let’s pretend that you can’t walk without the assistance of a cane. I, being a jerk, come up and give you a shove. It’s a shove that might tumble a person who walked unassisted. But with your impairment, you easily tumble to the ground. Then when you yell at me, I say, “Hey, it wasn’t my fault. It was the cane’s fault!”
That would obviously be insane, right? That would be the mark of a sociopath. And yet this would be similar to the chain of logic that has led the Q hive, multiple celebrities and even the president to claim that 94% of COVID deaths are not really from COVID.
This weekend, a tweet went viral claiming the CDC had “quietly” updated it’s data to show that only 6% of the COVID-19 deaths are actually from COVID-19. The other deaths attributed to COVID-19 are listed as having come from things like heart attack, pneumonia or kidney failure. The claim that the CDC had “admitted” that COVID-19 deaths were vastly exaggerated was then amplified by Gateway Pundit (aka, “The Dumbest Man on the Internet”1, lit up COVID denialist Twitter and was even RT’d by the President himself2. But is this true? Of the nearly 180,000 deaths attributed to COVID-19, are only 9,000 from the virus? Did most people actually die with the virus rather than of the virus?
No.
First of all, the CDC did not quietly anything. They have been publishing this data for months. Someone ignorant of how data works suddenly noticed it and drew the wrong conclusion.
Second, what the CDC lists is not what people are claiming it is. “Cause of death”, as listed on a death certificate or medical record is … complicated. I’ll let Orac, in a holy-cow-you-should-really-read-the-whole-thing post explain:
To understand how the CDC table was tabulated, let’s elaborate on the Tweet above. You have to understand that it was compiled from standardized death certificates. It’s been a long time since I’ve had to fill out a death certificate—thankfully!—but I still remember how they work. On the death certificate form, there is a space for the immediate cause of death and then several lines for underlying causes. In brief, death certificates are filled out by the medical certifier (who can be the physician who had treated the patient before death), who provides his best medical opinion regarding the cause of death. Part I of the death certificate includes the proximal cause of death, or what directly caused the death, and Part II lists conditions that contributed to the death:
For example, if a patient dies of respiratory failure due to acute respiratory distress syndrome (ARDS), which was the result of pneumonia, which was the result of COVID-19, the proximal cause of death was the respiratory failure, but contributing causes were ARDS and COVID-19, with the one farthest up the chain being the underlying cause of death under Part I. If the patient had hypertension or asthma, that would go under Part II. As I like to say, if you suffer a cardiac arrest due to blood loss after being shot, the cardiac arrest might have been the proximal cause of death, but you still died of a gunshot wound.
A few years ago, I lost someone to a drug overdose. The cause of death was not listed as drug overdose. It was listed as cardiac arrest. Drug overdose was listed as a contributing factor, along with several other things. Using the tortured logic of the COVID denialists, I would conclude that my friend died with a drug overdose rather than of a drug overdose even though they God damned well died of a fricking drug overdose.
In short, people do not generally die of COVID-19 itself. They die of the pneumonia it causes. They die because they can’t breathe. They die because their kidneys fail. They die because their heart stops. A severe case of COVID-19 results in an assault on many bodily systems and there are many ways for the disease to kill. Really, the error here is that 6% of the death certificates listed COVID-19 as the cause of death because they really shouldn’t; they should be listing the exact path COVID-19 used to kill.
The reality of COVID-19 as a mass murderer can be seen in the cold equations of death statistics. The CDC’s data shows that something like 200,000 more Americans have died to this point in the year than die in a typical year, mostly of things like pneumonia. Yet somehow that data was missed by the Qidiots when they started conspiracy theorizing.
The less conspiratorial interpretation of the CDC’s data has been with us for some time and goeth thusly: “Well, OK, those people died of COVID-19. But they had comorbidities!3 They were old! They were obese! They had pre-existing conditions! Let’s just protect those people and the rest of us can get on with our lives!”
The role that comorbidities play in COVID-19 has never been concealed. The first evaluations out of Wuhan indicated this was a big factor. But let’s take a step back here. The biggest comorbidity for COVID-19 is hypertension. One-third of American adults have high blood pressure. And a huge part of that is genetic. There are many people out there who are watching their diets, getting exercise, avoiding salt … and they still have high blood pressure because they are genetically pre-disposed to it.4
You know how many Americans have pre-existing conditions? 40%. This ranges from 30% in Utah to half in West Virginia. That means 40% have a high risk of serious illness and death. And it’s not just the oldest 40%.
Even putting that aside, COVID-19 is a serious illness. There are people out there — healthy young people — still dealing with it months after their infection. The path of COVID-19 is not that you live or you die. Most people won’t die. But a huge fraction of those that survive will endure serious hospitalizations with all the stress that entails (you can read Andrew’s post on what it’s like to survive that sort of thing). A large fraction of those will have long-term health effects, health effects that will become “pre-existing conditions” if COVID-19 (or even just a bad flu) comes back for another round. Even if we are able to defeat COVID-19, “complications from COVID-19” may be listed as a contributing cause of death long for a long time afterward.
And even putting that aside, what the holy hell is being said here? That people with pre-existing conditions deserve to die? I wouldn’t say that, except we have a very long and ugly history, as a species, of saying exactly that. From biblical claims that leprosy was caused by sin to medieval claims that the Black Death was caused by witchcraft to modern claims that AIDS only killed gays and sex workers, we have a tendency to turn a blind eye to the suffering of our fellow human if we think they did something to deserve it. “You wouldn’t have died of COVID-19 if you weren’t fat” is one hell of rallying cry.
Really, this is COVID-19 denialism distilled into its utter essence. From the first week, they have been looking to minimize the problem or find a quick fix to make it go away. They tried denying it was serious, which didn’t work. They jumped on hydroxochloroquine, which didn’t work. They said the summer weather would end it, which didn’t happen. They said the only way to deal with it was herd immunity, but Americans balked at the idea of 1-2 million deaths (and it’s not clear that this would work). So now we’re back to denial with a sprinkling of herd immunity nonsense.
The simple reality is that COVID-19 is with us for the medium haul, if not the long one. As our understanding of it and our treatment courses improve, we will be able to ease off some of the restrictions. If we get a vaccine — even a weak one — that will allow us to open up a bit more. But wishful thinking and misreading CDC reports is not going to get us out of this.
I know a lot people hate the analogy of a war, but we’re in the equivalent of one. It’s going to be expensive, drawn-out, painful and messy. But we will get through it if we keep our wits about us. Misrepresenting a CDC report to claim that the virus didn’t slaughter tens of thousands of people is the opposite of that.
- Which is saying something.
- It was also disappointingly repeated by a couple of contributors at Reason who should frankly know better.
- Let me just pause a moment and note that one of the weirdest things about COVID-19 is the way complex scientific concepts have entered our common lexicon.
- And many may have undiagnosed hypertension because they don’t have insurance.
I’ve heard that before, let me see if I can remember where…
Report
Yep. That was the exact comparison I was hinting at.Report
Anecdotal story: The first night of working as an overnight administrator at a large medical center, there was an DOA in the emergency department. A elderly man was found unresponsive at his home by his son and the son called 911. EMS drove the non-breathing, no pulse individual to the nearest emergency department even though our Emergency Department was not a trauma center. Following the SOP for reporting and recording deaths, I contacted the administrator that dealt with deaths, death certificate, and the accompanying paperwork (referred to as a death packet). That administrator came in to use the paperwork to complete a death certificate. The attending physician in the Emergency Department wrote on the death reporting form that the patient died from cardiac arrest. We all had a laugh about the cause and told the physician to be more specific. So the physician added pulmonary failure and rental failure. So the individual died because his heart stopped and he stopped breathing. In other words, the physician and the rest of us had no clue what he died of and no one cared. That is why epidemiologist do not use death certificate data to perform mortality studies.
As a side note, when I was trained on the procedure to release remains to a funeral home, everyone involved said that the mortician would look exactly like what one would believe how they should look. That turned out to be true.Report
The legitimate mischaracterizations of death aren’t helping with the conspiridiots. I’m sure you could count them on one hand, but so long as you have car crash victims showing up on the list of Covid deaths, you’re going to enable questions of the legitimacy of the counts.
We seem to oscillate between overstating and understating the risk of Covid. Which may just be how we make sense of something so new and unknown.
I do think delineating those with co-morbidities (or multiple co-morbidities) from those without is important. It isn’t black-and-white, if-you-have-a-comorbidity-you-didnt-really-die-from-Covid-so-Covid-isn’t-bad. But it also isn’t irrelevant. The truth lies somewhere in the middle. The risk that Covid presents isn’t evenly distributed. That’s true of ALL diseases. So knowing individual risk profiles can be really, really helpful.
It does matter that NY and NJ didn’t know (or didn’t care, depending on who you ask) that the elderly were particularly vulnerable to Covid because there policies increased the risk to a highly vulnerable group, resulting in many deaths that probably could have been avoided. That doesn’t mean that Covid isn’t bad. It does mean that Covid (as a ‘thing’) can be made less bad by good policies informed by good science and data.Report
This is a pretty good article. Probably better than I’m giving it credit for. The under-estimators are misrepresenting the data and the news stories, and I’m glad people are calling them out, with supporting information.
But excess deaths are something that the over-estimators are relying on too much, and I wish the article hadn’t cited it. I’m only interested in balls and strikes. The 6% obviously died from coronavirus. The 94% had other ailments either before or after catching the virus, and the vast majority of them wouldn’t have died if they didn’t have the virus. But as near as I can tell, the “excess deaths” are called that because they aren’t classified. The difference between the 200k and the 180k could be because covid tests weren’t run, or because people are avoiding doctors and hospitals and not getting checked, or because doctors and hospitals have in a few cases been overwhelmed, or because of indirect effects of the shutdown (in particular, depression, substance abuse, and violence).
Another complaint: “The path of COVID-19 is not that you live or you die. Most people won’t die. But a huge fraction of those that survive will endure serious hospitalizations with all the stress that entails (you can read Andrew’s post on what it’s like to survive that sort of thing).” From what I understand, the largest fraction of those who survive won’t even realize they had it. The next largest fraction will have something like, dare I say it, a case of flu.
Two additional complaints. (I’m sorry I’m going hard on this article. I really do think it’s a net positive.) You probably shouldn’t say that hydroxochloroquine doesn’t work. It’s more complicated. But the bigger objection I have is your reading of the motives of the, let’s call them deniers. You accuse them of thinking “that people with pre-existing conditions deserve to die”, and I don’t think that’s fair at all. People are questioning if the commonly accepted total deaths are accurate. No one wants an 89-year-old with one lung to die, and no one’s blaming him (unless he was a smoker), but they’re saying that he wasn’t going to live much longer anyway. If it weren’t for the virus, he might only average a year. I note in this article you discuss the dangers to “healthy young people”, and only once mention the old, in the mouth of a denialist. But social policy involves making brutal choices. It’s natural to consider the death of a healthy young person a higher price than the death of a sick old person – every sick old person I’ve known thought so. So the deniers suspect the numbers are being padded, and that the numbers don’t reflect the reality that we should be looking at to make the right policy decisions.
So please, more balls and strikes, and less characterization of opponents.Report
Thank you for writing this, Michael. Really good stuff.Report