Hey all. Im a long-time lurker, first time poster. Man, yall are intimidating! I finally have a question that is burning in my brain so hot that im putting my big-boy Osh-Kosh Bigosh overalls on and im wading in to the SD deep end.
Anyway, what I’d like to know is if there are specific criteria in place to determine what is, and what isnt a Covid-related death? And if so, what are those criteria? It seems like such a murky area, and given the extreme hot-button political issue this pandemic unfortunately is, the criteria, or lack thereof, that determines this 200,000+ death toll from Covid-19 seems ripe for malfeasance.
And even more likely or open, for abuse are the state numbers. The specific numbers can dictate certain powers or abilities to be bestowed on leaders of states in distress. For example, hyperbolic example intended, lets say someone finds out they have coronavirus which has caused fairly severe complications. If this person then commits suicide, is that death considered a “Covid-related” death and added to the overall death toll?
Who would want the power to make a large percentage of the voters pissed off at you and/or bring the economy to a screeching halt? Who in your mind gains politically from this mess?
Well, to use my state’s leadership for example, a Democratic governor who has Presidential ambitions could see it as politically advantageous to have a very hands-on approach/record with the pandemic.
I believe the definition used for the 200k number is someone who
had a positive test for Covid or was presumed positive (generally meaning they had lots of close contact with someone who was positive and also had symptoms
Died while experiencing a known Covid complication (low O2, trouble breathing, stroke, etc.).
For a lot of reasons, the data are noisy. A few people who died of a stroke after getting a positive test might have died of a stroke anyway. Plenty of people have probably died of Covid without ever getting a test.
For a lot of reasons, the all-cause excess death numbers are a much better measure. It’s just a measurement of how many more peopled died this year than in an average year. This number is both
Larger than the official Covid death count, meaning the official count is likely undercounting deaths due to Covid
and
Includes changes in deaths that are indirectly caused by behavior change around Covid. For example, people are drinking more, so you might see more alcohol related deaths go up, but they’re driving less, so you’ll see car accident trauma deaths go down. This is useful for measuring the deaths caused by the pandemic rather than deaths directly caused by the virus
And just to provide an alternate motivation, a Republican governor who sees it as politically advantageous to keep “business as usual” to keep the economy going as much as possible would have motivation do downplay the seriousness of the pandemic to make their weak response seem like the most reasonable. One method to do that is make claims that the covid death numbers are exaggerated. Just recently I’ve seen the idea floated that if a person with covid dies, and they had any other disease at all (e.g obesity, high blood pressure), it should not be counted as a covid death – IMO an obvious ploy to cast doubt on the official numbers.
That is why the excess death numbers are useful. It may be a matter of opinion whether a specific death is a covid death or not, but it is much less a matter of opinion whether someone actually is dead. And something has been killing a lot more Americans than is typical since February.
It’s also generally believed that many people who are older, suffer from chronic conditions, and/or are simply very concerned about COVID, have been skipping or delaying their regular appointments with their doctors and other healthcare providers, and even avoiding going to see a doctor when they’re actually feeling ill. It’s likely that there are some excess deaths which are coming about due to this, as well.
How Covid deaths are reported varies from state to state. Here in Oregon there is no distinction from those who died of Covid and those who died with covid. Washington used to count the same way but they changed it and removed 3 homicides, 2 suicides, 2 overdoses. Oregon has no plans to change the way they count Covid deaths. so regardless of what you actually died from, if you test positive for Covid in Oregon, that is what you died from. Your wife shooting you in the head was probably a contributing factor, but if you tested positive you are another Covid death, at least in Oregon.
See that is what I dont understand. They don’t count deaths from other conditions this way. A person who is diagnosed with cancer who then commits suicide does not get their cause of death listed as “death from cancer”, its listed as a suicide. Right? Or am I mistaken. I mean, that seems like a bit of a rabbit hole if thats not the way its handled. I mean, any potentially terminal condition that a person has who ends up dying by totally unrelated causes could then be listed as dying from the condition, even tho that was not the actual ca
The point is not that they will (or should) be directly counted as having died from COVID on the micro level.
The point is that their deaths may have been indirectly influenced by the presence of COVID (which caused them to not seek medical care, which might have prevented their deaths at this time), and so, they contribute to the overall “excess deaths” during the pandemic.
But those people are directly counted towards the sum total of deaths from the Covid virus, are they not? And so I ask would someone just diagnosed with cancer or who suffers from bipolar depression who commits suicide have their death listed as caused by cancer or depression or would it be listed as a suicide? After all, their deaths were “influenced” by their medical conditions?
If they also had COVID, then it appears it may depend on the state in which they died (as per the link that @Dallas_Jones provided). That’s not what I’m talking about.
What I’m talking about, and what @iamthewalrus_3 was talking about, is “excess deaths.” That calculation, as I understand it, does not look at listed causes of deaths. What it simply does is look at the overall death rate from all causes (probably in “deaths per million people,” or a similar measure) for a population (a state, a country, etc.) over a period of time (say, January-September 2020), and compares it to the same time period from the previous year. It’s an overall measure of the overall effect, both direct and indirect, of the virus on deaths.
Well this is more in line with my question, it seems rather un-reliable or scattershot to leave the criteria for qualifying as a covid death up to each state. How could this then be unified in any reliable way into that 200,000 + number that is so daunting to see?
That the 200K is an undercount of the COVID deaths is what everyone is telling you. You appear to not hear it, but that’s still what is being said. For example, pneumonia deaths alone are so astronomically high this year that it appears at least 50k of them should have been attributed to COVID, but testing was never done.
No, we don’t know if COVID killed exactly 201,245.3252 people in the US as of 2020-09-21T08:21:34+00:00. We’re never going to get that number. The really important thing when you can’t get exact numbers is to be consistent with your methodology and then observe trends. Consistent doesn’t mean everyone has to agree how to count, it simply means don’t change methods all over the place.
With that understanding, and since the actual question in your title has an answer of “It varies”, what exactly would you really like to know? Are you curious whether the real number is greater or less than 200k? Is it whether we are on an upward or downward trend?
You know, I dont think my question was very well hashed out. I feel like i am wanting to know bits of several things. Btw, I do hear what is being said.
But if there is a different set of standards in place for each state for determining what counts as a Covid-related death, regardless if the overall numbers are an under-counted, doesn’t it strike you as being a potentially very inaccurate method of reporting these deaths? Or you’re saying having all the states use one set of criteria isn’t really what matters? Its continuing with the same methods of previous years and tracking the rise in deaths from the previous years using those same methods?
It’s not terribly accurate, but since those at the federal level don’t seem to want to push states to try and get on the same page, it’s what we have. If you have messy data and there is nothing you can do about it, you then focus on trends. So we can’t tell if it’s 250K dead or 260K dead, but we can tell whether it is improving or getting worse.
That’s different. Since we’re not performing autopsies on every death in American, we’re never going to have perfect counts even if we do agree on a definition. What this speaks to is excess deaths, which is different from the COVID death numbers being reported, and might be more accurate although also flawed. For example, the number of excess deaths from pneumonia this year is so high it is over 8 standard deviations from the mean of the past several years. We can probably assume that the majority of those extra deaths were COVID related, since we don’t seem to have another cause of a pneumonia pandemic. If John Doe’s lungs stopped working, was it due to COVID or the COPD he had had for 5 years? Excess deaths tries to help with that answer, albeit imperfectly. It should be noted that excess deaths this year exceeds the reported COVID deaths by quite a bit.
Now, if you have a better method of counting, share it with the world and maybe they’ll adopt it. In the meantime, this is where we are.
To complicate matters even further, I bet than in many states it depends on what the doctor who issues the death certificate thinks is the cause of death. And those numbers are also messy.
For instance, my father’s death certificate says “heart attack” because that’s kinda the default. Except he actually had an autopsy that showed he died of a pulmonary embolism. But his death is still in the “heart attack” bucket for all reporting purposes.
My husband’s uncle’s death was also listed as “heart attack” despite evidence that he died of a bowel obstruction.
It’s important to understand that all data is messy. I don’t think there are any situations where anyone counts to 200K and gets all the counts exactly right. Anyone who regularly works with data is aware of this.
So, each state (and perhaps each doctor) are using slightly different rules. And we know that there are a few “overcounts” --the homicide that was removed from the covid count, for instance). And we know that there is undercounting, too, since most people are never tested for covid, and we’ve seen a LOT of excess deaths from pneumonia, strokes, and senility, which are all conditions that covid causes and/or aggravates. (My uncle was tested for covid because he experienced sudden-onset of severe senility – like, in a week he went from really sharp to incompetent. He died of covid a week later.)
The key thing, imo, is to understand that fundamentally data is messy and imprecise. And also, that fundamentally, cause of death is, too. Yes, if a healthy young man is shot in the jugular and bleeds to death, it’s best to list that as “homicide” or “bullet wound”.
But lots of deaths aren’t that clear-cut. What about the guy who has a stroke while driving, and passes out and drives in front of a truck? Sometimes the autopsy is unclear as to whether he was still alive when the truck hit him. What is the cause of death? Even if he was still alive, perhaps it would make more sense to “blame” the stroke?
Or the cancer patient who falls and dies. But they fell because the chemotherapy weakened their bones and their hip broke. And if not for the cancer, they would have survived the fall? What’s the one true cause of death?
The answer, in many cases, is that there isn’t one true cause of death.