This isn't a "study", it's a statistical model. It reflects the assumptions put into it. To wit:
> Our analysis follows four key steps. First, for all locations where weekly or monthly all-cause mortality has been reported since the start of the pandemic, we estimate how much mortality increased compared to the expected death rate....Second, based on a range of studies and consideration of other evidence, we estimate the fraction of excess mortality that is from total COVID-19 deaths as opposed to the five other drivers that influence excess mortality. Third, we build a statistical model that predicts the weekly ratio of total COVID-19 deaths to reported COVID-19 deaths based on covariates and spatial effects. Fourth, we use this statistical relationship to predict the ratio of total to reported COVID-19 deaths in places without data on total COVID-19 deaths and then multiply the reported COVID-19 deaths by this ratio to generate estimates of total COVID-19 deaths for all locations.
And, what, you may ask, are the biases of the models' authors with regards to the percentage of excess mortality due to Covid itself? Excellent question:
> Deaths that are directly due to COVID-19 are likely underreported in many locations, particularly in settings where COVID-19 testing is in short supply. Most excess mortality is likely misclassified COVID-19 deaths.
> Given that there is insufficient evidence to estimate these contributions to excess mortality, for now we assume that total COVID-19 deaths equal excess mortality. For the reasons presented in this section, we believe that this is likely an underestimate.
As someone who has now spent the better part of his life creating statistical models, when you assume something is true, your model is likely to confirm your assumptions.
This is why statistical models are worthless without prospective validation. If you haven't shown that your assumptions are correct for the future, you're just making castles in the sky.
"Most of the downplaying I've seen has revolved around the idea that those who died would have, "died anyway" from some other ailment or that hospitals are finding any reason they can to attribute deaths to covid."
The authors of this are doing exactly the same thing, but from the other direction:
> Given that there is insufficient evidence to estimate these contributions to excess mortality, for now we assume that total COVID-19 deaths equal excess mortality. For the reasons presented in this section, we believe that this is likely an underestimate.
This is all simply noise. The way we'll find out the answer to this question is when someone pulls the death certificates for 2020, and actually tabulates the percentage directly caused by Covid.
The study is counting all excess mortality during the pandemic. This is a bad methodology, and I might add possibly a politically motivated one, but that's neither here nor there.
There are metrics all over the place showing some regions with no lockdowns having less excess mortality and mortality due to covid. Some of the lockdown measures in some places were counterproductive. Some people say all of them were, some people say none of them were anywhere (which makes no sense really), but the truth is probably somewhere in the middle: at least some measures taken in some places were the wrong call. This is to be expected, we didn't know what the disease was going to do and lots of us panicked. But conflating deaths from covid with deaths from the response and attributing all of them to the same cause is going to necessarily include deaths that would not have happened even in the pandemic if it weren't for the response. And we can quantify that, as an example, look at the disparity in excess mortality and covid death between California and Florida. California had a higher excess mortality and a higher covid death rate. Was that excess death in California due to covid, or due to a botched response? Again, probably a bit of both, but there's no way to pick the two categories apart, and so this entire methodology is not useful to quantify anything.
The purpose of the excess death model is to measure the real number of deaths without having to argue about whether they were caused by covid infection or not. It is simply the difference between the typical number of deaths in a given time and the real number of deaths -- unless you're saying that death certificates are being fabricated altogether?
> for now we assume that total COVID-19 deaths equal excess mortality.
This seems reasonable, although I'd suggest it's an underestimate of COVID-19 deaths.
Excess deaths counts have been consistently showing negative in countries during periods of time where they have coronavirus under control.
Other significant causes of death (including suicide, influenza, road traffic accidents) appear to be down.
As such, I'd assume that COVID-19 deaths are likely somewhat higher than excess mortality figures. I'm prepared to accept the assumption that they're approximately the same.
[Edited to note: the linked article discusses this with further detail and evidence]
I think that's exactly the criticism of this paper. They are precisely saying "total mortality due to COVID-19" = "total mortality from COVID-19". From the paper:
> for now we assume that total COVID-19 deaths equal excess mortality
Surely you'd agree that "2020 excess deaths are going to be predominantly covid" is a very reasonable hypothesis though, right? And you agree that "2020 has an extraordinarily high excess death rate", because that part is easily measured data, right?
Basically: your response seems a bit nitpicky to me, which tells me that perhaps you're starting from a perspective where you're assuming the article is wrong and looking for holes to poke.
Look, nothing about this paper's conclusions is surprising at all. At all. We all know true covid deaths are higher than reported, because all such metrics underreport, for all diseases (and frankly almost all causes of death). The only question is "By how much?". So you don't like 400k. What's your counter?
This is an excess mortality study, not going off of positive covid tests. And there's some meaningful research that suggests that those are relatively accurate because indirect deaths have gone up and down in a way that could be a wash (e.g. less traffic deaths because of shutdown, but more deaths from untreated illness because hospitals were overwhelmed).
And this is Academic modeling - they don't have an incentive to inflate the numbers.
"Summing up the excess mortality estimates across all countries in our dataset gives 3.3 million excess deaths. In contrast, summing up the official COVID-19 death counts gives only 2.1 million deaths."
Interestingly enough, their conclusion was that covid deaths are being undercounted:
"At the time of writing, the world’s official COVID-19 death count is 2.9 million. Our results suggest that the true toll may be above 4.5 million."
>the discrepancy between reported deaths and analyses of death rates compared to expected death rates, sometimes referred to as “excess mortality,” suggests that the total COVID-19 death rate is many multiples larger than official reports.
So...not as many people died as we thought were going to so we're now just going to arbitrarily say any excess death was a covid death just to make sure the numbers look like what we think they should?
These are the kinds of statistical shenanigans that have been going on all year and people wonder why people question 'the science.'
They claim COVID led to behavioral changes that lowered the expected volume of death, so the naive excess mortality is an undercount:
> Overall, the evidence suggests reductions of 615,000 deaths, or potentially more, stemming from behavioral changes at the global level. The main potential increases in excess mortality due to deferred care and increases in drug overdose and depression are hard to quantify at this point or are of a much smaller magnitude. Given that there is insufficient evidence to estimate these contributions to excess mortality, for now we assume that total COVID-19 deaths equal excess mortality
So basically they lowered the expected number of deaths, then claimed all deaths above expectation are COVID.
> Isn't this exactly what measuring excess mortality is for? But it's not reported anymore since there is none
It isn't reported because it takes more explanation than the media likes to give, and direct COVID-19 death counts are less complicated, not because “there is none” which is just false:
I won't be able to explain that, because this is how much explanation given
> machine-learning model, which estimates excess deaths for every country on every day since the pandemic began. It is based both on official excess-mortality data and on more than 100 other statistical indicators
let me get the premise of this right: they built a "machine learning" model, using empirical data from normal years, and magical "statistical indicators", to predict excess deaths in a time of unprecedented events. Then, when the model does not fit, the conclusion isn't that the "model" is full of shit, but the "COVID deaths are undercounted"?
You have to do your own math to compare the excess deaths for the period of week 13-15 (around a thousand extra deaths) and the reported covid-19 deaths (just over 500).
It's difficult to find any article that I can cite for the disparity, as pretty much no-one wants to report it.
> The present study estimates the burden of COVID-19 on mortality. The state-of-the-art method of actuarial science is used to estimate the expected number of all-cause deaths in 2020 to 2022, if there had been no pandemic. Then the number of observed all-cause deaths is compared with this expected number of all-cause deaths, yielding the excess mortality in Germany for the pandemic years 2020 to 2022.
> In 2020, the observed number of deaths was close to the expected number with respect to the empirical standard deviation. By contrast, in 2021, the observed number of deaths was two empirical standard deviations above the expected number. The high excess mortality in 2021 was almost entirely due to an increase in deaths in the age groups between 15 and 79 and started to accumulate only from April 2021 onwards. A similar mortality pattern was observed for stillbirths with an increase of about 11 percent in the second quarter of the year 2021.
> Something must have happened in April 2021 that led to a sudden and sustained increase in mortality in the age groups below 80 years, although no such effects on mortality had been observed during the COVID-19 pandemic so far.
> Our analysis follows four key steps. First, for all locations where weekly or monthly all-cause mortality has been reported since the start of the pandemic, we estimate how much mortality increased compared to the expected death rate....Second, based on a range of studies and consideration of other evidence, we estimate the fraction of excess mortality that is from total COVID-19 deaths as opposed to the five other drivers that influence excess mortality. Third, we build a statistical model that predicts the weekly ratio of total COVID-19 deaths to reported COVID-19 deaths based on covariates and spatial effects. Fourth, we use this statistical relationship to predict the ratio of total to reported COVID-19 deaths in places without data on total COVID-19 deaths and then multiply the reported COVID-19 deaths by this ratio to generate estimates of total COVID-19 deaths for all locations.
And, what, you may ask, are the biases of the models' authors with regards to the percentage of excess mortality due to Covid itself? Excellent question:
> Deaths that are directly due to COVID-19 are likely underreported in many locations, particularly in settings where COVID-19 testing is in short supply. Most excess mortality is likely misclassified COVID-19 deaths.
> Given that there is insufficient evidence to estimate these contributions to excess mortality, for now we assume that total COVID-19 deaths equal excess mortality. For the reasons presented in this section, we believe that this is likely an underestimate.
As someone who has now spent the better part of his life creating statistical models, when you assume something is true, your model is likely to confirm your assumptions.
This is why statistical models are worthless without prospective validation. If you haven't shown that your assumptions are correct for the future, you're just making castles in the sky.
reply