Yes, I know that article makes multiple claims about excess deaths. I cited it to show you that the claim about Florida pneumonia deaths isn't real, it's just a conspiracy theory. I'm not claiming everything published by the Tampa Bay Times is correct. I think that point landed, as you didn't mention it again.
Now the other claim made in that article, the one you're now citing, it isn't logically valid. It makes the following inference:
1. If more people die on average during the past weeks, the cause of death must be CV.
2. Therefore if the death count of CV doesn't match excess deaths, there's been undercounting.
It's invalid because it assumes lockdowns have no effect on mortality. This is a very strange assumption to make, but it must be widespread because I've seen people here on HN argue that "the economy" is some optional thing that shouldn't be weighed up against people's lives, for example.
In practice the economy is what keeps us alive. It's not just restaurants and movies. Damaging it hurts or kills people. We'd absolutely expect there to be excess deaths from a lockdown both during and after it ends, and there's plenty of evidence of that happening.
During a lockdown access to healthcare services become restricted or people became afraid to go to the hospitals, or even afraid to go outside. In the UK cardiac admissions to the emergency rooms halved overnight when the government started telling people to avoid hospitals and cancelled all "optional" operations (not really optional of course, just delayable with some risk). Some people pulled their own teeth out because emergency dentistry was closed.
And in fairness to the article, it does say that - in the very next sentence after the part you quote:
> Health experts say that likely includes some people who died of coronavirus but were never diagnosed as well as others who might have lived had the pandemic not kept them from getting care.
So basically they don't know.
> Please explain how covid death tolls are hugely inflated if an analysis that entirely ignores cause of death, and only focuses on population level statistics, shows that there's been an unexplained increase in death rate beyond what has been attributed to COVID
These are two related but different arguments.
COVID is a real virus that has increased the excess death rate, absolutely, although not really more than a bad flu season. "COVID deaths" are a count of all attributed deaths, not just the excess over the average.
Imagine if every single death was reported as a COVID death. Obviously everyone would know that isn't true because deaths of other causes don't stop because there's a virus around.
Now imagine that 80% of all deaths suddenly become reported as "died of COVID" whilst excess deaths went up by 10%. People would still call shenanigans because it'd be obvious that this couldn't be the case.
But this can be the case if the report is actually "died WITH COVID". Then suddenly it's possible for 80% of people to die with it, and also for excess deaths to only go up by 10%. In practice all numbers reported as "died of" are in reality "died with" so this is a real issue.
Consider a theoretical virus called disease X. It's highly infectious and infects 50% of all people, but actually causes no symptoms and kills nobody at all. Its mortality rate is zero. Imagine that a few powerful people didn't know that and believed otherwise, and mandated that all deaths where the virus was detected via some clever test be reported to them centrally, and those reports went out as "died of disease X". In such a situation the virus would act as effectively a random labelling function. People would just get this label in their bloodstream but it wouldn't actually affect them. Then it would be true that the death toll of disease X was hugely over-inflated, as it'd be zero in reality but via a chain of reporting errors would be reported as 50% of all deaths simply because it was present at the time of death.
Finally, reconsider the above argument, but where 50% is adjusted to be lower than the excess death rate. The argument doesn't actually change at all: it's invariant to how many people get labelled. Any reports of death of disease X are an over-inflation, even if those numbers are smaller than the excess death count.
Now, COVID isn't disease X. It does cause symptoms occasionally, in very rare cases it can cause death, usually people with very weak immune systems. But there's a whole lot of very compelling data showing it's pretty close to disease X. It labels a lot of people as "infected with COVID" even though it's not really affecting them much or at all. In that case, given the way reporting standards have been implemented, a whole lot of people will be reported as COVID deaths when in reality they died of something else. And importantly that statement can still be true even if the total number of "died of/with COVID" reports is lower than the excess death rate, if there are other factors also killing people, like lack of access to health care and increased suicides.
To really figure out what happened with COVID will take a long time. It'll become clearer in retrospect, indeed it's already a lot clearer than when this all began.
> Your weekly reminder that while the life expectancy at birth is 78 years, the life expectancy of an 80 year old person is usually still 5-10 years. Statistics can be weird like that.
Interesting. Good point. I don't think it changes the argument above though.