Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Estimation of total mortality due to COVID-19

healthdata.org

Post reply on HN
Print view View original

321–330 of 387 posts

Re: Estimation of total mortality due to COVID-19

#321

Post by kens »

lamontcg wrote: post #262
timr wrote: post #20

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 IHME has been wildly wrong yet highly cited for over a year now. I've still got screencaps of their models that predict around 90,000 total deaths in the USA by the end of 2020. They were lifting a sigmoid curve off of China's numbers and applying it to the United States entirely blindly. It didn't even predict the shape of the decline off the first peak correctly since it was implicitly assuming the US would tak…

Yes, I don't understand why the IHME is still taken seriously after their awful models a year ago where they were repeatedly, confidently wrong.

Link predicting deaths drop to 0 by June 21, 2020 (i.e. last year). Spoiler: they did not. https://ktla.com/news/nationworld/influential-ihme-model-pro... The original paper https://www.medrxiv.org/content/10.1101/2020.03.27.20043752v...

Re: Estimation of total mortality due to COVID-19

#322

Post by pmoriarty »

Compare 900k dead from COVID-19 (so far) to U.S. deaths in:

- Vietnam[1]: 58 k

- World War Two[2]: 407 k

- World War One[3]: 117 k

and due to the huge numbers of people in the U.S. who are refusing to get vaccinated, wear masks, or participate in social distancing, the U.S. deaths from COVID-19 are probably far from over.

[1] - https://en.wikipedia.org/wiki/Vietnam_war

[2] - https://en.wikipedia.org/wiki/World_War_II_casualties

[3] - https://en.wikipedia.org/wiki/World_war_one

Re: Estimation of total mortality due to COVID-19

#323

Post by timr »

spekcular wrote: post #304
timr wrote: post #258

Earlier quoted context omitted.

Crude methods are fine. Fantastic models are made with linear regression. No matter how crude your methods, if you don't validate your model prospectively, it is a mathematical fairy tale.

Maybe I'm just ignorant of the relevant statistical methods, but how do you suggest models be validated prospectively ? All the model validation techniques I'm aware of come after the model is fit on the data (or a subset of it), like testing against a holdout set, or k-fold cross-validation, or checking R^2.

> testing against a holdout set

Not just a holdout set -- a holdout set of prospective data, where you actually predict the future, blindly, and see how you do.

Obviously, you have to wait a little while to gather such data before you make big claims with your model. Doesn't mean you get to skip it.

It's not even clear to me that the authors did a cross-validation here, or even bothered to fit the free parameters of their model based on any kind of empirical data at all. At least with regard to percentage of excess deaths due to Covid, I'm not sure how they could...the parameters seem to be unrelated to any empirical data.

Re: Estimation of total mortality due to COVID-19

#324

Post by sweezyjeezy »

klmadfejno wrote: post #271
sweezyjeezy wrote: post #50

Earlier quoted context omitted.

I think it's kind of unlucky in some ways that Covid affected the old/unhealthy so disproportionately. I feel like we would have responded in a much more assertive and effective way had that not been the case. More people who were not in the risk groups would (ironically) have been saved, and we'd be a lot further on the road to recovery. Not a lot of young people know a huge amount of people in the risk groups, it's…

We would have reacted more assertively because that would have been a much bigger problem. We're lucky it was this way.

I didn't necessarily mean a bigger death count, just a more equally distributed one (like the Spanish flu). I think the end outcome could have been better for all demographics, and the world would be further on the road to recovery.

With Covid it was easy people to think it wasn't really a big deal, it reduced the appetite and compliance necessary to keep the pandemic contained.

Re: Estimation of total mortality due to COVID-19

#326

Post by fastball »

TameAntelope wrote: post #320
fastball wrote: post #307

Earlier quoted context omitted.

The problem is we don't know , and "both sides" don't seem willing to admit that. What I can guarantee is that of those 95% who died with COVID, some percentage of them would have died anyway, under almost identical circumstances, except without COVID. However, it is also true that some of those people would have lived if they hadn't contracted COVID, even though COVID was not the "direct" cause of death. For how muc…

We can use deaths-above-replacement (among other things) to determine how many more people died than normal from all causes, and then whittle down the obviously-not-from-contracting-covid deaths from there. The submission goes into very specific detail about how they did this, and they account for many of the potential confounders related to doing this analysis. I believe we can be much more precise than you're givin…

It is of immense value to say this when you have people in this very thread saying we might as well count everyone who had COVID as a death from COVID.

Also one of us is confused, and I don't think it is me. From what I've gleaned from this article, the 900,000 figure is not (as you say) "determining if contracting COVID was the direct cause of someone's death", but rather trying to figure out how many people would have died if this had been a "normal" year, i.e. if we could go to a parallel universe and make COVID never exist. This is absolutely not the same thing, as such a scenario will also count deaths as a result of our response to the virus as a death "due to the virus". This is my impression as they seem to be including things like suicide (criteria "c") and people who died because people with actual COVID were taking up hospital beds (criteria "b").

Re: Estimation of total mortality due to COVID-19

#327

Post by fastball »

newacct583 wrote: post #309
fastball wrote: post #137

Earlier quoted context omitted.

That's... exactly the point? Dying with COVID does not mean you would be alive today if you hadn't contracted COVID.

Exactly! So rather than try to puzzle out this undecidable proposition for nearly a million individual deaths, let's just look at aggregate statistics. Did more people die on the whole during the pandemic than we would normally expect? Does that line up with a model that they probably died from covid? And... that's exactly what the linked article is doing. And yeah, it turns out that all these people (on balance) wou…

The linked article seems to be including suicide as a death resulting from COVID, when the increase in suicides is far more likely to be caused by our response to COVID, not having COVID.

It is definitely not only looking at whether people would have died sans covid in their system.

Re: Estimation of total mortality due to COVID-19

#328

Post by xkjkls »

fastball wrote: post #307
xkjkls wrote: post #158

Earlier quoted context omitted.

That's what he is attempting to imply is a massive overcounting in the data, but additional conditions that people point out as COVID deaths also having are mostly caused by COVID-19. Oh, they died of viral pneumonia after contracting COVID-19; pretty sure they would be in a different scenario had they not caught COVID-19.

The problem is we don't know , and "both sides" don't seem willing to admit that. What I can guarantee is that of those 95% who died with COVID, some percentage of them would have died anyway, under almost identical circumstances, except without COVID. However, it is also true that some of those people would have lived if they hadn't contracted COVID, even though COVID was not the "direct" cause of death. For how muc…

Just because we don't know exactly doesn't mean we aren't able to create good estimates. We don't know how many people die of any disease. But if you have a large enough dataset, and analyze it well, you can make good determinations of the approximate scale of each factor. The paper linked in this article does that, and concludes approximately 900,000 died of COVID who would not otherwise.

Re: Estimation of total mortality due to COVID-19

#329

Post by fastball »

xkjkls wrote: post #328
fastball wrote: post #307

Earlier quoted context omitted.

The problem is we don't know , and "both sides" don't seem willing to admit that. What I can guarantee is that of those 95% who died with COVID, some percentage of them would have died anyway, under almost identical circumstances, except without COVID. However, it is also true that some of those people would have lived if they hadn't contracted COVID, even though COVID was not the "direct" cause of death. For how muc…

Just because we don't know exactly doesn't mean we aren't able to create good estimates. We don't know how many people die of any disease. But if you have a large enough dataset, and analyze it well, you can make good determinations of the approximate scale of each factor. The paper linked in this article does that, and concludes approximately 900,000 died of COVID who would not otherwise.

Two different ideas are being conflated here.

One is the OP's idea, "deaths due to COVID". This idea is "how many people would have died if COVID had never existed". However, what a lot of people hear when you say "deaths due to COVID" is "this is the number of people that died as either a direct or indirect result of having COVID in their system". But that is not the 900,000 number being cited.

It is made even worse by the HN headline, which is "death toll from COVID", which even more strongly implies the above.

Re: Estimation of total mortality due to COVID-19

#330

Post by jollybean »

nonameiguess wrote: post #218
jollybean wrote: post #186

Earlier quoted context omitted.

It's also greater than all of US military combat deaths for every war combined i.e. Civil War, WW1/2, Korea, Vietnam etc.. And think of how much concern (legitimately so) we have for them.

US Civil War death toll is estimated between 620,000 and 750,000, which is higher than the official Covid toll so far. It was also much more highly concentrated in a small number of states in a country that was far less populated.

That's not 'combat deaths' though, that's 'total deaths'.

In those wars, a huge number of people died from things other than direct combat.

Here are the numbers [1]

[1] https://en.wikipedia.org/wiki/United_States_military_casualt...

Post reply on HN