Hacker News new | ask | show | jobs
by velcii 1778 days ago
Some convincing citations would be nice, if you can provide them..
2 comments

His number is based on a naive death count year over year. When the death rate is normalized for population age distribution his "excess" shrinks to near 0.
Could you expand on what you mean by "normalized for population age distribution"? Setting aside the fact that the study I'm citing is far from naive, I'm not sure how you can math away the simple fact that an extraordinary number of people have died over the past year, well in excess of prior years.

The graph on this page, for example, is quite stark (US only): https://www.cdc.gov/nchs/nvss/vsrr/covid19/excess_deaths.htm

Simply put, there's an extraordinary number of people becoming old at this exact time. The baby boomers.

Any "excess" death calculation which doesn't account for this natural spike in death is not comprehensive whatsoever.

It's a naive calculation because it doesn't include projections we can accurately make from understanding the current state of the population. It treats the population characteristics as a black box and only uses an extrapolation of past death events as a projection. We can also accurately predict death from birth.

You can't be serious. You're saying that the abrupt, almost overnight spike of all-cause mortality by 40% (!!) in April 2020 in the USA was... old age?

Did you look at the CDC graph? Do you see the very obvious wave of extra mortality, and how far it deviates from previous years? Do you see how the excess death comes in several huge waves, waves that perfectly line up with Covid-19 waves? Hell of a coincidence, don't you think?

I'd love to see this demographic analysis you cite. It must be a very sophisticated model, to perfectly predict a regular annual pattern (deviation <1%) for the past several years, abruptly followed by the wild 50% swings and chaos of the pandemic period, so accurately that you can confidently say that no one's died of Covid.

No, I’m not saying that. I’m saying your number is bad for a specific reason. I also never said no one’s died of covid.

If you wanted to see that data I’m sure you’d find it. Actuaries do this kind of work.

You said "When the death rate is normalized for population age distribution his "excess" shrinks to near 0". Then you said that baby boomers were getting old and the "excess death calculation... doesn't account for this natural spike in death". I cannot possibly read this in any other way than a claim that nearly all Covid-19 deaths are old age deaths.

I doubt I'll find this actuarial data, since it doesn't exist. Your proposition is ridiculous on its face - a cursory glance at the CDC graph is enough to see that. The swing in death rate is enormous and lines up perfectly with the Covid-19 waves. Actuarial data, no matter how comprehensive, doesn't say things like "40% more deaths in April". Look at the CDC graph. Look at it.

Actually, it's so ridiculous I have a hard time believing that you really believe it. Between the way you deny claiming things we can all see you claimed, and the way you've created that account just for this thread, I really think you must be trolling here.

Yes, thank you :)
Certainly. Here's the most comprehensive global analysis I could find:

http://www.healthdata.org/special-analysis/estimation-excess...

The discrepancy is of course smaller in more developed countries with better testing.