Having trouble finding a particular statistic about the US death rate

[Silent3]

According to this web site: https://www.cdc.gov/mmwr/volumes/70/wr/mm7014e1.htm

There were 3,358,814 deaths in the US in 2020, 377,883 attributed to COVID. This was a death rate of 828.7 people per 100,000, or 0.8287%.

But a death rate presented like this is offset by population growth, which, even if low (about 0.5%) isn't insignificant in this kind of statistic.

What I'd like to know is this: Out of 100,000 people living in the US at the start of a year, who aren't born during that year, don't leave the country, and don't move into the country, what number can be expected to die by the end of that year?

Since average life span is less than 100 years, this death rate has to be over 1%. What value over 1%? That's what I can't locate.

[PoindexterOglethorpe]

1. This link may be helpful.

https://247wallst.com/special-report/2020/07/29/how-many-people-are-alive-from-the-year-you-were-born/

As it says, you can see year by year, starting in 1936, how many people are still alive as of 2020. It's not quite what you're asking, but I have found it fascinating. Most people live well into their senior years.

[kcr]

3. That was interesting

A rare I-don't-feel-so-old moment on the internet. Over 90% of people my age are still alive

[muriel_volestrangler]

2. That figure, and the total population at the beginning of the year, gives you roughly what you want

the U.S. Census Bureau projects the U.S. population will be 330,222,422 on Jan. 1, 2020 (projection made on Dec 30th, 2019, so we'll use it as accurate)

Yes, a few people will move into the country during 2020 and also die in 2020, but very few. I think it's reasonable to offset that with those who move out during 2020 and also die during 2020.

So 3,358,814 died per 330,222,422, or 1,017 per 100,000. That's different from the 828.7 per 100,000 (not per million) the CDC gives, because the CDC's number is "age-adjusted" - and the method for that adjustment is no doubt complicated (and the main purpose for the adjustment will determine that - is it to allow reasonable international comparisons, comparisons in the US over time, comparisons between demographic groups in the US, or is there a method that satisfies all of those?)

So the value you want is the Crude Death Rate - available from the CDC here: https://www.cdc.gov/nchs/nvss/vsrr/mortality-dashboard.htm

Their latest figure measured over 12 months was 980.9, from Oct 19-Sept 20; that's up from 827.3 for Oct 18-Sept 19.

So it is about 1%, now. It's not significantly over 1%, as a simple "average life span is under 100, eg 82" might suggest, because the US population is still growing - younger generations are larger than older ones, so the number of people at greatest risk of dying is smaller in proportion. So, for instance, Japan, with a well-known long life expectancy, has a crude death rate in these estimates of 11.44 per 1000, UK 9.41, and USA 8.35 (I suspect the CIA numbers have not taken Covid into account for their 2021 estimates). Those numbers indicate Japan's population is on average older than the UK's, which is older than the USA's.

[Silent3]

4. Thanks. This is definitely closer to what I'm looking for...

...but I was hoping to see something that also removes the effect of population growth, a statistic that extrapolates what the death rate would be in with steady-state population level.

I went back and fixed where I'd typed "million" instead of 100,000 too.

[muriel_volestrangler]

5. I *think* what you'd want is the simple reciprocal of life expectancy at birth

because that works out the chances of a fixed population, using current figures, surviving to 1, and then 2, and then 3 and so on.

That would mean a death rate of 1,269 per 100,000 in 2019 (from life expectancy of 78.8), and 1,285 in the first half of 2020 (from 77.8).

I *think* that would imply a steady-state population - you get exactly the same number of people born each year, and the same dying at age 0, 1, 2 etc. all the way up. But we'd need a population statistician to know.

[Silent3]

6. That makes sense

I figured it was a number in that range I was looking for, but I wasn't sure if the calculation would be that simple.