How many people are expected to die in America this year? If you go to popular sites, you will find something like this.
These are observed deaths with the dotted line a very crude guess for expected deaths. Observed - expected deaths = excess mortality, and that is what everyone is so concerned about these days. Two vocal poles frame the discussion; one side thinks COVID causes all the deaths, and the other side, thinks vaccines do.
To understand excess mortality, first know it is typically age adjusted. In other words, a 10 year old might have a 1 year risk of death of 0.000097%, and an 88 year old 13.1%. So knowing people’s ages is essential to know expected death rates.
Each year, as the American population ages (not as many babies as the boom) we recalculate how many 1, 2, 3, 4, etc etc. year olds we have and their 1 year risk of death, and then we have an expected risk of death for the population. We can then compare this to the observed death for that year.
For a simple case, let’s just imagine 1000, 65 year olds. Say 14 died in 2017, 16 died in 2018, and 15 died in 2019. Now 2020 and 2021 come along and we disrupt all of life and COVID comes, and 25 and 24 die. Let’s say 22 died in 2022. How many are EXPECTED to die in 2023?
…Pause and make your guess….
If you go back and avg (14,16 and 15) that isn’t quite right is it? Extra people who were 62, 63, and 64 back then (and were supposed to be 65 next year) also died in the last 2 years, and these are not deaths at random. Neither are they the most sick people. They had to be both vulnerable and in networks that got COVID. These people are removed. So what should the new expected death rate be among 1000, 65 year olds who turn 65 this year?
Hmmm. Should it be 13 or 14 because the surviving 65 year olds now are healthier than prior years? or should it be 17 or 18 because a few very vulnerable people didn’t get COVID but otherwise took it very easy, hiding at home, and now are more likely to die? Which is it? Or should it be something else entirely?
Now let’s say that we observe 24 deaths this year out of that 1000. Having picked your expected number of deaths, is that a surplus? Now, what is the cause, COVID, or sequalae from lockdown, or some iatrogenic injury (as many on the internet fear— the vax!!!)? And how would you know.
What is my point?
Before getting into endless debates about how to interpret numbers, we have to think about where the numbers come from. What was directly measured, and what is a guess.
Expected death is a guess. What should it be after 2 seasons of above avg deaths and completely different ways of life? Who is the person turning 65 today, and are they the same as the person who turned 65 in 2017,18,19.
It is easy to believe those groups are largely similar as nothing major happened in those years, and observed deaths changed just slightly (likely as the population aged), but hard to know how the surviving 2023 65 year old compares to them.
And then, how would one know what is driving differences? …
Keep reading with a 7-day free trial
Subscribe to Vinay Prasad's Observations and Thoughts to keep reading this post and get 7 days of free access to the full post archives.