Looking at Worldometers 7-day averages from early on when cases were still low and building, the highest early death counts were in mid-April 2020, with 7-day daily death counts peaking at
2,280 on April 18.
Picking a number of days to work backwards is necessarily arbitrary, but 14-17 days is a good rule of thumb to estimate how long after infection a COVID death took place. Go back 14 days to 4/4/2022, and the 7-day case count was around 27,967. Go back 17 days to 4/1/2020 and the 7-day case count was 22,652.
For our purposes here, the difference between the denominators 27,967 and 22,652 will yield significantly different results. Let's split the difference and go with
25,309.
Do we want to account for speculation that cases were undercounted in April 2020? Remember, this is way before any kind of home testing was done. Unfortunately, precisely what percentage of cases were missed then is an unknown. However, some approximations have been run --
here's one Columbia University released early in 2021. The Columbia one is especially helpful because it reaches back to the beginning of the pandemic. They posit that there were about 200,000 - 225,000 actual cases in early April 2020 (
hard to pinpoint it at the graph's resolution). Again, let's split that difference and say there were about
212,500 actual cases per day in early April, and adopt that figure as our denominator. That means we took our split-the-difference
detected 7-day case count of 25,309 and multiplied by
8.396 to get 212,500.
Let's calculate our mid-April 2020 death rate -- 2,280/212,500 = just a smidge over 1%, 0.01073,
1073 out of 100,000.
Now. In spring/summer 2022, many researchers are estimating that detected daily case counts need to be multiplied 7-10 times to yield something close to a true daily case count. Well, how about that? Our 8.396 true-case multiplier from above almost exactly splits the difference between 7 and 10. So let's employ 8.396 again.
7-day cases one week ago (see my post above) was 131,917. 131,917 * 8.396 =
1,107,575 (rounded down from 1,107,575.132). Taking mid-July 2022's 7-day case rate of 1.1 million and applying the mid-April 2020 death rate would yield:
1,107,575 * 0.01073 = 11,884 deaths per day (rounded from 11,883.628)
That's a reasonable answer to
Leeroy's proposition "
the same death rates today as pre-vaccine with today’s numbers of infection" considering what we know and what we can reasonably estimate. Around 11,884 deaths per day.
The actual 7-day death counts from last week topped out at
393 on 7/13. They will likely rise to over 400 over the next few days as the numbers get adjusted. But for now, we can say that the death rate in April 2020 was about 30 times the death rate today. Not sliced and diced by age or comorbidities or anything like that. Just everything all lumped together and averaged out.