Post #1344: COVID-19 trend to 12/10/2021

Posted on December 11, 2021

The winter wave flattened out this past week.  The Northeast continued to see growth in new cases, but not the Midwest.

Increasingly, what we have is part of a winter wave.  The Northeast is repeating last year pretty closely:

But other states that were major contributor’s to last year’s wave are not having a winter wave this year.  Here’s California, home to about 10% of the U.S. population:

Last year at this time, California was above 100 new cases / 100K/ day.  Now they are around 20.

And, at some point, winter’s going to be over.  Last year, the winter wave peaked on January 13.  If this year were a repeat, that would only leave just over four weeks in the current winter wave.  This year’s wave started late, owing (I think) to generally warmer weather.  But if this year is anything like a repeat of last year, that still doesn’t leave a whole lot of time.


How all this will shift with Omicron remains to be seen.  Based on my last post, my best guess is that we will have many new cases but few hospitalizations or deaths. 

All we can do at present is look at what’s happening in South Africa.  South Africa’s hospital data and official deaths data are not ideally suited to tracking COVID-19 severity.  Their hospitalizations counts are mainly picking up cases where COVID-19 was incidental (not the cause of admission).  And their official death tolls are thought to understate the true death counts considerably.

That said, all I really need is for them to track cases and deaths consistently over time.  That will let us compare (e.g.) the relative case mortality of Delta (3% in South African data) against Omicron (which seems to be much less, based on the calculations in my last post).

My point is pretty simple.  Per the official data, day after day, we’re seeing 10,000 to 20,000 new COVID-19 cases reported.  And just a handful of COVID-19 deaths.  If that goes on for another week or two, either the South African data are just total nonsense, or somebody has to take note of it.

Nerd that I am, I have bothered to track down the details of the South African death certificate process.  You need one in order to bury somebody, so it sure seems like there isn’t going to be some sort of huge lag involved.

Below, this is what I’m talking about.  This, from the WHO:  See the big empty circle?  Either South Africa just isn’t reporting those deaths, or they take a long time to report those deaths (not evident in historical data), or those deaths aren’t occurring.

Source:  WHO.

I’m so much of a nerd, I even went and checked earlier version of this graph via the Wayback Machine.  Historically, there was no long lag between the case data and the deaths data.

Source:  Wayback Machine.  Red notes are mine.

Basically, those deaths aren’t showing up.  And if they are occurring, they now should start to appear in the data.  I think I’ve ruled out any sort of systematic data reporting issue with South African deaths data.  Historically, their data reporting lag looked no longer than that of the U.S.

Sooner or later, people are going to have to notice that those deaths aren’t there.

And that’s why I say, I expect to see a lot of cases, but not a lot of hospitalizations or deaths.