Post #1361: COVID-19 trend to 12/20/2021, looks like the U.S. Omicron wave is here.

Posted on December 21, 2021

New COVID-19 cases are now spiking in some areas of the U.S.

Before I get into all that, perhaps a bit of a calming exercise is in order. 

Focus on some happy thoughts

More cases, fewer hospitalizations, in the U.S. so far.  I’m going to start with this chart, from CDC, showing new COVID-19 cases and hospitalizations through 12/19/2021.  Note that even as cases are rising, total hospitalizations are falling. 

Source:  CDC COVID data tracker, accessed 12/21/2021

Sure, that could be a fluke.  On the other hand, I’ve plainly said that I expect our Omicron wave to be short, sharp, and of vastly lower average case severity than Delta.  If, as it appears to me, Omicron has a vastly lower hospitalization case rate than Delta, then this is what we should be seeing, as Omicron displaces Delta.  So maybe it’s a fluke, but it’s what I’ve been waiting to see.

As a footnote, continue to keep in mind that most U.S. hospital statistics are patients admitted with COVID-19 (regardless of whether the COVID-19 is clinically relevant to the admission), not patients admitted for COVID-19 (Post #1349).  Up to now, that hasn’t mattered.  But to the extent that Omicron generates a lot of asymptomatic cases, we may end up in the same situation as parts of South Africa, where three-quarters of hospitalizations were “with” not “for” COVID-19.  As I noted in that earlier post, for the U.S. data cited above, it’s ambiguous.  That particular statistic is something akin to a count of cases with COVID-19 present on admission.  If hospitals apply rapid (antigen) tests at time of admission and discover asymptomatic COVID-19 cases on the day of admission, we’ll be in the same situation as South Africa.  We’ll end up counting a bunch of COVID-19 hospitalizations that have nothing to do with the actual treatment of COVID.

Extremely few deaths in the South African Omicron wave.  Deaths only show up with some time lag, compared to cases.  But not with an infinite time lag.  Although all the deaths from this wave have not yet been counted, at this point, we can be pretty sure that the final case mortality rate for Omicron in South Africa was a tiny fraction of the case mortality rate for Delta.

Source:  WHO COVID-19 dashboard for South Africa, accessed 12/21/2021

Rapid end of the South African Omicron wave. A second graph worth focusing on a bit is the one showing the rapid end of the South African Omicron wave.  In the past five days, their daily new case count fell from 25,000, to 8,000.

Source:  Just Google South African COVID

The current U.S. trend

Data source for this and other graphs of new case counts:  Calculated from The New York Times. (2021). Coronavirus (Covid-19) Data in the United States. Retrieved 12/21/2021, from”  The NY Times U.S. tracking page may be found at

There has been a sharp increase in daily new cases over the past few days.  And that’s starting to be more broadly-based than just the Northeast.

This past week, the South Atlantic region firmly joined the U.S. winter wave.  Prior to this, there had been rising cases only in the northern part of that region (e.g., Delaware, DC).  But this past week, daily new cases in Florida more than tripled, from 9.5 to about 35 new cases per day.  Because Florida is such a populous state, that pretty much determines where the South Atlantic average is headed.

That said, the extremely sharp spike in the South Atlantic region above is due in part to a data issue in Maryland.  Per the New York Times:  Dec. 5, 2021 to Dec. 20, 2021: Maryland reported many cases on Dec. 20 after resolving technical issues that prevented the state from reporting updates for more than two weeks.

Here’s how it looks from the perspective of this year, versus last year.  But don’t let the graph fool you.  Last year, the steep “ups” were mostly recoveries from data reporting outages over the holidays.  (I.e., they weren’t “real”, in the sense of steep increase in actual new cases).  Right now, by contrast, that’s a real increase in cases.  

In short, this last week seems materially different from what we saw last year at this time. We never saw this fast of a true increase in cases.  Couple that with the CDC’s upwardly-revised estimated of Omicron, and it’s a good guess that this is, in fact, the start of the U.S. Omicron wave.

How does that timing work out vis-a-vis Great Britain?  Our first case (12/1/2021) was only about four days after Great Britain’s first case (11/27/2021).  By eye, let me mark the start of the U.S. Omicron wave as 12/17/2021.  The curve for Great Britain kinked upward on or about 12/13/2021.

So yes, that’s just about right, for the timing of it.  In hindsight, the start of our Omicron wave is about four days after the start of the British Omicron wave.

Well, it’s not a surprise.  But it’s still a shock.

An important technical note on the shape of the British versus South African Omicron waves.

This is a brief note on the shape and duration of the Omicron wave in South Africa versus Great Britain.  It matters, because if we’re going to guess the near-term course of the pandemic in the U.S., based on what Omicron did in those countries, we have to be clear about what we’re looking at.

South Africa’s wave took about 3.5 weeks, from first case to peak.  Because they had almost no cases just prior to that, we could see the entire wave.  It fully looked like a wave that was 3.5 weeks long.  And seems to be ending abruptly.

In Great Britain, by contrast, they were in the midst of an ongoing wave of Delta.  They now appear to have (maybe) reached a peak.  That peak (if it is a peak) is occurring three weeks after their first case (11/27/2021), similar to the South African experience.  And for whatever reason, it does not appear to be a sharp peak as in South Africa.

But, by eye, looking at the graphs, we only really see the final week of that wave, as the rapid growth of Omicron finally peeked out from beneath the pre-existing upward trend in Delta.  So what looks like a “one week Omicron wave” in Great Britain really isn’t.  It’s the same three-plus weeks that they saw in South Africa.  It’s just that the first two couldn’t be distinguished from the ongoing wave of Delta.

Will the same apply to the U.S.?  Will we actually reach our peak or plateau just  few days from now?  No way to be sure.  And, in fact, we won’t even be able to tell.  See next section.

Blind for the Holidays!

We should expect to have at least a week of rapid growth of this sort, based on what occurred in Great Britain.  Possibly more, as we’re considerably more spread out than they are.

But, we’re not going to see that.  And that’s because the Christmas and New Year’s holidays are going to interrupt our data reporting, as they did last year.

Based on what happened last year, the last reliable data point we’ll get will be Thursday December 23rd.  And then it’ll be, well, early 2022 before we get another good fix on the actual trend in new cases.  And if this is the start of our wave, by that time, the wave will be well over with.

For that period, I’m going to start paying a lot more attention to the hospitalization series above, and less to what will probably be a declining count of new cases.  The hospitalizations appear to be counted reliably regardless of holidays.

I’ll do a separate post on the CDC’s revised Omicron numbers, next.