Post #1363: COVID-19 trend to 12/21/2021

Posted on December 22, 2021


First, a little international perspective.  Just a few graphs to calm the nerves.  Then the current U.S. carnage.

To cut to the chase, for the a whole, daily new cases increased by about 25 percent in the past four days. 


Before you go on to the U.S. data, just glance at a couple of recent developments that were in the news today.  The gist of both is that Omicron really does result in much lower severity of illness.  Before, we only had crude observational data to suggest that.  Now we’re getting the more nuanced and controlled analysis that can’t be easily dismissed with a few mumbles about vaccination rates and such.

First, South Africa has done an apples-to-apples comparison of cases with Omicron and cases with the prior variant.  Same population, same time period.  They just sorted them by type of COVID.  The hospitalization rate for persons with Omicron was 20% of the rate for other strains.

"Compellingly, together our data really suggest a positive story of a reduced severity of Omicron compared to other variants," said Professor Cheryl Cohen of the National Institute for Communicable Diseases (NICD), one of the study's authors."

Source:  This reporting, Reuters, published 12/22/2021.

I will point that the same figure — 20% — keeps popping up every time I have been able to calculate an on-the-fly number from the various European Omicron outbreaks.  As an example, see the top of Post #1359, 12/20/2021.

Second, the U.K. government’s official study of this issue also shows that Omicron results in milder case, on average (Politico, 12/22/2021).  But they won’t release that study until after Christmas, so we don’t have a number on how much less severe the average case is.


A bit of perspective on the duration of the Omicron wave:  South Africa, Great Britain, Norway, Denmark

These four countries appear to have been hit early with a significant surge in Omicron cases.  In some countries, there was no significant pre-existing winter wave, so the rise in cases is clear from the outset.  In others, the first week or two of increase in Omicron gets lost in the pre-existing Delta case load.

My point here is that many (but not all) of these Omicron waves are shaping up to follow a common pattern.  Sometime in the second or third week following the first case, new cases skyrocket and (ludicrously) dire warnings are issued.  And then, in most cases, so far, for reasons completely unknown, sometime in the fourth week — around 3.5 weeks after onset, say, on average — cases peak or plateau.  And that development somehow doesn’t get much press coverage.

South Africa: Peaked about three-and-a-half weeks after the first case was detected.  Still seems to have peaked (i.e., new case counts are still falling).  Almost no deaths were attributed to Omicron.  The case hospitalization rate appeared to be one-tenth of that in their prior Delta wave.  Not much more else to say.

Great Britain seems to have plateaued around three weeks after their first case. Great Britain detected its first Omicron case on 11/27/2021, putting them a few days ahead of the U.S.  They were already in the middle of a bad winter wave of Delta.  In the second week of their Omicron wave, new cases skyrocketed.  Daily new cases more or less doubled in a week.  At that time, you would have read headlines like these in the British press.  (N.B., Great Britain has a population of about 60M.)

Source:  The Guardian, 12/14/2021



Source:  The Guardian, 12/18/2021


That week of extremely rapid case growth has been followed by (nearly) a week of zero new case growth.  With another day of data, Great Britain’s Omicron outbreak it still looking as if it has plateaued. 

(N.B. 90,000/day in 60M population is a national average of 150 / 100K /day — almost twice the new case rate of the last year’s winter wave in the U.S.)

Norway seems to have plateaued around the third week following their first case.  Norway discovered its first cases of Omicron on or about 12/1/2021.  They immediately snapped into action, reimposing indoor mask mandates, speeding up booster shots of vaccines, and so on.  They had one superspreader event where 140 people got infected from a single celebration, but none required hospitalization.

And they too saw a rapid increase in Omicron cases and total cases.  To the point where their official government summary made the prediction below.  (N.B., Norway has a population of about 5M.)

Source:  Government of Norway, posted 12/14/2021.

Some time around the third week of what was, in fact, a pretty horrendous outbreak — new case counts plateaued.

(N.B. 4000/day in 5M population is a national average of 80/100K/day — about the same as the peak of our winter wave last year.)

Denmark has not plateaued yet, and has an extremely high rate of new cases per day.  (But only a modest rate of new hospitalizations.)  Denmark found its first two cases of Omicron circa 11/28/2021.  Like Norway, they saw rapid spread, aided by superspreader events at parties.  But unlike Norway, they haven’t plateaued yet:

(N.B., 10,000 cases per day in a population of 5.8M is about 170 new cases / 100K / day, or more than twice the peak our winter wave last year.)

That said, you ought to take some comfort in the contrast between these two graphs from Denmark.  The top is total cases, the bottom is a count of persons in the hospital.  Compare last winter’s wave to this winter’s wave.  The case hospitalization rate now is less than one-fifth if what it was historically.  (Which, now that the South African study is out, seems normal, and not at all a surprise.)

U.S. as of 12/21/2021

 In a nutshell, things look pretty grim, in terms of the rate of increase of total cases. 

But right now, the simple count of cases is a really stupid thing to focus on. 

I’ll do a separate post on that.   But briefly:  Up to now, the count of cases was, in some sense, a good proxy for total illness burden.  More cases implied more illness.  And every time a new variant replaced an old one, we found that it produced cases that were, on average, just about as severe.  So that produced some indicator of total disease burden from COVID that was consistent over time.

But now we’re getting hard evidence that Omicron cases have much lower severity, on average.  As we transition from Delta to Omicron, the simple count of cases no longer gives us a measure of illness burden that is consistent over time.  Or, to put it crudely, one Delta case is worth about five (or so) Omicron cases.  At least, in terms of likely impact on the hospital system.

That said, total cases is what everybody tracks, so who am I to be different.  At least for today.  And so:

Daily U.S. COVID-19 cases are up about 25% in the past four days.  A handful of states are going vertical.  Some others are seeing upturns.

But large areas of the country still seem mostly unaffected by Omicron.  Because the U.S. is so big, the three-and-a-half-weeks-and-done scenario above probably won’t play out here.  It’s like we’re the equivalent of several different European-sized countries, all starting our Omicron waves at staggered times.

So we may be seeing these steep growth rates for some time yet.

And now the data.  I’ve revised all this to start with the start of the U.S. winter wave.  And I’m not tabulating the last column as the total increase since 12/17/2021 — the point where the Omicron wave really took off in the U.S.

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 12/22/2021, from”  The NY Times U.S. tracking page may be found at

The Northeast continues along the steep upward path that it was already on.  Rhode Island is now getting close to 120 new cases / 100K / day.

But the big story has to be the rest of the U.S. East Coast, what I’ve termed the South Atlantic region.  As a whole, they were barely involved a week ago. Now they are a primary driver of growth.

That’s driven by three factors.  First, Florida is how showing rapid case growth.  Due to its large population, that matters greatly.  But both DC and MD are now showing near-vertical spikes in new cases.  For MD, that is due in part to a data reporting glitch.  In DC, that also looks like a glitch in the data, but no sources are reporting that.

I’m going to end with the graph of COVID-19 hospitalizations through 12/20/2021.  As I noted in an earlier post, this is going to be our only consistent indicator over the holidays, as both testing and reporting will drop off between (say) 12/23/2021 and somewhere in the first week of 2022.  Note that, despite the spike in cases, new hospitalizations are actually down a bit.

Source:  CDC COVID data tracker.