Post #1503: COVID-19 trend to 5/5/2022: Double digit increases everywhere.

Posted on May 6, 2022

 

Now 20.4 new cases per 100K per day, up from 19.4 yesterday.  Still rising about 25 percent per week.

And, at this point, there are double-digit weekly increases across all of the six geographic regions that I have used to track this pandemic.

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 5/6/2022, from https://github.com/nytimes/covid-19-data.”  The NY Times U.S. tracking page may be found at https://www.nytimes.com/interactive/2020/us/coronavirus-us-cases.html

 

The implication is that while this second Omicron wave or Omicron sub-wave really hasn’t hit the middle of the country yet (as shown yesterday),  that may now be starting to change.

 

This latest wave — presumably the B.2.12.1 wave, which clearly needs a snappier name if it’s ever going to catch on in the media — is well-organized in the Northeast and Midwest, with nearly all states lining up on a common growth trend.

(N.B., the “blocky” nature of the lines is due to the sporadic reporting of data.  My default is to continue the last true “reading” of the numbers until new data arrive.  The result is a bunch of plateaus (no new data) followed by jumps (fresh data).  There’s no doubt some smarter way to fill in those gaps.  But with enough time, you arrive at the same place whether you got there by climbing a ramp or a set of stairs.)

For the remaining regions, the average is up, but there no such homogeneity across the states.

I’m not sure there’s any clear interpretation of that, but casually, by eye, that appears to line up with when these regions’ case counts began to climb.  As they pass the low point, they slopes of the lines are mixed.  Once they’ve been climbing for a few weeks, it seems as if all states line up with a common growth rate.

If that’s true, then give it a couple more weeks, and all the graphs should be as tidy as the first two.

A note on mask use

Case rates are going up, but mask use is coming down.

And, as irrational as that may seem, that’s completely normal.  Throughout the pandemic, mask use has been reactive.  It has followed the peaks and troughs of new case counts, by, oh, about a month, on average.  This seems to be the average time it takes for the change in risk-of-infection to make it into the collective American consciousness.

Source:  Mask use from Carnegie-Mellon COVIDcast.  Case counts from NY Times data cited above.

Just to give one example, at the start of the U.S. Delta wave (second arrow from the left, above), cases (orange line) began rising around 6/24/2022.   Mask use (blue line) did not begin rising until 7/22/2022, or roughly one month later.  This, despite Delta having been advertised as resulting in more severe infections, well in advance of it becoming the dominant U.S. strain.

For the current B.2.12.1 wave, U.S. cases began rising around 4/10/2022.  If history repeats itself, we ought to see that mask use curve now flatten and turn upward.

Alternatively, what with all the vaccination and boosters, and all the prior infections, and the anti-mask-mandate court victories, maybe this time will be different.  Maybe we’ll now see continued declining COVID-19 hygiene in the face of rising new-case counts.  That could be interesting.

A brief note on hospitalizations.

As of now, the CDC COVID data tracker runs about 5 days behind the raw data, in tracking hospitalizations.   The raw DHHS unified hospital data set is current through 5/5 (as of 10 AM 5/6/2020), the COVID data tracker only lists hospitalizations through (effectively) about 5/1/2020.  (Nominally, it shows data through 5/3, but for some reason, it subtracts two days from the reporting date as shown on the raw file).

In part, those lags are reasonable.  The data tend to be somewhat under-reported for the last two days of reporting.  And the final date on the file typically just replicates the data from the next-to-last date (for reasons I cannot even begin to fathom).  So, by waiting a few days, the CDC doesn’t have to revise their numbers constantly as the raw data file gets updated every day.

In addition, the raw data show a weekly pattern due to variation in completeness of data reporting over the course of a week.  Small hospitals are generally excused from daily reporting of new cases over the weekend, and they then dump three days’ worth of new cases into their Monday reporting.

Anyway, FWIW, I downloaded the file (referenced above) and got something of a surprise.  Best guess, all things considered, they raw data series, when it is reported by the CDC, will appear to have gone from the current 2200 new hospitalizations per day, to the eye-candy level of 3000 new hospitalizations per day.  It’s not quite there, on this chart, but it should be at that level by the time the CDC gets around to reporting data that are nominally for 5/5/2022.

Source:  Analysis of DHHS unified hospital data set

My only points are that a) the hospitalization rate appears to be picking up speed, to match the new case rate, and b) that’s probably going to get reported as some sort of “jump” in hospitalizations, which is not the case.  If you smooth out the weekly variation in data reporting (orange line above), hospitalizations are picking up smoothly, and appear to be ramping up to match the rate of daily new cases.

Business-as-usual makes for boring news.  But there’s a reason that it’s called “as usual”.

Addendum:

B.2.12.1 is just soooo boring.  As is, say, Omicron sub-wave,  No wonder nobody is taking this seriously.  If you can’t even put a name on it, discussion is difficult.

Omicronic wave (or -chronic) wave?

Omicron microwave?

New-micron wave?

Omicron, Part Deux?

Omicron speed bump?

Omicron after-wave?

Omicron rebound wave?