Post #1406: COVID-19 trend to 1/18/2022, post-peak.

Posted on January 19, 2022

 

Today almost every state reported data.  This makes the numbers somewhat less uncertain than they were yesterday.  The rate of new cases per 100K per day is unchanged over the past seven days.  I’ll mark the peak of the U.S. Omicron wave as 1/16/2022, almost exactly one month from the date on which new case growth began accelerating.

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 1/19/2022, 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

There’s a little dip in the line caused by the King Day holiday.  That said, the end of the line is now in the correct place.  And U.S. new case counts are clearly headed downward.  You’ll see a little spike in the numbers seven days from now, again caused by the shifting of reported cases from Monday to Tuesday following the holiday.

Among the states, New York and New Jersey are notable for a new case count that continues to plummet.  They are now down 40 to 45 percent in just over one week since they peaked.

Wisconsin is notable for the opposite, or maybe it’s just data reporting.  Cases are up 80% in the last week.  Wisconsin now exceeds 400 new cases / 100K / day, second only to Rhode Island.  But you have to temper that with reporting that the Wisconsin Department of Health has been dumping a large backlog of cases into the current case counts.  I have no way to sort out what fraction of the apparent surge in cases in Wisconsin is real.


One-tenth of the U.S. population had a COVID-19 infection in the last month.

If it seems like everybody you know was getting a COVID-19 infection last month, there’s a reason for that.  You can see the Omicron wave in perspective in the graph below.  This shows both of the pandemic years, starting on April 1 of each year.

Let me try to put the size of this into perspective.

On the up side of the Omicron wave, from start (12/17/2021) to peak (1/16/2022), the CDC recorded 15.6 million new U.S. COVID-19 infections, per the COVID data tracker.  Based on the most recent seroprevalence survey data (now through November 2021), the U.S. has averaged 2.13 actual infections for every one recorded.  Multiplying those together, I estimate that the true number of infections for the Omicron wave so far is about 33 million.  Or one-tenth of the U.S. population.

And this wave isn’t over yet.  If it’s roughly symmetric, we’ll see that many cases again on the downslope.  Suggesting that by the time the Omicron wave is finished, about one person in five in the U.S. will have had an Omicron infection.

What this does for “herd immunity” has now become impossible for an amateur like myself to say.  That’s for two reasons.

First, there’s no longer any simple way to count up persons with either vaccination or prior infection.  A significant fraction of these Omicron infections are either breakthrough infections or re-infections.  Just adding up all vaccinations and infections produces a vast over-count of persons with some immunity.

Just to get some handle on the extent to which the simple sum of infections plus vaccinations is an overcount of persons, in the U.S., to date, there have been (per the CDC data tracker)

  • 142M million actual infections (my calculation)
  • 209M people fully vaccinated.

For a total of 351M people, as the simple sum of full vaccinations and estimated total infections.  This exceeds the U.S. population of roughly 330M.  And that doesn’t even count the people who are only partially vaccinated.  My point is that there have now been enough breakthrough infections and re-infections (and persons getting vaccinated after an infection) that the simple count of persons now exceeds the U.S. population.

Second, prior infection with some other variant, or vaccination-but-not-booster, doesn’t provide much immunity to Omicron anyway.  So this is no longer even remotely like a classic herd immunity calculation, where you are assumed to be immune once you’ve recovered from an infection.

You may or may not have noticed that people don’t even talk about herd immunity any more.  Once you get to the point where prior infections don’t confer much immunity, and the virus of interest has an R-nought of something like 15, the whole concept pretty much goes out the window.