Post #1400, Part 2: Endemic Omicron

Posted on January 22, 2022


In this post, I’m trying to guess what the world will look like after the current Omicron wave ends.

In a nutshell

  1. The consensus of informed opinion is that COVID-19 will become endemic to the U.S., just another one of the many diseases continuously circulating in the population.
  2. How that will work, exactly, nobody seems to be able to tell you.  I can’t quite get my mind around Omicron (or its successors) just fading into the background, given that it’s both extremely infectious and good at evading the immune system.
  3. Best guess, if you are smart enough to stay fully vaccinated and boostered, your overall risk from Omicron won’t be materially different from your overall risk from flu.  I do a rough cut to show that in this post, and plan to do a more systematic job of that tomorrow.
  4. Why, then, is Omicron stressing out the U.S. hospital systems in ways that the flu never does?  One key is in italics above.  That’s due to the high burden of illness among the unvaccinated.  This really is, still, mostly a pandemic of the unvaccinated.  The second factor is simply the sheer volume of weekly new Omicron cases, which I estimate to be four times the  volume of symptomatic flu cases in a typical peak week of flu season.
  5. And the upside of that is that if you are fully vaccinated, right now you are not facing risks from Omicron that hugely greater than those from flu.  In fact, most of your excess risk isn’t due to the virulence of Omicron compared to flu, it’s due to the high prevalence.  There’s just a lot of Omicron going around right now, compared to flu during flu season.  Once we get past this peak, as long as Omicron remains the dominant variant, in the long run, severe illness risk from COVID-19 risk, for the fully-protected population, should be no higher than the risk from flu.

Source:  Calculated from Virginia COVID-19 data by vaccination status, week ending 12/25/2021.

Looking past the end of the Omicron wave.

Now that Omicron is getting ready to peak in the U.S., it’s time to start thinking past the end of the Omicron wave.

If there is an end.

It appears that the overwhelming scientific consensus is that we’re stuck with COVID-19 permanently.  As in, 90% of qualified scientists thought it was going to be end up endemic here in the U.S. (Reference), just one of many diseases constantly in circulation in the population.  And that consensus dates back a year, when we were merely dealing with the native (Wuhan) strain of it, not the vastly more infectious Omicron strain.

Once upon a time, I figured the 2021/22 winter wave would be the end of the COVID-19 pandemic.  That wasn’t just wishful thinking, or mindless analogy to the 1918 flu pandemic.  My calculated guess was that by the end of this 2021/22 winter wave, nearly everyone would have been either fully vaccinated or infected.  Throw that level of immunity into your basic math for epidemics, chuck some reasonable estimate of infectiousness (“R-nought”), and presto, the pandemic should end.

That involved some wishful thinking.  But I really couldn’t contemplate the alternative.

But Omicron changed the math quite a bit.  Not only is it vastly more infectious than prior strains, it’s able to avoid existing immunity to a far greater degree.   Put those new parameters into the basic pandemic equation and it’s hard to see an end to the pandemic.

I don’t think you even need to bring up the unvaccinated to reach that conclusion.  (Although they they certainly aren’t helping things.)  My guess is that the slow decay of natural immunity over time would continuously generate enough new carriers to keep the disease in circulation, given how contagious it is.  Plus, we don’t have a vaccine good enough to put this particular genii back into the bottle anyway.

If the R-nought for Omicron is somewhere around 15, that means you have to stop 14 out of 15 chains of infection in order to bring this pandemic to a close.  If we take no other precautions against spread of disease, that would require that more than 93% of the population have perfect immunity to Omicron.  It’s not possible to achieve that when vaccine plus booster is only perhaps 70% effective in preventing symptomatic infection with Omicron.

But somehow, even though I believe the scientific consensus on this, I can’t quite get my mind around how “endemic COVID-19” is going to work. 

These are certainly examples of diseases that emerged in the U.S. over the past few decades and are still here.  (Emerged meaning that they weren’t here before.)  They are endemic — just part of the background of everyday life in the U.S.A. now.

AIDS.  Zika.  Multi-drug-resistant tuberculosis.  Lyme.  West Nile.  Legionnaire’s disease.  Dengue.  E. coli that can kill you.  Hantavirus.  Methicillin-resistant Staphylococcus aureus (MRSA).  And so on.

The trouble is, Omicron is qualitatively different from any of those diseases listed above.  And it is different from common highly-contagious diseases that we currently control with long-lasting vaccines, such as the numerous formerly-common diseases of childhood (measles, mumps, rubella, varicella, and so on).

And Omicron is qualitatively different from flu, in that it’s vastly more infectious.  A typical estimated R-nought for seasonal flu is somewhere around 1.5.  For Omicron, it’s about ten times that.

As a result, I can’t find any obvious model for how “endemic Omicron” would play out.   I can’t quite wrap my head around how the world will look with a disease that is:

  • currently quite common.
  • Airborne, so requires no vector and requires no physical contact for infection.
  • About as contagious as a disease can be (I did not come across any diseases with  estimated R-nought materially higher than 15, which is best-guess for Omicron).
  • Sometimes causes acute illness and death (more on that below).
  • Still frequently undergoing major mutations.
  • Able to bypass immunity developed from prior infections with other strains.
  • And for which vaccine-induced immunity fades with half a year.

I don’t think there’s another disease in existence today that matches those characteristics.  And so, I’m having a hard time figuring out how we could possibly have a stable, background pool of that, constantly circulating at low levels in the population.  Something about that description of an endemic disease just doesn’t quite line up with Omicron’s ability for explosive growth due to its high R-nought (infectiousness), combined with its ability to evade much of the immune system.

What happens immediately after the Omicron peak?

We can look at South Africa to see that they’ve had a fairly long “tail” to their Omicron wave.  They peaked around 12/17/2021 — just about the time the U.S. got started.  Cases fell rapidly for about two weeks.  And then the rate of decline slowed.  Four weeks after peak, they’ve still got about 25% of their peak case rate.

The U.K. appears to be following roughly the same trajectory so far.  They are less than two full weeks after their peak, and cases have fallen from about two-thirds, from 200K per day at peak to 70K per day.

If the U.S. were to follow the same trajectory, and if we’re hit our peak this week (say Wednesday 1/19/2022) at around 250 cases per 100K per day, we’ll still be looking at:

  • 80 new cases / 100K / day for around February 1.
  • 60 new cases / 100K / day around February 15.

Just for comparison, in 2021, during the mid-summer lull, we had almost two months when the new case rate never exceeded one-tenth of that.  Those were the months when (e.g.) I went back to going to the gym, and so on, due to the low risk of infection.  Months where I would say we could approach normalcy.

The point is, if you won’t feel safe until there’s relatively little virus in circulation — say as little as there was last summer — you’re going to have to keep your guard up for some months yet.

I realize that I keep talking about the peak of the Omicron wave as something to look forward to.  But, in reality, it’s just another way of saying that this is as bad as it gets.  If we follow the South African trajectory, there will still be plenty of opportunity for infection for at least a month after the peak.

And in the long run? There are too many unknowns right now.

Let me just pretend for the time being that Omicron is the final and most successful mutation of COVID-19.  And so, as the winner of that competition, that’s the one we have to live with.

If I had to pick out the single largest unknown in how “endemic Omicron” works out, it would be whether Omicron can readily re-infect people after an Omicron infection.  It’s already well-established that it has a high re-infection rate among those who have recovered from some other strain of COVID-19.  British research seemed to show that prior infection provided almost no protection against Omicron (reference).  And we’re now seeing the same sort of high reinfection rates that were first observed in South Africa.  Below is a graph from Missouri showing that almost eight percent of recent cases are reinfections.


But nobody knows (yet) whether Omicron can reinfect people readily after a prior Omicron infection.  (Or, if so, I haven’t seen it.)  If Omicron can readily reinfect individuals following a prior Omicron infection, then the population will never achieve much in the way of immunity to Omicron.  We might develop immunity to severe disease, but not immunity preventing any infection.

The second big unknown is how effective the new Omicron-specific vaccines will be.  One is already in production and is slated to be available in March (per this reporting).  I have not seen any data on how much more effective the new vaccine is.  (And, per this reporting, manufacturers are reluctant to jump in for fear that the vaccine will be made obsolete by yet another mutation of COVID).

Let me sum it up to this point.

  • The scientific consensus is that COVID-19 will become endemic.  That is, it will always be circulating at low levels in the population.
  • How that’s going to happen, nobody can tell you.
  • I’m skeptical that we’ll reach some nice, stable background rate.  I think the combination of airborne + extremely infectious + high levels of immune escape just begs to result in outbreaks.
  • Nobody can even start to guess what the long run will look like until we have some handle on whether an Omicron infection confers significant immunity against Omicron, and on how effective the new Omicron-specific vaccines will be.


Comparison of risks between Omicron and flu

Since nobody can tell you what “endemic Omicron” will look like, let me turn it the other way around.  How different are the risks now posed by Omicron and by common seasonal flu? 

I’m not ready to put up the numbers on this one yet, so this is just a teaser for a more complete analysis.  I hope to do a more refined set of numbers as the third and final post in this series.

I already looked at this issue crudely in Post #1364At that point, with that crude comparison, I could already see that the numbers were in the same ballpark.

Now I want to take the most recent U.S. data and ask a very specific question:  How different are the risks to a person concerned enough to get fully vaccinated?  So I’d like to know the risks faced by an individual who gets an annual flu shot (for flu), and an individual who is vaccinated and boostered (for Omicron).

That turns out to be a fairly involved task, because most of the data we have for either disease is for the population as a whole.  So in my final post in this series, I’m going to take the raw numbers and try to “back solve” for the risk faced by the prudent and fully-vaccinated individual.

But I can already tell you that the answers are shaping up to offer some pleasant surprises.  Mainly, as far as I can tell, the case mortality rate for Omicron, for a fully-vaccinated individual, now appears to be roughly the same as for seasonal flu.

Let me do the quick-and-dirty cut of the numbers here, to show you were I’m headed.

Start from the CDC’s estimates of the illness burden of flu, on this CDC web page.  Here, I’ve just ignored the statistical uncertainty (the 95% confidence intervals) and taken the median of values for the past ten US flu seasons.

In a typical year, in the U.S., 1.4% of persons with a symptomatic case of the flu end up in the hospital, and 0.13% die.  And there are about 30M symptomatic cases.  So those are the benchmarks for something we can routinely live with.

The question I want to ask and answer is, how does that case mortality rate compare to the average fully-vaccinated and boostered individual with Omicron?  That’s going to take some back-solving from the observed data.

But just crudely, let me pull out some mortality data from Virginia, putting a two-week lag between case counts and death counts to account for the median time from infection to death.  Here I’m looking at Virginia data broken out by vaccination status, on this web page.

For the past three weeks, the highest mortality rate observed for fully-vaccinated individuals in Virginia was 0.12 per 100K population, for the week ending 1/1/2022.  (Earlier weeks show substantially higher rates, but that’s reaching back into the Delta era.)  Going back two weeks, to the week ending 12/18/2021, the fully-vaccinated population contracted Omicron infections at the rate of 111 per 100K population.  Therefore, my two-week-lag case mortality rate for the fully vaccinated population of Virginia is (0.12/111 =) 0.11%.

Compare to the 0.13% from the table above, for flu.  It’s really not that different.

That’s one week of data, that doesn’t account for flu vaccination, and so on and so forth.  On the other hand, “fully vaccinated” is a mix of those who only have two shots, and those who have also gotten a booster shot.

So it’s a rough cut.  But I think this demonstrates that, once infected, Omicron’s risk for a fully-vaccinated person is probably just about on par with the risk from seasonal flu.

Why does the overall severity of illness from Omicron appear much worse that from flu?  Aside from a larger number of total cases, it’s due entirely to the vaccine-stubborn population.  If you’re smart enough to get vaccinated and boostered, the only excess risk you face from Omicron relative to flu arises because there’s such a high Omicron infection rate right now.  And not because the average case of Omicron has higher severity of illness than the average cost of flu, for the fully-vaccinated population.

Source:  Calculated from Virginia COVID-19 data by vaccination status.

Addendum:  But are there really vastly more new COVID-19 cases each week than there are weekly flu cases in a typical year? Interestingly, the answer is no, there are not.  More, yes.  Vastly more, no.

Right now, at the peak of the Omicron wave, the U.S. is identifying roughly 5.5 million new COVID-19 infections per week. You’d have to guess that for every identified case, there’s another one that was not formally identified.  So that would yield about 11M total new COVID-19 cases each week, in the U.S.

By contrast, the CDC estimates (above) about 30M flu cases in a typical year.  By looking at the weekly data for a typical year (I choose 2017-2018), the peak weeks of flu season typically account for 9 percent of the year’s cases.  Doing the math, in a typical peak flu week, the U.S. gets roughly 2.7M symptomatic flu cases.

The upshot of that our all-time peak Omicron week generates only about 4 times as many cases as our typical peak flu week.