Post #1341: COVID-19 trend and the most under-reported bit of yesterday’s news.

Posted on December 8, 2021


Trend?  ‘Bout the same as yesterday.  The trend numbers I’m sketching in are “connect-the-dots” estimates:  One end is 11/24/2021, and the other end is the most recent data point.  Based on those two points, new cases are now rising at 13% per week.

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 12/4/2021, from”  The NY Times U.S. tracking page may be found at

Public service announcement

In addition, below, I review some research published yesterday out of South Africa, and a followup published today from Pfizer.  The gist of that is:

Get your booster shot.  Now.

Why?  Based on test-tube results, two-doses-of-vaccine provides next-to-no protection against infection from Omicron, although two doses might still provide some protection against severe disease.  You need three doses to mount an antibody response against Omicron.

On a societal level, this has huge implications for the expected rate of spread of Omicron in the U.S.  We’re sitting here thinking that something like 60 percent of the population is fully vaccinated.  Nope, not against Omicron.  Against Omicron, effectively only those with the booster are fully vaccinated.  (Although, it remains possible that two doses will still reduce the prevalence of serious illness, even if they don’t stop infection).  And then, the vaccine might not even work as well against Omicron as against Delta.

So, we thought we were here:

Against Omicron, we’re actually here (below), if we’re lucky (if the three-shot regimen is somewhere near as effective against Omicron as two shots are against Delta.)



I’m not prone to hyping the dangers of COVID-19.  I think the facts speak loudly enough.  Right now, more-or-less eight percent of the persons diagnosed with it end up hospitalized, and something in excess of one percent end up dying from it.  It’s prevalent enough to have become the third leading cause of death in the U.S. in 2020 (reference).   It crowded hospitals enough that “crisis standards of care” (a.k.a., just let some die) were invoked for at least two state hospital systems (ID, AK).

And, for good measure, you can catch it just by breathing the air.  If that air was  contaminated by an infected individual talking or even breathing nearby.  Even if that person appears and feels completely healthy at the time of infection.

U.S., Deaths per 1000 population, 1958 to present:

(Source:  USAfacts.)

This has been an event unprecedented in out lifetimes. And it’s ongoing.

Even as you don’t want to claim that this is the end of the world, you don’t want to swing too far in the other direction either.  It remains a potentially deadly disease that needs to be respected for what it is, and not dismissed as if it were the common cold.

Yesterday I read some reporting on new research out of South Africa.  Read it, read the commentary, moved on.   More bad news, ho-hum.

This morning, I woke up and said:


But what’s the big deal, anyway?  We’re vaccinated now.  Combine that with some COVID hygiene (N95 mask, avoiding large public gatherings) and you’ll probably be fine.  This may still be a problem for society, but as individuals, those of us who have the option to (and good sense to) protect ourselves can do so.

Which brings me to the most under-reported bit of yesterday’s news.  Let me now:

  1. Describe the experiment and results,
  2. Tell you why I think they matter greatly.
  3. Tell you why I think that virtually all the popular press reporting has downplayed it or missed the point.
  4. Summarize and finish.

I’m going to do this from memory, having read through the reporting yesterday.  I’m therefore not providing references.  Google it if you want to read the popular-press reporting of the original research.  I hope that my overnight digest of the research and commentary makes no material errors on the facts.

1:  Research and results.

Researchers in South Africa took blood from twelve individuals.  Let me call them Group A and Group B:

  • A:  Vaccinated:  Six persons had received two doses of COVID-19 vaccine
  • B:  Vaccinated and infected:  Six persons had survived a COVID-19 infection AND had received two doses of COVID-19 vaccine.

Those researchers took cultures of two different types of COVID-19.

  • One was the  prior prevalent strain of COVID-19 in South Africa
  • One was Omicron.

They then tested to see whether the antibodies in the blood would bind with the virus.  If they’ll bind with it, that’s a pretty good indication that your immune system would deal with the virus quickly. I.e., that you’d be able to kill the virus before you got very sick.

The results can best be thought of as a simple 2 x 2 experiment.

My understanding of the results is shown below:

Obviously, the key cell — and the one that will be most scrutinized — is the one I’ve labeled “No” above.  No, antibodies generated from vaccination-only more-or-less failed to attach to Omicron.  And by “No”, (or, if you prefer, “barely”), the researchers categorized the magnitude of the effect as “41-fold” less.  Inverting that, antibodies generated by existing vaccines were triggered by Omicron about 2.5% as well as by the prior strain.  Whereas antibodies from prior infection (plus vaccination) seem to stick to Omicron about as well as they stuck to prior strains.

(Edit:  This portion of the finding has been validated by Pfizer — see last section.)

2:  My interpretation.

Bear in mind that I have just an educated layman’s understanding of the immune system.  But as I understand it, based on those results, the most likely scenario is that two doses of the existing vaccines provide (pick one):

  1. No protection against infection with Omicron
  2. Almost no protection against infection with Omicron
  3. A yet-to-be-determined level of protection, but for damned sure, we wish this experiment hadn’t turned out the way it did.

My best guess is 2. above.  I might settle for 3.  But best guess, 2.

Here’s why, based on my understanding of the immune system.  In short, if those antibodies don’t “recognize” Omicron, then your immune response to Omicron is going to be quite slow.  Which, to me, means that you’ll have to get sick in order to develop a full immune response to Omicron.

Cartoon version of how antibodies work:  Antibodies are the sentries of your immune system.  They are little balls of proteins, constructed to fit some specific other protein, much like a key for a lock.  The “lock” that they fit will be a protein on some bacterium or virus that your immune system defeated in the past.  And they contain a self-destruct switch.  When they stumble across that specific protein they are designed to fit, they raise a little protein flag that says to the rest of your immune system, please destroy whatever I’m attached to.  And the rest of your immune system then carries out the destruction.

The point is, this is how you beat back a bacterium or virus without becoming  sick. Your antibodies stick to it, they raise the alarm, and your immune system does the rest.  This is how a vaccine “protects you from infection”.  It allows you to beat back any pathogen that you (e.g.) accidentally inhale, before it can get a real foothold.

If your existing antibodies don’t stick to the pathogen, your immune response is much slower.  Which means that the pathogens reproduce before you start killing them off.  Which means that you get sick.

3:  What you’ll read in popular press reporting on this topic.

So far, literally everything I’ve read in popular press reporting of these results is some version of “oh, don’t worry about it”.  Or, perhaps, “Don’t panic”.  And, that may be true, or it may just be reporters misinterpreting statements by scientists.  Here’s my take on what has been reported.

3.1  Oh, we expected that.  Sure, academics expected that the vaccines would be less effected against Omicron due to the many mutations that Omicron has.   Those mutations change the shape of the proteins on the surface of the virus, and so make it less likely that the existing antibody “keys” will fit those new protein “locks”.  (See “Cartoon version” above).

But I don’t think anybody expected a “41-fold” lower response.  Lower response, sure.  That much lower?  Expected?

3.2  We didn’t get zero response, and that’s good.  Again, sure, academics working on vaccines can look at the non-zero antibody binding and say, great, that gives us something to work from in our next generation of vaccines.  (If the response had been literally zero, they’d kind-of be working from scratch).

But celebrating a non-zero antibody response only makes sense in the context of the time (100 days minimum, per Pfizer, referenced below) to take to create, test, and authorize a new vaccine against Omicron.  Which, near as I can tell, given that Omicron is already here, is on-order-of about two months too late to do us any good with this year’s winter wave.

3.3  There are other parts of your immune system, antibodies aren’t the whole story. 

Absolutely true.  My understanding is that these other systems work quite slowly, compared to antibodies.  So, saying that other parts of the immune system are there is cold comfort, I think.  I think that’s equivalent to saying “so, you’ll have to get some what sick if you are infected with Omicron”.

Plausibly, this “other parts of the immune system will still work” is the reason that absolutely everyone says:

3.4)  Well, the vaccines will probably still protect against severe disease from Omicron, even if it doesn’t protect against infection from Omicron.

Which is great, and everybody says that the vaccines are particularly good at protecting against severe disease now.  But that this is based on the assumption that even though your antibodies don’t recognize Omicron, the rest of your immune system will.  And there’s no empirical evidence to support that yet.  This is not quite pure wishful thinking, because in the canonical case you would expect some immune response spillover to a closely related organism (as Omicron is closely related to, but not identical to, prior strains of COVID-19).  But it’s also a statement unsupported, as yet, by evidence.

Edit:  Pfizer (below) says that 80% of what the rest of the immune system will react to remains unchanged between prior strains of COVID and Omicron.  Their assumption is that the slower channels of immune response will kick in for Omicron and prevent most severe illness.

3.5:  Well, vaccination plus booster ought to work, even if vaccination alone did not.

The assumption here is that those who were vaccinated alone simply didn’t have a high enough level of antibodies in the blood.  But that flatly ignores the fact that those same people seemed to have plenty enough antibodies to bind with the prior variant.

In other words, the folks making this statement have commingled the amount of antibodies with the type of antibodies.  The study didn’t show that the vaccinated individuals lacked for antibodies.  To the contrary, the control portion (the test with existing strains) demonstrated that they had an adequate binding response to the existing strain.  (In addition, they might have had lower antibody levels, but that’s not the main point.)  As I read it, the main thrust of the Omicron-versus-prior-variant comparison is that the vaccinated-only population has the wrong type of antibodies to bind to Omicron .

Edit:  Today’s results from Pfizer appear to confirm this — see below.  They claim that with two doses, they see roughly the same results as were demonstrated in South Africa.  They found a 25-fold reduction, instead of 41.  But they also claim that with three doses, all is well, more-or-less.  See final section.

(I confess I find this result to be absolutely confounding.  But that’s what they report.  Two shots gives you what appears to be almost no protection.  That’s immediately after onset of full immunity (three weeks after second shot).  But a third shot, and everything’s jake?  Clearly there’s something going that I don’t understand.  But the results are what they are.)

4:  Epilogue:  Panic early and often.

This result actually makes pretty good sense.  I would expect — and I guess most scientists would expect — that the spectrum of antibodies generated by vaccines would be narrower and more specific than those generated by actual infection.

With the vaccines, your body was exposed to some specific fragment of the “spike” on the outside of the initial strain of COVID-19.  Vaccinated individuals would then develop antibodies that key into that specific protein fragment.  Those who were infected, by contrast, were exposed to the entire organism, and likely developed antibodies that would “key into” different parts of the virus.

Thus, vaccine-generated antibodies should be less robust to mutations, compared to antibodies generated by infection.  If the mutations in Omicron just-by-chance change the shape of the key proteins that your body reacted to in the vaccine, then … well, the antibodies from the vaccine may no longer lock into that specific protein. Whereas antibodies from infection may have been developed for a much larger subset of the proteins on the surface of the virus.  Some of them might still work even with a significant number of viral mutations.

In short, it’s not a big surprise that the vaccine-derived antibodies might be less robust to mutations than the infection-derived antibodies.

Unfortunately, the nut-o-verse is going to go crazy over that.  That’s just begging to be a major “I told you so” event for those who advocated letting the infection rip through the population in the first place.  The same individuals will, of course, forget that the only way to have the population acquire infection-derived antibodies would have been to have a whole lot more people get hospitalized and die from COVID-19.

Obviously, before we go into full panic mode over this test-tube result, we’d really want to see the actual bottom-line number:  What is the efficacy of the (Pfizer, Moderna, J and J) vaccine against Omicron? 

And yet, best guess, we’re at least weeks away from anybody being able to produce a credible estimate of that.  And, based on the South African growth rates, a few weeks from now Omicron will be the dominant variant here.  We’ll get a crystal-clear and definitive answer to this question only in our rear-view mirror.

Where does that leave us?

First, three aspects of Omicron all make sense as a package.  They are all reflections of a virus that has mutated to be better at dodging the immune system.

1: We now have a much-more-contagious version of COVID-19 spreading around the world.

2: It seems to be able to avoid existing natural immunity pretty well.  The re-infection rate in South Africa was characterized as “three times that of the prior strains”, and in absolute terms, about 8% of South African Omicron infections are re-infections.

3: Now we have some test-tube evidence that it avoids vaccine-generated immunity even better.  We don’t yet have that definitive figure on how much less effective the vaccines will be.  But, to me at least, if your antibodies don’t kick in, then that means at the least, you’re likely to get infected.  Whether or not vaccines still produce lower likelihood of severe illness, without aid of antibodies, is an unanswered question, I think.

Pfizer’s most recent research seems to say that you’ll be OK as long as you get your booster shot.   But I’m still not quite convinced that it’s as simple as that.  Again, see below.

Our last hope is that it produces only mild infections.  But even there, popular press reporting misrepresents expert opinion.  I’m sure you saw headlines yesterday that Fauci said it looks like Omicron produces less severe infections.  In fact, he said no such thing.  What he actually said is that there is no evidence that it produces more severe infections than the current strain.  And so, what we have for evidence of that is the impression of a handful of South African physicians, based on a population of patients that is vastly younger than the population for the prior strain.  Given that, there’s no way to tell ” by eye” that the new strain is, in fact, less virulent (as opposed to looking at a generally healthier and more robust younger population).

4.1 Implications, or, is it time to refresh your TP stockpile?

Bear in mind that all the data you see are a couple of weeks after-the-fact.  That’s due to the lag between infection and reporting, the time it takes to produce genetic sequencing of samples.  And so on.

I’m not quite at the point of building back my toilet paper stockpile.  But I am considering several non-toilet-paper-related changes.  I’m going to be even-more-circumspect in public places for the next couple of weeks, until more information develops.  I’m switching back to my best-sealing N95.  And so on.

I think this most recent piece of research out of South Africa raises a big red flag for the vaccinated population.  And, guess what — by the time that’ll be  definitively resolved, Omicron will already be the prevalent strain in the U.S.  So if you act, you’re going to have to act based on incomplete information.

If you’ve relaxed your habits a bit, figuring that you’re well-protected by vaccination, now might be a good time to tighten them back up.  Just for a while.  Just until we know whether or not this extremely contagious form of COVID-19, that is far better than its predecessors at defeating natural immunity, really can blow right past your vaccine-generated antibodies.  Or not.

5  Addendum:  Today’s research released by Pfizer.

This new development boils down to two doses plus booster seems to work fairly well against Omicron.   So, get your booster.

Today there’s a big headline that changes weirdly depending on where you view it.  In Google News, the headline is “Two vaccine doses may not provide sufficient protection against Omicron coronavirus variant, Pfizer says.”  But in the actual article, the headline reads “Protection against Omicron coronavirus variant improves with three vaccine doses, Pfizer says”.

The key wording:  “Data indicate that a third dose of … increases the neutralizing antibody titers by 25-fold compared to two doses against the Omicron variant; titers after the booster dose are comparable to titers observed after two doses against the wild-type virus which are associated with high levels of protection.”

But I note that the three-shot regimen is only about half as effective against Omicron as it is against Delta:

The geometric mean titer (GMT) of neutralizing antibody against the Omicron variant measured in the samples was 154 (after three doses), compared to 398 against the Delta variant (after three doses) and 155 against the ancestral strain (after two doses).”

Each of those, by the way, is immediately after full immunity is reached (e.g., 3 weeks after second shot, or one month after third shot).  I’m pretty sure that means that whatever protection you thought you had against any infection from Delta, you’ve got much less protection against any infection from Omicron.

The upshot of all of that is:

  1. Pfizer validates that two doses of vaccine generates almost no antibody response to Omicron.
  2. Pfizer finds a good antibody response to Omicron after three doses (but not as good as the response to Delta).  Crudely put, based on Pfizer’s research, three doses provides about half as much protection against Omicron as it does against Delta.  (It invokes half the level of neutralizing antibodies).  So even with that third shot, you might want to re-evaluate your routines.
  3. You might still get protection from severe disease from the slower-acting portions of your immune system.