Early in the pandemic, a small group of epidemiologists recognized that a significant portion of the population would likely refuse vaccination. Those scientists teamed with virologists to develop an alternative way to achieve herd immunity at modest cost. With NIH backing, they used recombinant-DNA technology to create a genetically modified (GMO) version of COVID-19 that was far less lethal, but easier to spread, than the naturally-occurring variants. After extensive testing, they released this new strain of COVID. The new GMO-COVID is now displacing the earlier variant, and thereby reducing mortality and morbidity and hastening the end of the pandemic.
Source: My keister. And why not? Generically speaking, half the nonsense you read on the internet comes from the same place.
The paragraph above is science fiction. We weren’t that smart, or that fast on our feet. Arguably, such a thing isn’t even possible in real time, given current technology. In any case, the legal liability issues alone would have gotten any such projected spiked before it started.
But can we view Omicron this way? Maybe Mother Nature has done for us that which we could not do ourselves. Maybe she cooked up a benign-but-highly-spreadable version of the virus.
Maybe we should embrace Omicron as our path to salvation from the pandemic. I’ve been seeing a lot of talk that boils down to that, at the moment.
Or maybe not. Right now, evidence is sparse. Cases appear to be less severe, but that’s based solely on casual observation of selected segments of the predominantly young South African population. It also appears to be far more infectious than Delta. So that it should produce more cases than Delta.
Before concluding that we should embrace Omicron, somebody should do the math to see whether or not it offers us a beneficial tradeoff between number of cases and typical harm per case. (Somebody sensible, I mean. The nut-o-verse has already incorrectly decided that Omicron is harmless.) I’d settle for something as simple as the expected number of hospitalizations with Delta, versus the same estimate with Omicron.
But let’s be real here. Based on all available evidence to date, we’re probably going to embrace Omicron the way Louisiana embraces a hurricane. That is, without a lot of choice in the matter. If the South African experience is any guide, debate over whether or not we should welcome Omicron is pointless. It’ll be the dominant strain here in a month or two anyway. Just in time for the (somewhat delayed) peak of our winter wave.
In this post, I’m just trying to get a handle on the facts.
- How much more rapidly does Omicron spread, compared to Delta.
- On an apples-to-apples basis, how much less severe is an Omicron infection compared to the current Delta variant?
Obviously, things are still a little murky at this stage. That means that the details matter less than usual, and I have no problem giving my summary judgments up front, then grinding through the details.
At the moment, near as I can tell, Omicron is:
- Vastly more infectious than Delta. Putting aside all the casual estimates, the one hard analysis I found suggested that it’s two to three times more infectious than Delta. That would put the R-nought for Omicron somewhere between 10 and 15, making Omicron as infectious as mumps or chicken pox. That’s consistent with the rapid spread in South Africa, where Omicron displaced Delta about three times faster than Delta displaced the prior variant there (Beta). That’s also consistent with the near-immediate reporting of large clusters of cases (from airline passengers sharing a cabin, from one U.S. wedding), something that did not appear until much later in the spread of Delta.
- May or may not be somewhat less severe in its average impact per infection. People are still being hospitalized for it, and are still dying from it. That just appears to be happening at a reduced rate. How much reduced, that’s the question of the hour, because so much of the South African spread is among the youngest, and because the South African health care system may have a different propensity to hospitalize people. That is compounded by the relatively high fraction of cases that are reinfections (estimated 8%, based on one analysis.) Anecdotally, doctors there suggest that this is mostly resulting in milder cases. But that hasn’t stopped (e.g.) hospitalizations from going up as this has spread (see below).
Based on the information available today, it’s premature to suggest that we embrace Omicron. But it’s not totally beyond the pale, yet. Given the current state of knowledge, it’s still possible that Omicron will turn out to be the lesser of two evils, relative to Delta.
It all depends on just how sick it makes you. And I don’t think there’s any crystal-clear information on that yet. The main problem is that, at present, the it’s mostly spreading among the younger population of South Africa. And this grew so fast that it’s still too soon to see the ultimate death rate.
When the situation is murky and the stakes are high, the bullshitters come out to play. So I thought I’d start with the stuff that’s clearly wrong, as a way of clearing the decks.
Wrong 1: Of course Omicron must be less severe …
You’re going to see a lot of arguments that “of course” Omicron must be less severe, this is a natural consequence of the evolution of the virus, and so on. Near as I can tell, claims that natural selection must/will/does/shall drive new variants to be less deadly is directly contradicted by data from the prior variants.
All you have to do is take five minutes to calculate the hospitalization and mortality rates for the last few variants. Like so:
Source: Calculated from the CDC COVID data tracker, accessed 12/5/2021.
Bear in mind all the other changes occurring over that time period (vaccination of the elderly, and so on). To a rough approximation, one way or the other, twice as many diagnosed cases translated to twice as many hospitalizations and deaths, for all the prior strains. Anybody who tells you that evolution must lead to less severe illness either hasn’t taken the five minutes it would take to check that, or chooses to ignore it.
In any case, if Omicron results in vastly less severe cases, that will be something completely new in this pandemic. We’ve had our fill of viral evolution to date, and that mythical less-harmful variant of COVID-19 hasn’t appeared yet. Not to the degree that anybody would call these later variants “safe”.
Wrong 2: Omicron doesn’t kill people.
This is being spread all over the mainstream media. And we owe it all to one intemperate and ill-thought-through statement by one World Health Organization official, that was then taken completely out of context.
So, every time you see the claim that Omicron won’t kill you, it’s always in the context of “38 countries, no deaths, according to WHO”.
Digging deeper, this is attributed to a 12/3/2021 presentation by Maria Van Kerkhove, the WHO’s lead person for Omicron. (This CNBC reporting has most of the details.)
The gist of the Van Kerkhove’s discussion is clear: WHO is seeing a full spectrum of case severity, and it’s too soon to say whether or not Omicron cases are more or less severe than Delta cases. That’s in part because we’re looking at a sample of cases biased toward younger people and healthy travelers, and in part because it takes time for the deaths to appear.
If she ever said “no reported deaths” on camera, I surely can’t find it. It’s not on any of the clips of that press conference that I can see, and the only direct attribution I see is to reporting by Sputnik, which is the Russian state news agency (and so, hardly a source to be trusted).
As far as I can tell, the notion that Omicron is safe, no deaths have occurred, and so on, is plausibly Russian disinformation. I can’t find that WHO official actually saying that, everybody offers the exact same quote, and that quote appears to have originated with Russian state-sponsored media.
Back in the real world, people keep forgetting that Omicron is now the dominant strain in South Africa. That it has taken just one month achieve that dominance. That it takes about two weeks, on average, to die from COVID. And that people continue to die from COVID-19 in South Africa.
Odds are pretty good, then, that people are dying, from Omicron, in South Africa. But we can’t prove it, because the South Africans haven’t literally sequenced the COVID found among their most recent decedents.
It does appear true that the case hospitalization (and by inference, eventual mortality) rate is lower under Omicron than under Delta. But that would be true in any case, for an outbreak focused almost entirely on college-aged and younger individuals. As has been the case in South Africa so far.
The bottom line, for me, is that the odds are overwhelming that Omicron can be deadly. The WHO’s only official statement seems to have been that they are seeing the full spectrum of case severity. If they in fact said “no deaths”, all that means is that among the handful of international travelers known to have contracted Omicron, no report of deaths within that population has yet reached WHO.
All you have to do is look at South Africa to say, yes, hospitalizations are up. Particularly hospitalizations for COVID among children (though many of those seem to be people hospitlized with COVID, not for COVID). Or note that hospitalizations are up sharply at the epicenter of the Omicron epidemic in South Africa. Or trust the WHO’s estimate of rising hospitalizations in South Africa. If it’s severe enough to hospitalize you, the odds are overwhelming that it’s sometimes severe enough to kill you.
If you still want to believe that it’s not lethal, just take one minute to watch the entire clip from the supposed source of the “no deaths” quote, Van Kerkhove, above. The entire gist of what she’s saying is that the experience to date is not enough to allow us to form a firm judgment regarding severity.
Throughout, bear in mind that all we really have is data from South Africa. What occurs in their situation — with a far lower vaccination rate than in most of the developed world — may or may not be indicative of what will occur in the U.S. and other developed, reasonably-well-vaccinated countries.
Below, I’m going to summarize the evidence as it appears for South Africa, and then anything else I can get. Then maybe I’ll offer some vague discussion about differences between there and the U.S.
1: New case growth is extremely high in South Africa. For example, this reference. (Or, maybe, quadrupled in the past four days might get the point across better.) This has been characterized as Omicron is spreading about twice as fast as Delta ever spread.
2: Omicron has displaced Delta as the dominant strain in South Africa (reference). That, by itself, argues that Omicron is more transmissible than Delta.
3: Omicron displaced Delta in about a month, or about one-third the time it took Delta to displace the prior strain. It took less than four weeks for Omicron to displace Delta in South Africa (reference, though I know I have seen a better reference elsewhere.)
By contrast, South Africa’s experience with the growth of Delta was much like that of the U.S. — it took more than three months for Delta to displace the prior variant (Post #). That suggests that Omicron represents a leap in infectiousness, over Delta, compared to the difference between Delta and the prior strain.
4: Initial numerical estimates of “R-nought” confirm that Omicron represents a leap in infectiousness, over Delta.
The best and most complete analysis so far is this non-peer-reviewed study of the South African experience. The only link I have to it is via Twitter. The document itself is accessible on Google Drive. The study is:
“Omicron Spread in South Africa Growth, Transmissibility, & Immune Escape Estimates,” on behalf of the South African COVID-19 Modelling
Consortium. It’s undated, but the document was created on 12/3/2021.
I have to admit that I don’t understand a key portion of this (“immune escape” as opposed to spread), but the key graph is the following. This shows estimates of the transmissibility (R-nought) of Omicron compared to Delta. I’ve added an arrow at the midpoint of all the modeled estimates to show that, best crude guess, Omicron appears to be more than twice as transmissible as Delta.
Source: Cite above, red arrow and circle mine.
Assuming that’s about right, then we can generate the following table of estimated R-noughts:
- Original variant … 2.5
- Alpha ………………… 3.5
- Delta ………………….. 5.0
- Omicron ………….. >10.0
5: An extra for the geezers in the audience. The estimated R-nought for Omicron is similar to that of mumps or chicken pox.
- Smallpox has an R0 of 3
- Polio has an R0 of 4-6
- Mumps has an R0 of 10-12
- Chickenpox has an R0 of 10-12
- Pertussis has an R0 of 15-17
- Measles has an R0 of 16-18
Source: Vaccines today.
Those of us who predate the vaccines for mumps and chicken pox will recall that when (e.g.) chicken pox hit an area, more-or-less all the kids got it, more-or-less all at the same time.
And so, best guess, among the un-vaccinated and un-masked population without prior infection, under business-as-usual, this new strain is going to spread about like chicken pox used to spread.
Those of you who are under 60 probably can’t relate. For those of you over 60, I think that’s enough said.
But note that the U.S. differs from South Africa in two key respects. First, almost 60% of our population is fully vaccinated, compared to about 28% of the South African population. Plausibly, this is going to slow down the spread of the new variant in the U.S. relative to South Africa. Second, the seasons are reversed — they are heading into summer, we’re heading into winter. So far, winter is the best season for COVID spread. So unlike South Africa, we’re heading into the part of the year when COVID, in general, is easier to spread.
Which effect will dominate, I have no clue. But to a first approximation, I’d guess what took one month in South Africa is unlikely to take more than two months here. So I expect Omicron to be a factor in our (weather-delayed) winter wave.
Severity of illness
I’m simply going to have to punt on this question. Other than the two pieces of debunking above. The reason is the lack of data.
There are too few international cases so far to allow any inference from those. And, those are self-selected to be, by and large, healthy active people who have been traveling abroad.
Within South Africa, we’re handicapped by three factors:
- This is mostly spreading among a very young population. In fact, I believe that the age group most strongly affected at present is infants (those 5 and under). That’s a population for whom vaccination is not yet approved.
- This has grown so fast in South Africa that, by and large, deaths will not have shown up yet. It takes a couple of weeks for the deaths to catch up with the cases.
- We don’t know how hospitalization practices in South Africa compare to those in the U.S. Thus, South Africa appears to have had some increase in hospitalizations, but what that indicates about severity of cases is hard to determine.
But most telling, I think, is the the early data from most variants has ultimately been misleading with regard to severity.
This is the gist of statement by the WHO official reference above. It was also true for Delta, which initially appeared to cause exceptional cases among the young.
“During the early days of the Delta outbreak, there were reports that the variant was causing more serious illness in children than did other variants — an association that dissolved once more data were collected, Çevik says.” (Link to source given just above).
Just look back at Delta. Even as late as early September 2021 (!), there were significant un-answered questions about the severity of Delta. (This, per reporting in the New York Times).
Absorb that for a second, and realize that anybody who tells you that they know, right now, that Omicron produces less severity of illness is probably just kidding themselves. That NY Times article is just after the peak of the Delta wave, and the research on Delta’s severity was still uncertain at that point.
Here’s what I think I know for sure:
- Omicron infection brings some risk of hospitalization. At least, based on hospitalization standards for South Africa. So, for sure, it’s not harmless.
- Prior evolution of the virus did not reduce its virulence. That’s the graph in the debunking section above.
- Every responsible party seems to be saying that it’s too soon to tell.
I would characterize the present situation as anecdotes. We have anecdotes that it results in less severe cases, on average. We have the observational data from South Africa, consisting largely of a very young infected population. We have some untrained individuals who want to form a firm conclusion, based largely in their political ideology. And we have more-or-less all the experts saying that it’s too soon to make that conclusions.
I’m sticking with the experts on this one. There’s a hint that Omicron is less severe than Delta. But there have been misleading hints before. It’s just too soon to tell.