Post #1386: COVID-19 trend to 1/4/2021, new infections still doubling every seven days.

Posted on January 5, 2022


New York and Washington, DC still appear to have plateaued.  Otherwise, for the U.S. as a whole, new infections are still doubling every seven days.  With the exception of NY, DC, and Maine, new cases are rising rapidly everywhere.  The U.S. now stands just shy of 170 new COVID-19 infections / 100K / day.

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 1/5/2022, from”  The NY Times U.S. tracking page may be found at


This is a quick recap of what I think I know about Omicron in the U.S.  These are all estimates that I’ve documented, to some degree, in prior posts.  I just want to put them all in one place.

Omicron spreads vastly more readily than Delta.  Where Delta had an R-nought of maybe 5, estimates for Omicron now seem to be 15 and up.  (That is, absent an immunity or precautions, if every person infected with Delta would have gone on to infect five others, every person with Omicron would have gone on to infect 15 others.)  To put that in perspective, typical seasonal flu in the U.S. has a R-nought of maybe 1.5.

Research shows that two shots of vaccine or a prior COVID infection provide little or no protection against having a symptomatic Omicron infection  Near as I could tell, from both test-tube studies (where they assayed antibody response to Omicron) and the best observational studies (where they analyzed odds of infection by vaccination/prior infection status), the protective effects were so low that systematic research could not rule out literally zero protection against Omicron infection.  I believe the highest estimate I’ve seen is 30% protection from two shots of vaccine, but that was the outlier.

Full vaccination plus booster shot appears to provide about half as much protection against Omicron as against Delta.  Plausibly, it reduced your odds of any symptomatic infection by about 45 percent, relative to an unvaccinated person, all other things equal.  Might as well say “half”, because that makes it easier to think about.  Highest estimate I’ve seen is 70% protection, but that was the outlier.

None of that, by the way, seems to appear in U.S. observational data.  At least, not yet.  In theory, we should start to see infection rates for the vaccinated and unvaccinated moving closer to each other.  But I don’t see that happening.  E.g., For Virginia, for the week ended 12/25/2021, unvaccinated individuals were still being infected at a rate twelve times higher than those who are vaccinated.  That’s about what it has been all along.  I’m not quite sure what to make of that.  I believe the body of research showing little or no protection against Omicron is more-or-less unimpeachable.  For now, I’ll chalk that up to time lags, or some sort of behavior effect.  Perhaps those who are unvaccinated are simply a whole lot more careless in general, regarding risk of COVID-19 infection.

Omicron produces cases with a lower average severity than Delta, but estimates of that keep changing. Based on what occurred in South Africa, Omicron seemed to produce much less severe illness, on average.  But in Europe and America, Omicron appears much closer to Delta in terms of objective measures of severity of illness.

And, in theory, vaccination and prior infection should provide some protection against serious illness, even if they do not protect against having some symptoms of COVID if infected.  That appears to be because even if your antibodies can’t do much about Omicron, the slower-moving portions of your immune system are still primed against it.  So if you’ve had vaccination or infection, the idea is that you’ll be able to deal with Omicron before it puts you in the hospital.

Here’s what I think I have observed so far, in the U.S.

The U.S. case hospitalization rate for Omicron is about two-thirds that of Delta.  That’s just straight-up arithmetic, based on hospitalization data reported through 1/3/2022, on the US DHHS unified hospital file, available at this link.  The more recent seven days showed a case hospitalization rate of 4.2 percent.  Previously, under Delta, the U.S. case hospitalization rate was about 6.5 percent.

There are a few caveats worth noting.  First, that’s for all cases, including pediatric cases.  The hospitalization rate for adults would be significantly higher than that.  Second, because the number of cases is growing so fast, that 4.2% could be an understatement of the true value.  Even a single day of data lag between the two time series (cases, hospitalizations) would move that number around quite a bit.  Finally, that two-thirds-of-the-Delta-rate estimate is (barely) within the span of estimates produced by British research.  It’s vastly higher than research out of South Africa.  Plausibly, a country’s propensity to hospitalize people plays some role in that.

Likelihood of ICU use, once hospitalized with Omicron, appears to be about two-thirds that of Delta.  That’s harder to assess, because that’s not reported on a per-admission basis.  The Federal hospitalization file used above only shows the count of patients in the hospital and in the ICU on a given day.  At the peak of the Delta wave, 30% of patients hospitalized for Delta were in the ICU.  By contrast, for the week ending 1/3/2022, 20% of patients hospitalized (presumably for Omicron) were in the ICU.

As with the case hospitalization rate, this one may be affected by timing.  That is, if it takes a few days to require an ICU level of care, the rapid increase in cases means that I might be understating the true (eventual) ICU use rate.  I have no way to assess that given available data.

Slapping those two estimates together, you end up with a “case ICU rate” for Omicron that’s about 40% that of Delta.  That is, if you get infected, the likelihood that you’ll end up in the ICU with Omicron is only about 40% of the likelihood that you’d end up there with Delta.

Finally, for mortality rate, every analysis says that it’s much lower under Omicron.  If I had to guess, 80% lower.  But it’s too soon to assess that for the U.S. I can take a rough cut of that from Great Britain.  For the past seven days, they have averaged 131 COVID deaths per day.  Two weeks ago, they averaged 87,000 new COVID cases per day.  Doing the arithmetic, that’s a crudely-estimated case mortality rate of under 0.2%.  Which compares to a U.S. figure, for Delta, of about 1.2%.