The U.S. now stands at an average of 70 new COVID-19 cases per 100K population per day. That’s down 42% in the last seven days. No states are above 200 cases / 100K / day; 43 states are below 100.
The rate of decline is slowing, just a bit, but the rate of nuttiness remains constant. Apparently, large swaths of America are shocked, shocked to find out that in response to low current risk and rapidly falling cases, governors of many states are dropping mask mandates. E.g., this reporting, attributing mask mandate repeals to pandering politicians. I wonder what they think the alternative would be? Masks forever?
In fairness, much of the reporting has it right. E.g., mandates fall as we approach normality. Mandates are eased in response to lower infection and hospitalization numbers.
(And let me contrast dropping a mask mandate, where, with forethought, public health officials make some reasoned judgement regarding risk to the public, versus what we’re in the midst of doing in Virginia, which is initiating a free-for-all regarding in-school mask use, based on a fact-free campaign promise made by the governor about half-a-year ago, done with no reference to current conditions, no reasoned assessment of risk, and no clear guidance to schoolchildren. We may eventually end up in the same maskless place. That said, you can reach the ground floor of a skyscraper by walking down the stairs or jumping off the roof. It’s smarter to have a government that takes the stairs. We have one that can’t even acknowledge that the stairs exist.)
In this case, this blog was only a couple of days ahead of popular press reporting (Post #1426, 2/7/2022). I’ll chalk up the short lead time to the speed with which Omicron cases counts continue to fall.
As a prelude to my next post, just how low are current risks? CDC has now put up a very nice summary of what is known about vaccine effectiveness, as part of the CDC COVID data tracker. Using (what I believe to be) clinical trials data, they estimate that the three-shot vaccine regimen is about 90% effective in preventing hospitalization from Omicron.
With that new information in hand, and accepting that as hard data, right now, the average vaccinated and boostered American faces about the same risk of hospitalization from Omicron as from flu in a typical flu season. That’s how low risks have fallen, from the combination of low case severity, plummeting case numbers, and an effective three-shot vaccine regimen.
If that’s true, then why are COVID-19 hospitalizations (~12,000/day) higher than typical flu season hospitalizations (~2,100/day)? Almost all the Omicron hospitalizations are among the unvaccinated, followed by the less-than-fully-boostered. Even now, this remains mainly a pandemic of the unvaccinated.
Finally, this is not new. COVID-19 risk has temporarily dipped below typical flu risk before. Your risk from COVID-19 falls to a low level every time COVID-19 new case counts get low enough. Late last summer, case counts fell low enough that the average fully-COVID-vaccinated American faced lower hospitalization risk from COVID-19 than from typical seasonal flu. The only difference now is that Omicron is much less virulent. We now obtain that flu-equivalent risk level, for the fully vaccinated-and-boostered, at a fairly high Omicron case count.
Details follow.
Trend to 2/9/2022
Here’s my standard graph of the trend. A couple of days ago, the rate of decline was 45% for the past seven days. So the rate of decline as slowed, but only a bit.
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 2/10/2022, 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
The lines are starting to flatten out, in the graph above, but that’s a consequence of the “natural units” scale. There, a constant rate of decline graphs as a slowly-flattening line. To visualize the percentage rate of decline, you have to graph it in a log scale, where a constant rate of decline translates into a straight line.
If there is a slowdown, it’s not much. Looking in detail (not shown), I can attribute this one-day slowdown to upticks in the data for two populous states, Florida and California. Plausibly even this small slowdown is a bit of a fluke.
There is still nothing that I would call strong evidence that the decline in new U.S. cases is slowing down. The actual case count (blue) still overlaps the constant-growth projection I made last week:
As a consequence of that strong continued downward trend, we’ve now fallen further off our peak than the U.K. has, despite a somewhat earlier peak for the U.K.
Source: Johns Hopkins University via Google Search.
Risk of hospitalization from COVID-19 versus flu, the short version.
Let me start off as simply as possible. If I go to the CDC’s web page regarding disease burden of flu in the U.S., a typical U.S. flu season results in about 380,000 hospitalizations. This is per the CDC’s estimate.
Source: CDC disease burden of flu, Table 1.
It only takes a bit of additional figuring to break that down into average flu hospitalizations per day, during flu season. I went through that in Post #1426. It works out to 2100 flu hospitalizations per day, during flu season.
If we get to the point where there’s fewer than 2100 COVID hospitalizations per day, then, by definition, the risk of hospitalization from COVID is lower than the risk from flu, on a typical day in flu season, for the average American.
If we look at the recent history of the pandemic, there was a brief period mid-summer 2021 when COVID hospitalizations fell below that rate:
Source: Taken from CDC COVID data tracker, accessed 2-10-2022
So, as a simple matter of logic, with no further arithmetic involved, the average American’s risk of hospitalization from COVID, at that time, was less than that for flu, for a typical day in a typical flu season.
But what about vaccinated individuals? This has been a pandemic of the unvaccinated for some time now. By mid-June 2021, more than 150 million Americans were fully vaccinated, and the two-shot vaccine regimen was an estimated 85% effective against hospitalization from the then-prevalent COVID strain, as of June 2021 (source for both: CDC COVID data tracker). Without dwelling on the algebra behind it, this means that, all other things equal, the risk of hospitalization for fully-vaccinated individuals was about 25% of the overall average risk, and the risk for the unvaccinated was about 175% of the overall average risk. (Check: 1 – (25%/175%) =~ 85%).
(These are theoretical rates based solely on the efficacy of the vaccine. The actual observed difference between in hospitalization rates between the vaccinated and unvaccinated populations is even larger than that. Presumably, that’s due to all the other behavioral factors associated with vaccine refusal. E.g., it’s a pretty good guess that those who refused vaccine are also less likely to wear a mask, social distance, avoid crowded indoor situations, and the like.)
As of mid-summer 2021, an overall average COVID-19 hospitalization rate of 8400 cases per day would translate to a rate of 2100 per day for the vaccinated population, all other things equal. That means that for this period, the COVID-vaccinated population of the U.S. faced lower hospitalization risk from COVID than they faced in a typical U.S. flu season:
If we were to redo that calculation today, several things would change. The three-shot regimen appears to be roughly 90% effective against hospitalization with Omicron. Two shots appears to be about 57% effective against hospitalization. But the population shares are different. The boostered population is only about 27% of the total, versus about half of the population being fully vaccinated in June 2021.
Again, without dwelling on the algebra: All other things equal, at present, a boostered individual would expect an Omicron hospitalization rate that is about 18% of the overall average. Full-vaccinated but not boostered individuals would expect a hospitalization rate that is 79% of the average. And an unvaccinated person would expect a rate that is 183% of the average. (Check: 1 – (18%/183%) =~ 90%).
That means that when overall U.S. average COVID-19 hospitalizations hit (2100 / .18 = ) ~11,700 per day, the boostered population faces no more risk of COVID-19 hospitalization than it faces from flu, on a typical day in the average U.S. flu season.
As you can see, for the U.S. as a whole, we are very nearly at the level of Omicron hospitalizations where Omicron risk is the same as flu risk, for the boostered population.
There are some nuances to deal with yet. And a few numbers to clean up. But that’s the gist of it. Right now, the U.S. is at 70 new COVID-19 cases per 100K per day. At some rate not too much below that, we’ll reach the point where the fully-boostered population faces no higher risk of severe illness from Omicron than they do from typical seasonal flu.
There’s one obvious point, that I should add, in that these are the rates we observe doing what we do. They are conditional on not just the vaccination rates, but on the entire current level of COVID-19 hygiene.
Masks — particularly high-filtration masks — reduce your exposure to COVID-19. Staying out of crowded indoor situations, ditto. You’re going to have to wait until Omicron cases drop a bit further before your maskless, socializing COVID risk equalizes with your typical flu risk. How much more, that’s hard to quantify.
But the point remains the same. The combination of low average severity, plummeting cases, and effective three-shot vaccine regimen means that the boostered population of America is rapidly getting near the point where Omicron poses no higher risk of serious illness than typical seasonal flu does. In that context, widespread removal of mask mandates makes sense.
