We’ve now topped the record for U.S. daily new COVID-19 cases, set at the height of last year’s winter wave. And the growth in that number isn’t even slowing down yet. Continue reading Post #1372: New U.S. COVID-19 record, no reduction in new case growth.
I’m just trying to get a handle on where things are likely to be headed in the U.S. for the near term. I’m just going to present the last three months of new cases, with a little commentary. These plots all come from Google searched for COVID and the country’s name.
Here’s the game I’m playing: How hard is it to interpret what you are seeing as a peak in new cases? In other words, how much wishful thinking is required to say that a country has already passed its Omicron peak.
South Africa is an example of a short, sharp peak. No wishful thinking required. There were almost no cases of COVID-19 in circulation just prior to the Omicron wave there, so that’s pretty much a pure Omicron wave. And it’s receding at just about the same rate that it grew.
Norway has peaked, but more gradually than South Africa. Again, not much imagination required to interpret the graph below as showing a peak. But Norway was already having a winter wave of Delta when Omicron came along. The early part of that wave is Delta, the latter part is Omicron. That might explain the relatively slow ramp-up compared to South Africa.
Denmark hasn’t reported post-Christmas data yet. Too soon to tell. Maybe a hint of a peak just prior to Christmas. But with no post-Christmas data, it takes a lot of wishful thinking to call that a peak.
Great Britain’s first post-Christmas data looks high, but that one bar is three days’ worth of data. With just a bit of wishful thinking — break the last bar into thirds and fill in last three slots with that — yeah, maybe that looks peak-ish. But it’s too soon to tell. That relatively low first post-Christmas data point might be an artifact of testing availability and behavior over the holiday.
Canada’s COVID trends tend to mirror those in the U.S. (or vice-versa, depending on which side of the border you reside). I can see a peak there, if I just focus on the shape of the blue bars, not the seven-day-moving average line. Another few days ought to determine that one way or the other.
U.S.A. data reporting is so ragged, it’s hard to say much. But, trying to be positive here, take that last blue bar, break off a piece to fill in the reporting hole two days’ prior, and things haven’t really changed much in the last six days. Sure, the seven-day moving average is still rising sharply. But it’s not completely out of the question that (e.g.) growth in new cases has at least slowed.
As you may recall, about a week ago, the CDC announced that Omicron accounted for an estimated 73% of new U.S. COVID-19 cases, for the week ending 12/18/2021. Continue reading Post #1370: CDC estimate of Omicron share revised sharply downward
I thought that we’d break the U.S. record for daily new COVID-19 cases today. But we’re just a touch short. Looks like we’ll almost certainly break it tomorrow.
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 12/28/2021, 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 good news is that the official hospitalization counts out of the CDC show no proportionate rise in hospitalizations. Note that, as scaled, this is the first time in the pandemic where the hospitalizations line is below the new cases line.
Source: CDC COVID data tracker.
You’re going to see people dismiss this as “too soon to tell”, but my best analysis says that’s incorrect. (Or, for that matter, just look at the peak from January — looks like the lines were pretty much in sync then.) I went over all this a coupe of posts ago. There are significant lags between (e.g.) infection, symptom onset, and hospitalization. But that’s not the point. Near as I can tell, on average, for the two lines shown above, reported new hospitalizations lag reported newly diagnosed cases by one day, on average.
For sure, people are still being hospitalized and dying. But at a lower rate, per case, for Omicron than for prior variants.
The problem is, while the hospitalization rate is lower, it doesn’t appear to be anywhere near as low as early research suggested. But I think I have the answer for that, in the next post.
That said, this wave isn’t stopping, it’s continuing to accelerate. And before the words “post-Christmas-surge” leave your lips, please note that a) this wave started before Christmas, and b) owing to the typical lag between infection and reporting, the cases showing up today were infected well before Christmas.
My expectation of a short, sharp wave, based on what occurred in South Africa, appears incorrect. Sharp, yes. But so far, no strong indication that this is going to be sort. South Africa and Norway did indeed have a short, sharp peak. Great Britain and Denmark did not. As of today, it looks like the U.S. is going to fall into the “not” group.
We can hope that this last uptick is some sort of reverse holiday effect. Maybe a lot of people decided to get tested before joining relatives for the holidays. If so, that would be unique to this year. That certainly didn’t happen last year. But if so, that will wash through the data in just a few days.
As you can see from this final chart, cases are up in all regions. The daily new case rate has not-quite-doubled since the start of the Omicron wave on 12/17/2021.
Bottom line: The holidays don’t seem to be making a dent in the growth of reported new U.S. COVID-19 cases this year.
Tomorrow, when all the states report new data, you’ll see that the U.S. has broken the record for daily new cases, set at the peak of the last winter wave.
For now, here’s my best guess for the U.S. trend through 12/26/2021. This is the actual U.S. average through 12/23. Beyond that, it’s based on the growth observed in the ten states that consistently reported data from 12/23 to 12/26.
Based on that hybrid estimate, daily new cases have now roughly doubled in the nine days since the start of the Omicron wave. Best guess, as of today, the U.S. averages about 76 new COVID-19 cases / 100K / day.
We will get actual data from almost all states tomorrow, at which point I’ll replace this “hybrid” estimate with the actual data.
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 12/27/2021, 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.
That said, I think it’s finally dawning on a lot of people that the simple count of cases doesn’t tell you much, if you’re replacing Delta with a far-less-virulent Omicron. I’ve made that point here, and I’m now seeing that in mainstream media (example).
As far as I can tell, there has still been no commensurate surge in hospitalizations. The CDC still has not updated its COVID-19 data tracker, so this graph is my calculation from the US DHHS hospitalization data itself.
Source for hospital data: Calculation from this U.S. DHHS file.
I believe this is consistent what what’s being observed in those states that produce timely hospitalization data. So far, even for states where new cases are skyrocketing, there has been, at most, am increase in hospitalizations that is far less-than-proportionate to the increase in newly diagnosed cases.
As of Christmas Day 2021, there had not yet been any jump in U.S. COVID-19 hospitalizations to match the recent jump in cases. Continue reading Post #1367: COVID-19 hospitalization trend to 12/25/2021. No big uptick, and other surprising aspects of U.S. COVID hospitalization data
My wife just invited her sister over for Christmas Day. Without bothering to ask if I thought that was advisable. Said sister-in-law having possibly been recently exposed to COVID at work. My son is going out to the local bar to hang out with his buddies. Ditto. I’m still going grocery shopping in person, where maskless customers remain an increasingly common sight.
Meanwhile, the U.S. trend for new Omicron cases where I live (Washington DC suburbs) remains more-or-less vertical. As it is in several other locations around the nation.
I think I can connect the dots.
At six days into the Omicron wave, we ought to be close to our darkest hour, based on what we’ve seen in other countries. And that’s working out for me. Continue reading Post #1366: COVID-19 trend to 12/23/2021.
The bottom line is this: There were a lot of new cases of COVID-19 diagnosed yesterday. But you probably already knew that. What’s more newsworthy is that there aren’t a lot of new hospitalizations. Not anywhere you look. Consistent with the best available research at this time. And, while it’s too soon to count the deaths, it’s a good guess there won’t be many deaths, either. As I see it, with Omicron, our winter wave of COVID has morphed into something akin to an unbelievably bad flu season. Lot of people sick. Lot of work time lost, events cancelled, business closed, and other economic losses. But not a lot of severe illness or deaths, per infection.
Despite Omicron’s lower case severity compared to prior strains, it remains a really good time to do what you can to avoid exposure. Because nobody wants to be sick. But the odds of getting extremely sick, if infected, seem to be vastly lower under Omicron than under the prior strains.
See Post #1364 if you want the straight-up crude comparison between Omicron and flu, in terms case hospitalization rates. The big difference now is that Omicron spreads ten times faster than flu. (Literally: R-nought is about 1.25 for seasonal flu, but between 10 and 15 for Omicron).
Tomorrow will likely be the last day we get any clear data on what’s happening in the U.S. Not only are states shutting down reporting for the holiday’s, even the CDC’s COVID data tracker website won’t update tomorrow. So after this, everything will be guesswork until after 1/1/2022
What the heck is “calming” about about focusing on the hardest-hit states? Well, for one, you can be glad you’re not there. But more to the point, they provide a great illustration of what you’re not reading in the news, despite the incredibly rapid growth in cases. You’re reading about the huge increase in cases. But you aren’t reading about overflowing hospitals and morgues. And that’s a good thing, as these things go.
Source: New York Times.
Omicron appears to be hitting metropolitan areas first, and spreading from there. That follows prior waves of new variants. They enter the U.S. via international travel from already-infected areas. And so appear first and in greatest number in places with lots of international travel.
The worst hit area, so far, by far, appears to be Washington DC. That’s depicted above, via the New York Times. As you can see, the daily new COVID-19 cases shatter all prior records for the city. They are very nearly as high as have been observed in any state over the entire pandemic.
That’s so nearly vertical that I would normally think it was a data error, but there’s Maryland right alongside it. And about half the population of Maryland lives in the DC Metro area.
That’s what you’ll see reported. But as important is what you’re not seeing: Reports of overflowing hospitals. And you aren’t seeing that because it’s not happening.
I downloaded DC’s data from their dashboard. They track “patients in hospital”, which is the number of persons who are inpatients, and have COVID-19. They don’t appear to track daily new admission to the hospital, for COVID-19. Here’s the picture of last year’s winter wave, and the Omicron wave:
Source: Calculated from Washington DC COVID-19 dashboard, accessed 12/23/2021.
In Washington DC, for the past seven days, the count of persons:
That’s probably a slight understatement of the seriousness of the situation. Plausibly, some of the persons just diagnosed were not sent immediately to the hospital, but will show up in a few days.
But I think that’s the gist of it. At the Delta hospitalization rate, the current rash of Omicron cases would have been an unprecedented public health disaster. By contrast, at the Omicron hospitalization rate, it’s not even close to maxing out hospital resources.
Unlike DC, Rhode Island had been in the grip of a serious Delta winter wave before Omicron came on the scene. Accordingly, what you see in their data is a blending of rates, as Omicron displaces Delta.
And, as Omicron displaces Delta, despite rising new case counts, the number of COVID-19 cases in Rhode Island hospitals has started to fall:
Source: Rhode Island Department of Health
If you wanted to be contrary, you could point to the short timeframe of the decline. It could be a blip. But Omicron has only really taken off in the U.S. in the past week. So that short timeframe is about right.
New York state is third in the nation, with 110 new COVID-19 cases per 100K per day. And, much like DC, it wasn’t having much of a Delta wave prior to the onset of Omicron.
But the center of the wave in New York is New York City. Let’s look there.
By now you know the drill. Below is a graph of new cases, followed by a graph of new hospitalizations for COVID-19. The hospitalization data are incomplete. And it’s a short timeframe. But when you see the same thing time and again, it’s probably real, and not an error in the data.
Source: Government of New York City.
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 12/23/2021, 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 rate of U.S. new COVID-19 cases / 100K / day is now up 37% since the start of the Omicron wave on 12/27/2021.
If we go back to the pre-Thanksgiving basline, cases are rising at an average rate of 16 percent per week.
As I stated in a just-prior post, it makes no sense to compare the simple count of Omicron cases to Delta cases. It’s like comparing a count of $5 bills and $1 bills, without noting that the $5 bills matter a lot more.
But it’s what everybody does. And all the better information — hospitalizations, deaths, and possibly some other indicators of morbidity — all arrive with some time lag. So I’m counting cases, along with everybody else. Even though it makes no sense.
The F-word in this case being flu.
This whole debate about how COVID-19 compares to influenza has been so poisoned by the nut-o-verse that it’s hard to have a rational conversation about it.
But things have changed a lot since that early disinformation regarding COVID being no worse than flu.
COVID hospitalization and mortality rates in the U.S. have fallen since that initial New-York-centered outbreak. And now, it appears that the latest mutation of the virus is resulting in far fewer hospitalizations and deaths.
And so, what was once a right-wing fantasy (“no worse than the flu”) is, in my opinion, no longer so far fetched.
Without beating this into the ground, I’m just going to show you the numbers that have me thinking about this.
First, you have to look up what the CDC gives for the closest analogs of case hospitalization rate and case mortality rate for the flu. You can find the CDC estimates at this link.
There are significant differences in methods between a “case hospitalization” or “case mortality” rate for COVID versus the flu. For COVID, a case is a formally diagnosed case — more or less, a positive test result. By contrast, the CDC’s “case” for flu is an estimate of all symptomatic cases. So the denominators differ. One is all persons with a positive test, the other is an estimate of all persons that had symptomatic disease.
The deaths for COVID are fairly straightforward, but deaths for flu involves considerable estimation, as reporting of flu diagnoses on death certificates is not as clear-cut.
So there are difference in methods.
That said, let me just lay out what appear to be the facts.
Per the CDC, eyeballing their chart, I estimate the following for a typical flu season:
For Delta, most recently, this is my recollection of the current rates:
If those aren’t exactly right, they’re close.
But now, per the South African study released today , the current estimate is that Omicron has about one-fifth the case hospitalization rate of Delta. If that’s true, then for the U.S., we should eventually find that:
Compare to flu above.
Further, I’ve already noted the near-total lack of deaths in South Africa from Omicron. So while I won’t attempt a formal case-rate calculation, it’s not beyond the pale to suggest that the Omicron case mortality rate is at least 80% lower than Delta, if not more.
Once upon a time, nobody was vaccinated against COVID, nobody had any immunity, and there were no effective treatments for it. Hospitals didn’t really know how to treat it. By contrast, just about everyone except newborns had some prior exposure to flu, 60% of the population got a modestly-effective vaccine each year, and hospitals had decades of experience treating the worst cases. During the initial outbreak, in the U.S., the case hospitalization rate for COVID was easily 10x that of flu, case mortality rate was nearly 100x that of flu. At that time, people who said that COVID was no worse than flu were just flaming idiots.
At present, most are vaccinated for COVID, others have some immunity via prior infection, hospitals have gotten considerably better at treating it, and there are some modestly effective treatments (monoclonal antibodies, anti-viral-replication drugs) available. Even then, with that, if you’d said Delta is no worse then flu, you’d have been an idiot.
Looking forward, if Omicron really results in cases with such lower average severity, you know, I don’t think I’m at all crazy for saying that Omicron’s case hospitalization and case mortality rates are starting to look flu-like.
There’s still an issue of total illness burden. Not just because a third of our population still has no immunity to it. But also because there is no comparison in this simple-minded analysis that looks at the average symptomatic-but-non-hospitalized case. Nobody talks about “long flu”, for example, but “long COVID” is a real thing.
That said, at some point, it looks like we’ve going to have to live with this in circulation. Assuming Omicron sticks around, and the South African estimates pan out, we probably need to start thinking of this the same way as flu. Get your vaccine every fall, stay home if you’re sick, and hope for the best.
I hope it remains legal and customary to wear face masks during winter flu/COVID season. This is common in Asian countries, and in hindsight now just seems like common sense. I don’t want to be denied the right to give myself a little extra protection.
But the bottom line is that if it reaches the point where the risks from flu and COVID are not so different, and the population has some resistance to it, we’re going to have to start using the F-word in this context, and just ignore the history behind that.
First, a little international perspective. Just a few graphs to calm the nerves. Then the current U.S. carnage.
To cut to the chase, for the U.S.as a whole, daily new cases increased by about 25 percent in the past four days. Continue reading Post #1363: COVID-19 trend to 12/21/2021
