Yesterday, a friend asked about recent news reports of an uptick in COVID-19 cases. How serious is this, really? She is in somewhat fragile health, so it’s not an idle question.
My gut reaction — based on the reporting I had scanned — was, not. It’s not a big deal.
But my second reaction was, I’m not sure what information is still available.
So that’s what this post is about: The current state of information on COVID-19. Because COVID, like flu, is seasonal, with an end-of-year peak. And we will eventually get to the end of 2023.
In a nutshell: The only remaining consistent and timely information is on inpatient hospitalizations for COVID-19.
Still no shortage of 🐴💩
Weirdly enough, we still have plenty of COVID disinformation and misinformation circulating. You’d think we’d be over it by now. But, thanks largely to the Republican presidential candidates, there’s still an active market for being angry about COVID-19.
I recently read about a leading Republican candidate for president railing against “federal lockdowns” for COVID. I found this bizarre for two reasons.
As for “federal”, this appears to appeal to the collective amnesia of Republican voters, who apparently have forgotten that restrictions on commerce were imposed by the Governors of the States. For example, restrictions in Florida were imposed by … wait for it … the then-governor of Florida. Who is currently running for President. In part, by railing against those “lockdowns” … wait for it … that he himself imposed.
As for “lockdowns”, I can’t speak for other states, but here, I was always free to go about “essential” business, such as going to work, buying groceries, and so on. The list of essential businesses open to the public was long. None of which were ever shut, here in Virginia, during the pandemic. I find it hard to characterize what happened here as “lockdown” when I could run out to the liquor store for a fifth of Jack any time I chose. (As opposed to what happened in Communist China, for example. They had real lockdowns.)
That isn’t to say that there weren’t restrictions. But to my eye, they were, by and large, rational, and as fact-based as possible. In many states, bars were the first thing to close, and the last thing to re-open. In Virginia, bans on large public gatherings essentially shut down in-person church service, for those churches who obeyed those bans. (Near as I can tell, nothing happened to churches who ignored that). And, after a few well-documented super-spreader events in church services, somehow, passing on large church gatherings for the duration seemed reasonable. Particularly in the pre-vaccine portion of the pandemic.
Meanwhile, the Surgeon General of Florida has gone from recommending that healthy children not get a COVID-19 vaccine (reference, 2022), to outright Looney Tunes claims that getting a COVID-19 vaccination causes you to contract COVID-19 (reference, of a sort, 2023).
In short, even now, the horseshit keeps flying.
So, is there any hard data on COVID-19 any more? Or do the political opportunists and the crazies now own the field entirely?
News Flash: Temperatures are falling …
… as we move toward winter. That means that the incidence of new COVID-19 cases should start rising.
Cases for most coronaviruses reach a minimum sometime mid-summer. This was known well before COVID-19, and is shown in the graph above for Stockholm, Sweden.
I have many prior posts looking at seasonality of COVID-19, but the gist of it is, COVID-19 does the same thing. It’s not quite so clear, as the rapid mutation of during the pandemic, plus the introduction of vaccines, gave peaks and lulls that were not related to the seasons.
But abstracting from those, COVID-19 does the same thing as most other coronaviruses. There were some oddities early on, such as a mid-summer peak the first year, in very hot-climate states, suggesting that the move into indoor air-conditioned spaces lead to greater spread.
But once things had settled down, the pattern was mid-summer low, year-end peak. Which is, for that matter, the same as most acute respirator illnesses, such as flu.
After the mid-summer minimum, well, what do you think minimum means? Cases will rise. Reporting on the fact that COVID-19 cases were rising is not exactly the worst of journalism, but certainly in the neighborhood of that.
Upshot: We should expect COVID-19 cases to be rising now. And they’ll probably continue to rise, right on into the new year. So the mere fact that they are rising is hardly a news flash.
Do we have adequate information to detect a big winter surge?
Information on the number of COVID-19 cases never was much good.
A good chunk of cases (call it 16%) were asymptomatic. (Roughly the same proportion as for flu, in a typical season). So those individuals themselves didn’t even know they had it.
A good chunk of cases (half) were never reported, because individuals were not formally tested with a PCR (DNA) test. Back when the CDC still tracked this via testing blood draws (a sample of convenience), the ratio of likely true cases, to officially-reported cases, was about 2 to 1.
As over-the-counter (antigen-based) testing became widespread, a large fraction of cases was not officially reported because there was no way to report positive antigen tests. At some point, the ratio of likely true cases, to officially-reported cases, rose to about 3 to 1.
As vaccines were introduced, the severity of contracting COVID-19 dropped precipitously. Instead of being a near-death-sentence for older, frailer individuals, likelihood of death was reduced roughly 10- to 20-fold by vaccination. So the count of cases in the post-vaccine era reflected far less burden of serious illness compared to the count in the pre-vaccine era.
Don’t forget that at the peak, COVID-19 increased U.S. deaths about a million a year, from roughly 2.6M to 3.6M deaths per year. It was the largest increase in deaths — and largest reduction in estimated U.S. lifespan — since the 1918 Spanish flu. It was, in certain months, the leading cause of death, and for the entire year, I believe it was the second leading cause of death. (And people who naysay that have no clue how death certificates are filled out.)
Finally, the widespread presence of re-infections means that it’s hard to count much of anything at all. But those re-infected are like those vaccinated — the likely health consequences per case are much reduced from what they were at the start of the pandemic.
The only remaining consistent and timely information is on inpatient hospitalizations.
If there’s a big surge in cases in the community, how will we know? Conversely, if the news media are making a mountain out of a molehill, how can we tell that, objectively.
I’ve seen a lot of conflicting, shoot-from-the-lip opinions about that, so I decided to do something this morning to set myself apart from the Average American. I’m going to take a few minutes to do my homework before I form a firm opinion about this issue.
Rather than go through the TL;DR detail, you’ll have to trust my judgment on this. Near as I can tell, we probably do have enough information to be able to spot any truly large surge in cases. But that’s based almost entirely on the rate of inpatient hospitalizations.
Short answer on the COVID information that is still available:
- Deaths: Good, not timely.
- Inpatient hospitalizations: Good, timely.
- Outpatient hospital visits: Not clear.
- Test positivity: Not clear.
- Total cases: Probably not.
The real eye-opener to me is that, while you still see case count data from the states, you have to read the fine print. For example, Virginia continues to maintain case count data, based on reported tests. But … it’s possible, based on the wording of the footnotes to the data, that they only do that when they track cases for public health purposes, e.g., for outbreaks in prisons, schools, and whatnot.
That said, the Federal government continues to require hospitals to report admissions with COVID-19. In prior posts, I already worked out that this is almost entirely admissions for COVID-19. That is, the vast majority of cases included in those counts were in fact hospitalized for symptoms of severe respiratory infection. And then there is a small minority of cases hospitalized for something else, and found to have COVID upon testing. I cannot be sure, but I’d bet that cases hospitalized for respiratory infection are all still tested for COVID-19, just as they would test for any other common pathogen.
The upshot is that we’ll know, in a timely manner, that there is a surge in cases, when we see that occur in the data on daily new COVID-19 hospitalizations. The time lag there is short, and the severity of illness is roughly constant over time (e.g., sick enough to require an inpatient standard of care).
Source: CDC COVID data tracker, annotations mine.
The downside is that this is almost entirely the oldest old.
Luckily — from a data perspective — any broad-based surge in cases in the community will likely kill off a lot of old folks. (Which is why, in part, that failing to vaccinate children in the heart of the pandemic was not very smart.) And those folks will get hospitalized on their path toward death. Thus giving us timely warning that there is a surge in cases.
In any case, if you’re worried about a resurgence in cases this winter, keep your eye on the hospitalization numbers. It’s not like that’s ideal. It’s more like that’s all we’ve really got. But that should be enough to raise a red flag if there really is a big surge in COVID-19.
