The U.S. is now down to 37 new cases per 100K population per day, down from 38 at the end of last week. Daily new hospitalizations have fallen below 6000 per day. Deaths remain around 350 per 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 8/2/2022, 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
Really, the only change I’m noticing these days is that all the scare stores have left the news, except for long COVID. Near as I can tell from scanning the news, all the “experts” who asserted that BA.5 was The Worst Thing Ever have quietly faded back into their well-earned obscurity.
For long COVID, all of the popular press stories seem to have the exact same structure. They start with an interview with an individual who has it. Then they assert that this is a Huge Problem for the U.S., that it’s causing labor shortages, that it’s a coming disaster for the U.S. disability payments system, and so on.
All of which — other than than the one person interviewed — seems to be based on complete and total innumeracy (that is, the ability to do simple round-numbers calculations). This, mixed with an absolute naivete about methods and uncertainties health services research and epidemiological research. Combined with a willingness to turn a blind eye to the only reasonably-well-controlled study of this issue that exists, from the U.S. CDC.
At any rate, I’ve gone over this before, last time in Post #1524. If you take the numbers that are being thrown around about the prevelance of long COVID, and do some simple math, you end up with results that are absolutely not credible (innumeracy). The numbers just don’t add up, at least not as they are being interpreted in the press. And then if you look closely, the numbers on long COVID come from two unreliable sources (naivete).
Want to take the acid test on how credible the numbers are? Do you know 10 people who’ve recovered from COVID? If you believe the numbers, then, on average, between three and four of them ought to have significant lingering COVID symptoms.
Does that match your experience? If not, then who are you going to trust, the supposed experts, or you own lying eyes?
One way being used to get bad information is to ask people with COVID whether or not they feel like they have any of a laundry-list of symptoms. But not ask the same questions of a control group. Do I sometimes feel tired, or moderately depressed, even though I’ve never had COVID? Yep, sure do. So how can you take the absolute fraction of COVID survivors who answer “yes” to those questions, and infer that their troubles are due to COVID? The answer is that you can’t.
And when you actually do seek out a reasonable control group, what you find is that, to a close approximation, a cross-section of those who survived a COVID infection look no different from those who recovered from what appears to have been some other minor respiratory infection. Like so:
Source: Adapted from CDC COVID data tracker.
Above, from Post #1557, contrast the rate of symptoms represented by the orange bars. One chart represents COVID survivors, the other is the control group. Having a hard time figuring out which is which? That’s my point.
The other way being used to get bad information is to do a study of people who were sick enough to be hospitalized for COVID. And then never check for that, and certainly never acknowledge that you’ve done that. Instead, pretend that you’re looking at the average COVID-19 case. And so extrapolate the rate of lingering symptoms for the tiny subset of persons who were hospitalized, and assume that rate applies to the entire fraction of the U.S. population who has had COVID.
I ran through that in Post #1524. I admit that I spent my career doing analysis of health care claims data, so I knew what to look for. That said, even a tiny bit of attention to the key numbers would tell you that the study that is now widely cited for the prevalence of long COVID looked at only a tiny subset of COVID cases in their target population. And that the study was drawn from health care claims (bills) from several large hospital systems.
Long COVID is a real phenomenon. For that matter, long flu is a real phenomenon. And I would bet that for any potentially dangerous viral infection, there is some fraction of survivors who suffer permanent or at least long-term damage.
I have sympathy for people who have been struck down by it.
Ultimately, the only difference between that and (say) being a polio victim is that everyone recognizes that post-polio-infection paralysis is a serious condition. And we all acknowledge that the incredible bit of bad luck that a person would end up with permanent and sever damage from that particular viral infection.
But to assert that a third of U.S. COVID cases end up with long COVID is not really different from asserting that a third of us have polio-based paralysis of some sort.
It’s exaggeration. It’s hype. Somewhere within the hype there is a tiny grain of truth, in that long COVID is real.
But, as it is used by modern media, given the current stable situation, it’s really little more than the last scare tactic that can be thrown in your face as click-bait. Of the health issues that I worry about, long COVID doesn’t even make the list.
