NewsWire: 8/9/21
This essay went out this morning on Hedgeye's "Early Look." We are issuing it here for all our subscribers. It is an updated version of "Long Covid Casts a Long Shadow." Watch my discussion today with Hedgeye President Michael Blum here.
"The whole town rushed outside to celebrate this crowded minute when the time of suffering had ended and the time of forgetting had not yet begun."
--Albert Camus, The Plague
What is “Long Covid”? Most Americans know by now is that it is unpleasant, spans a variety of symptoms, and affects some recovered Covid-19 patients. That’s about all they want to know.
I’m going to explain what most Americans don’t want to know but will soon find out. It’s that Long Covid affects a sizeable share of recovered patients, has a long half-life, and is going to challenge our health and generate headwinds to the US economy for years to come.
New research is gradually expanding our understanding of Long Covid. The most recent study, by a research nonprofit with access to U.S. private insurance claims data (FAIR Health), boasts gigantic n-size. It selected the claims records for nearly 2 million Americans who tested positive for Covid-19 between February and December of 2020 and then examined what happened to them later.
Verdict: 23% of positives came back to see a provider for a Covid-related condition more than 30 days after their test--the most common definition of "Long Covid." The rate was certainly higher for the 5% of positives who ended up hospitalized for Covid (50%) than for the 40% with only mild symptoms (27%). Yet the rate was substantial even for the 50%+ of positives who were asymptomatic (19%).
This 23% is consistent with a more rigorous statistical analysis conducted by Washington University researchers of positives within the Veterans Health Administration. This sample was also quite large (n=73K). The researchers compared it to a control group that was statistically created to be identical to the positives, except for the fact that the controls did not test positive. Their result: Positives (more than 30 days after test) were 20% more likely to require outpatient care at any time and had 47% more provider encounters in any 30-day period.
These percentages may underestimate the magnitude of Long Covid since many people can experience debilitating symptoms and still not see a doctor--especially for "soft" symptoms like fatigue, anxiety, depression, confusion, and insomnia. A large, survey-based study of positives in California and Nevada found that 42% reported at least one Long Covid symptom after 30 days and 24% did so after 90 days. Long Covid after 90 days is a more durable problem because at that point symptom prevalence decays much more slowly.
The newer and larger studies may begin to narrow the wide range of estimates of just what that prevalence is. While WHO's most recent advisory issued in February offers a low figure (10%) for symptoms after 12 weeks, WHO concedes that the published figures range very widely depending upon the methodology of the study. Its summary of various studies, most of them with small n sizes, demonstrates that this range is indeed all over the map.
Which sorts of maladies does Long Covid encompass? Almost everything under the sun, as all of the studies make clear. Victims suffer from respiratory problems (difficulty breathing, reduced lung capacity, reduced O2 diffusion); metabolic disorders (insulin resistance, high blood lipids; anemia; liver and kidney damage); cardiovascular issues (hypertension, rapid heartbeat, cardiac inflammation, clotting); digestive complaints (abdominal pain, nausea, gastric issues); nervous system disorders (severe headache, body aches, "brain fog"); and overall emotional and physical distress (chronic fatigue, anxiety, depression).
How long does Long Covid last? Due to the bewildering multitude of symptoms--and to the fact that some symptoms may get worse even while others get better--no one yet has a clear idea. The general consensus (see also here) is that the prevalence declines steadily from 30 days after infection to 90 days (3 months), but that symptoms thereafter stubbornly resist much decline.
The most exhaustive study to date on Long Covid's trajectory (global, with n=3,762) concludes that the afflicted are clustered into separate groups, and that those not recovered within 3 months present with a distinct mix of "moderate" symptoms that persist over time. This group "indicated that fatigue, breathing issues, and cognitive dysfunction (i.e., 'brain fog') were the most debilitating of symptoms."
Who is most likely to experience Long Covid? As I already mentioned, the likelihood goes up for those who suffer more serious symptoms during the acute infection. But recall: Even those who are asymptomatic after testing positive have a 19% risk of Long Covid--not that much lower than the 27% who are symptomatic but not hospitalized. There is also evidence that women (by gender) and whites (by race) are somewhat more likely to experience Long Covid.
Most of all, let me stress this: It is very clear that youth does not protect you from Long Covid. In fact, the risk may go down slightly at older ages. Most researchers say the most at-risk age range is 35 to 49, and positives in their late teens and 20s are probably more at risk than positives in their 70s and 80s.
Here's an age breakdown of perhaps the largest national survey of Long Covid symptoms, run by UK's Office of National Statistics (ONS). Note that the prevalence rate peaks at ages 35 to 49.
Several theories have been suggested to explain why the age profile of severe chronic Covid differs so much from that of severe acute Covid. One is survivor bias: An infection severe enough to kill someone past age 70 may allow that same person to survive at age 40--but suffer longer-term symptoms. Another is based on the premise that Long Covid constitutes a post-infection auto-immune disorder (just as Myalgic Encephalomyelitis or Chronic Fatigue Syndrome, which also occurs in women more than men, is often presumed to be a post-influenza auto-immune disorder). In younger people with more active immune systems, the ME/CFS disorder would therefore tend to be more severe.
I stress the undiminished risk of Long Covid among younger age brackets, because the dominant reason young adults offer for not getting vaccinated is their worries about vaccine side effects. Young adults often perceive the risk of vaccine side effects to be large compared to the danger they face from Covid-19 itself.
True enough: The young adult's risk of dying from acute Covid-19 is very small--though it's not nearly so small as the risk of dying from the vaccine, which is statistically indistinguishable from zero. Faced with such small odds either way, though, young adults can reasonably say: Hey, the odds of anything bad happening are tiny in any case, so why bother?
The young-adult risk of suffering from Long Covid, on the other hand, is quite substantial. And this ought to persuade young people to re-compute their odds.
Sooner or later, the vast majority of unvaccinated young adults will be infected. And once they are, as we have seen, their odds of incurring Long Covid are somewhere in the range of 2 in ten to 3 in ten. The odds of a serious or long-term side effect from the vaccine by contrast is 2 or 20 or (at the outside) perhaps 200 per million. That makes the vaccine roughly 100,000X to (at worst) 1,000X less risky than Covid-19.
And lest the young assume that their Long Covid symptoms would be relatively mild, the data don’t offer any such reassurance. To the contrary, the evidence suggests that younger adults are more likely than older adults to complain of cognitive dysfunction that prevents them from working. Ditto for complaints of palpitations, chest pain, and tachycardia. According to the FAIR Health study, 25% of all patients who saw doctors for post-Covid cardiac inflammation were ages 19 to 29.
Enough about the young. Let's pan out at this point and look at the broader question. What's the magnitude of Long Covid's impact on all Americans? Just how many people are we talking about?
Let's conservatively assume 20% of all positives eventually encounter Long Covid. As of the end of March 2021, the CDC's best guess is that 115M Americans have been infected. By now, we can assume that the number is up to at least 130M.
That translates into about 26M Long Covid sufferers. Over time, we can assume that some of this group will get better. But we can also assume that more of the unvaccinated will get infected, adding to the group. It is also possible that "breakthrough" Delta variant infections among the vaccinated may add to the group.
On this last variable—breakthroughs getting Long Covid—health experts assumed up until a few weeks ago that this flow would be relatively small. Today, we’re no longer so certain.
To begin with, estimates of the share of vaccinated people who become infected are ramping up quickly in light of (a) Israeli data on the rapidly declining efficacy of vaccines against the Delta variant among the first to be vaccinated, and (b) new CDC data on the stunning recent super-spreader event in Massachusetts—in which three-quarters of the propagation was among the vaccinated. Yes, vaccination will protect you from hospitalization and death. And that’s great. But it won’t protect you very well from infection.
Now we can move on to the next link in the chain: If you’re vaccinated and infected, what are your odds of getting Long Covid? Again, until recently, we had no data on this question. Now we have one study (again, from Israel) on 39 Pfizer-vaxed healthcare workers who were infected, mainly by the Alpha variant: Though most had mild acute symptoms, 19% had chronic symptoms lasting beyond 6 weeks. Admittedly this study does not tell us much—except that the breakthrough risk of Long Covid seems to be well above zero.
OK, now let’s move back to the main narrative… that is, to our 26M. A number this large is going to be consequential.
Think about the 47% rise in outpatient visits per week by Long Covid patients observed in the VA population. Then multiply that by 26M.
Or think about the impact on employment. In the global survey referenced earlier, of the 93% of Long Covid respondents who did not experience a full recovery by the end of the survey time period, 73% reported they were working on a reduced schedule due to their symptoms. That includes 22% who reported that they were unable to work at all due to their symptoms. Again, multiply these percentages by 26M. Could this explain a lot of the missing 3M in labor force participants that I recently discussed on The Macro Show? I bet it could.
As a demographer, I also think about mortality. Long Covid doesn't just make you feel bad. It can kill you--though the proximate cause of death will never be identified as Covid-19.
The Washington University study demonstrated with some precision that the mortality rate among all Covid-19 positive patients during the six months after their recovery from infection (that is, after 30 days) is 59% higher than among a demographically identical group that was never infected. This is the extra risk for all Covid positives--not just those with Long Covid. But we can reasonably assume that the risk of death is at least that high, and probably quite a bit higher, among Covid positives who experience chronic symptoms.
As I pointed out several months ago, we shouldn't be surprised by this finding. When SARS-CoV-1 hit the world back in 2003, there was widespread astonishment at its high case mortality rate. Yet another important feature of this coronavirus was seldom remarked upon: the high rate of chronic illness ("Long SARS") among the survivors. According to one study of a sample of SARS survivors, 30% had not returned to work two years after their infection; and average walking speed, a good surrogate indicator of expected mortality, remained far below normal. Why would we expect Covid-19, driven by another coronavirus (SARS-CoV-2), to be all that different?
Nor is it clear that Long Covid is all that different from the CFS/ME and related disabilities that are often noted after most severe viral outbreaks—and were widely observed for many years following the Russian Flu of 1889 and 1892 and the Spanish Flu of 1918-19. One thing we do know is that pandemic events (even excluding catastrophes like the Black Death) regularly inaugurate eras of less investment and lower real rates. That’s according to a recent paper by the San Francisco Fed. Another paper, by a team at the IMF, comes to the same conclusion, but it adds that they also trigger periods of greater social unrest and civil violence.
Did the long post-viral “shadow” of those earlier outbreaks cast a similar shadow over the post-viral economy? It’s possible.
Increasingly we read news stories discussing the Covid pandemic in the past tense--as if, in the vaccinated high-income world, it's pretty much over except for the outside risk of some novel variant breakthrough. All that's left for us now is to "rebound" back to where we were before.
But Long Covid paints a different picture, of a pandemic whose shadow is much larger than its death toll. It is, in reality, a pandemic that is still oppressing tens of millions of us and that will continue to affect our families, our healthcare system, our work lives, and our economy for many years to come.
| To view and search all NewsWires, reports, videos, and podcasts, visit Demography World. For help making full use of our archives, see this short tutorial. |
