Aleks pointed me to this article, which reports:
Epidemiologic data consistently show strong protection for young children against severe COVID-19 illness. . . . We identified 3,126,427 adults (24% [N = 743,814] with children ≤18, and 8.8% [N = 274,316] with youngest child 0–5 years) to assess whether parents of young children—who have high exposure to non-SARS-CoV-2 coronaviruses—may also benefit from potential cross-immunity. In a large, real-world population, exposure to young children was strongly associated with less severe COVID-19 illness, after balancing known COVID-19 risk factors. . , ,
My first thought was that parents are more careful than non-parents so they’re avoiding covid exposure entirely. But it’s not that: non-parents in the matched comparison had a lower rate of infections but a higher rate of severe cases; see Comparison 3 in Table 2 of the linked article.
One complicating factor is that they didn’t seem to have adjusted for whether the adults were vaccinated–that’s a big deal, right? But maybe not such an issue given that the study ended on 31 Jan 2021, and by then it seems that only 9% of Americans were vaccinated. It’s hard for me to know if this would be enough to explain the difference found in the article–for that it would be helpful to have the raw data, including the dates of these symptoms.
Are the data available? It says, “This article contains supporting information online at http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.2204141119/-/DCSupplemental” but when I click on that link it just takes me to the main page of the article (https://www.pnas.org/doi/abs/10.1073/pnas.2204141119) so I don’t know whassup with that.
Here’s another thing. Given that the parents in the study were infected at a higher rate than the nonparents, it would seem that the results can’t simply be explained by parents being more careful. But could it be a measurement issue? Maybe parents were more likely to get themselves tested.
The article has a one-paragraph section on Limitations, but it does not consider any of the above issues.
I sent the above to Aleks, who added:
My thought is that the population of parents probably lives differently than non-parents: less urban, perhaps biologically healthier. They did match, but just doing matching doesn’t guarantee that the relevant confounders have truly been handled.
This paper is a big deal 1) because it’s used to support herd immunity 2) because it is used to argue against vaccination 3) because it doesn’t incorporate long Covid risks.
For #3, it might be possible to model out the impact, based on what we know about the likelihood of long-term issues, e.g. https://www.clinicalmicrobiologyandinfection.com/article/S1198-743X(22)00321-4/fulltext
Your point about the testing bias could be picked up by the number of asymptomatic vs asymptomatic cases, which would reveal a potential bias.
My only response here is that if the study ends on Jan 2021, I can’t see how it can be taken as an argument against vaccination. Even taking the numbers in Table 2 at face value, we’re talking about a risk reduction for severe covid from having kids of a factor of 1.5. Vaccines are much more effective than that, no? So even if having Grandpa sleep on the couch and be exposed to the grandchildren’s colds is a solution that works for your family, it’s not nearly as effective as getting the shot–and it’s a lot les convenient.
Aleks responds:
Looking at the Israeli age-stratified hospitalization dashboard, the hospitalization rates for unvaccinated 30-39-olds are almost 5x greater than for vaccinated & boosted ones. However, the hospitalization rates for unvaccinated 80+ is only about 30% higher.
Herd immunity isn’t something one can sensibly argue for or against. It is a fact of nature. Why do you think each wave comes in and out?
Stressing about “long Covid” is pointless for the same reason. Everyone is going to get Covid repeatedly throughout their lives.
Daniel:
I don’t understand your logic that it is pointless to “stress” about something that is likely to happen. We’re all likely to get cancer too; that’s one reason why there’s so much research being done on cancer treatments. If something is very rare, then, yeah, I see the argument that it’s pointless to stress about it. But if it’s a common outcome, then that really is something to stress about, no?
This was my first thought too. It’s not like we should be stressing about shark attacks, or being struck by falling airplane parts. So don’t stress about the unlikely stuff. If we are also not supposed to stress about the likely stuff then hey, cool, nobody should stress about anything!
But it’s actually kinda-sorta right. We’re all going to die eventually, but it doesn’t make sense to stress out about that fact because what good does the stress do?
To the extent one gets to choose what to get stressed about, you should choose to worry about the things you have some control over. So I think the likely-unlikely axis is the wrong way to choose what to stress about; it’s the unavoidable-avoidable axis that carries most of the information. And in that context I think “long COVID” is certainly on the list of things worth stressing about. It may be true that in practice just about everyone is going to get COVID eventually, but each person has quite a bit of control over how many times that might happen and perhaps what strain to get.
Only tangentially related, but since I’m here I may as well tell this story. I have a friend in her early sixties who is in fair but not really good general health. She was pretty careful about COVID for the first year and a half but was among the first people I knew to get fed up with COVID restrictions and start getting careless. (From the standpoint of the country she’s not unusual at all, but most of my friends have been among the more COVID-cautious). Sure enough, sometime in mid-2021 she got COVID…and it was pretty bad. Not go-to-the-hospital bad, but bad enough that she was pretty much incapacitated for a month and then had several _months_ when she felt groggy and slow-witted and low on energy and so on. Eventually she was able to pretty much resume her normal life…and she still wasn’t careful about COVID, and she caught it again, and it was almost as bad the second time as the first time: another three weeks in bed, another couple of months of grogginess and bad sleep and fuzzy thinking and so on. She’s really just getting over it…but she still isn’t being careful about COVID! She’s right back to eating in restaurants and doing grocery shopping without a mask and so on. To me, I can see dropping all precautions if you think you’re in the category of people who are going to have a mild COVID response — I’m in that category myself, and if I didn’t hang out for a while almost every day with my 81-year-old mom I’d probably drop just about all precautions — but if you’ve had a terrible time the first time you’ve been infected, and another terrible time the second time, I’d think you’d be pretty careful about not getting it again! But not everyone feels that way. It’s strange to me.
Phil:
Yes, I agree that the things to stress about are the things that are not too rare and which can be changed. I assume that covid outcomes can be changed: there’s been lots of progress on vaccines and treatments already, right? Also, as you point out, even if it’s true that “everyone is going to get Covid repeatedly,” there’s still a lot that people can do to reduce the frequency of that “repeatedly” thing.
> Yes, I agree that the things to stress about are the things that are not too rare and which can be changed
Probably implied in both of your comments, but I think there’s another important vector: the severity of the impact from the risk. Due to some cognitive biases, low probability/high damage function risk is often among the hardest to get a handle on (for example climate impact resulting from the higher range of climate sensitivity), but (imo) severity of the damage needs to be a key, explicit consideration.
It depends on lifestyle.
If you are highly social and enjoy going to bars, restaurants and entertainment venues – as your friend seems to – then even though you need it more, caution is much more disruptive. If you’re not social and spend a lot of time outdoors, caution is easy, there’s almost nothing you need to do except wear a mask in stores and keep vaxed.
Well, perhaps “stress” was the wrong term. We’re all going to die, and it’s often pointless to tell ourselves it’s pointless to stress about that.
I don’t want to get into an argument here about “long Covid”, so I’ll say only that the condition is poorly defined, and there *really* isn’t good evidence that multiple reinfections increase likelihood of sequelae. The norm for other diseases is the opposite.
I’m not convinced that – except for a very privileged few who can have huge amounts of control over how they live – personal choices make much difference to how many times one gets Covid. How much do you think you could do to control how many colds you have?
So there’s a choice between focussing much of one’s life on avoiding Covid, or accepting that this isn’t going to work, and/or isn’t worth the candle. In as much as I consider these latter options wiser, I’ll say it’s “unwise to stress”.
> and there *really* isn’t good evidence that multiple reinfections increase likelihood of sequelae. The norm for other diseases is the opposite.
Can you elaborate?
From what I’ve seen we don’t know with COVID, let’s say with a 2nd infection, that the likelihood of developing long COVID decreases from that individual (i.e., 2nd) infection – but even if we did…
Let’s call the % who get long COVID from first infection X%, and the amount that % decreases with the 2nd infection Y%: it would still be X% + (X% – Y%).
Seems the BEST you could say is that subsequent infections don’t ever result in long COVID (or that Y% = X%) which certainly seems not to be the case).
Maybe a 2nd case prevents long COVID from developing long after a person recovered from a first case – but is there evidence of THAT with other diseases? in other words, are you really saying that people who have had repeat infections with other diseases are less likely to ever develop chronic sequelae (compared to people who have only gotten sick once)?
Hi, Joshua. I think the reifying of “Long Covid” is a mistake, but I’ll leave that aside for the sake of argument.
You write “we don’t know with COVID, let’s say with a 2nd infection, that the likelihood of developing long COVID decreases from that individual (i.e., 2nd) infection”, so I think we agree on the state of the evidence.
But I take your logical point: the probability of getting sequelae can only increase or (at best) stay stable with each new infection.
But in practice what’s needed is a cost-benefit analysis: is the probability of each new infection causing grave sequelae high enough to justify the extraordinary measures it would take to avoid getting re-infected? I’d say no.
Is there such a thing as herd immunity to this virus? There are lots of credible reports of recurrent illness and transmission of illness in people who got vaxxed, boosted, and had natural infections. This virus seems to throw up variants with novel immune pathways at a prodigious rate. Herd immunity may not be achievable.
Naturally, I do want to emphasize that vaccination does blunt the perniciousness of the disease which is valuable.
> Is there such a thing as herd immunity to this virus?
At some level, that depends on the definition of “herd immunity” your referring to. One of issues with this pandemic is that different people use different definitions.
I’ve seen where some have even gone so far as to argue that we’ve moved on and out of herd immunity as a function of the rate of spread (or around the R number 1 threshold).
Yea, I don’t get the “herd immunity” discussion.
Just like every other respiratory virus, you get temporary mucosal immunity that wanes after a few months to years (roughly proportional to disease severity). If enough people recently got exposed then the transmission rate temporarily goes down. If it was otherwise your mucus would be so thick with antibodies/etc that you couldn’t breath.
Then the related issue is intramuscular vaccines don’t even trigger that mucosal immunity. So they protect against the virus spreading through the blood but not infection/transmission.*
Everything has played out as expected based on prior knowledge and everyday experience, so I don’t see why there is any discussion. Yet there seems to be mass confusion surrounding the issue.
* There are reports of IgG antibodies leaking into the mucosa from the blood when levels are very high, but only to concentrations 1/1000 lower and without detectable neutralization. So maybe you will see reduced infections for a month or so due to that.
In February 2021, a U.S. medical school professor of nephrology (who is also a clinician) as well as two non-physician epidemiologists in Israel warned me that it was vital that everyone get vaccinated and continue to mask and isolate for an indeterminate amount of time. Well, long enough to eradicate COVID19 sufficiently that it does NOT become endemic. If it became endemic, they said that people would contract it repeatedly, with some portion suffering from long COVID and its sequelae such as cognitive impairment, shortened telomeres, and multi-organ injuries of varying degrees of severity, all of which would shorten life. Even those who didn’t get long COVID would suffer from similar sorts of permanent damage and have 10 to 20 years lopped off their otherwise expected lifetimes.
It was difficult for me not to take them seriously, because they are prominent but not medical ‘social influencers’. One is a Greek emigrant to my native New Mexico, and has an impressive knowledge of R, stochastic modelling, and even Andrew Gelman’s STAN thing. They all resemble Gelman in that they are bright, wry, cautious, quantitative, analytic, experienced, and highly educated. None are Trump supporters like deplorable me. (Gelman already knows about me, so I don’t try to hide it.)
Yet Daniel Hadas stated bluntly:
What YOU said in response seems well-supported. I am not going to ask, “But how could my nephrologist and the two Israelis not have known this, regarding vaccination for respiratory viruses?” Instead, I want to ask you and Daniel whether you concur with the grim prognosis of endemic COVID as described by my experts? IOTW, what are the implications of endemic COVID? Are we all going to die 15 years before we would have otherwise, even with innumerable vaccine boosters?
P.S. This is for Andrew. I was reading Tablet Magazine, the slightly right-wing but not neocon publication for Jews like me. In this July 2022 article about children, COVID19, and vaccination, it states that 75% of children in the U.S. have already been infected by the virus, via CDC Morbidity Mortality Weekly 71/17e3 (29 April 2022). Maybe living with children (for parents and grandparents) affords the same protection that living with indoor/outdoor pets, usually dogs, offers children: A more robust immune system due to continual low level exposure to dirt, dander, and assorted grime?
Post-post script for Andrew: Thank you for your 2007 era post about how E.T. Jaynes was not all that. I think you said “no guru”. I had thought similarly, but am sometimes overwhelmed by Jaynes fanboys. Your post validated my suspicions, and comforted me.
I don’t see why that would happen. The most unique thing about this virus was the hysteria surrounding it. Besides that we should assume it will play out like similar viruses. What are the reasons they believe this one is exceptional?
By that I mean everyone will get it every year or two, and if it gets into a nursing home or similar then it can send 10% or so of people over the edge. Then there will be a small minority of people that seem to have related long-term issues that no one can really explain.
I mean most of the time no one would bother to test for this or that virus, but when they do:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2095096/
So it comes down to what they think makes this virus unique, besides the response to it.
Anoneuoid –
> What are the reasons they believe this one is exceptional?
Seriously?
Reasoning from personal incredulity can be sub-optimal.
Maybe that over 6 million have died (not to mention how much severe illness) over a couple of years might have something to do with it?
Just a guess.
Of course, that doesn’t necessarily make the relevant biology “unique” along many dimensions, but the high transmissibility and virulence are characteristics of this virus that make it particularly notable.
I really doubt the people being referred to relied on a correlation = causation argument when obviously many things changed in response. We all know the pirates vs global warming meme.
Originally there was a lot of talk about “happy hypoxemia” and then later that the mechanism was probably blood clots (in particular “micro-clots” in the capillaries that mimic the gas bubbles seen in altitude sickness), but I haven’t heard much about that lately.
Anoneuoid,
Thank you. I don’t want to die any earlier than necessary.
Anoneuoid –
> I really doubt the people being referred to relied on a correlation = causation argument
I’m not entirely sure what that comment meant (I’m not even totally sure if it was a response to my comment).
But maybe you are saying that you think there’s no clear evidence that COVID causes severe illness and death, only that there’s evidence of an association – in which case the 6 million deaths “from” COVID (actually, I’m reasonably certain that’s a undercount) reflects a conflation of correlation and causation? Do you think that the association is only incidental?
Well all I said is that by default it should be like similar viruses, but that could be wrong of course.
Originally there were reports of the strange symptoms that I mentioned above, but for some reason that is no longer being focused on. It seems to me that is something that should be figured out.
But I would be interested in why the people you mentioned were so convinced about organ damage being common, etc.
Thera prolly an interaction effect (or positive association) between higher rate of COVID infections and less severe disease.
Not suggesting how much thar explains, but it may be a favor.
Arrrggghhhh.
There’s..that…it may be a factor
As always, I think theorizing about mechanism of causality should necessarily be a part of the discussion of interpreting this kind of analysis.
As for mechanism of causality…
Early on in the pandemic, when COVID “skeptics” were heavily promoting the effects of cross-immunity in support of minimizing the impact of COVID and questioning the benefits of NPIs in terms of minimizing spread, some heavy-hitter virologists were saying that “cross-immunity” would not likely minimize spread but might well reduce the severity of the infections that do occur.
Much has changed since early on in the pandemic. I don’t know if that view has changed – but it would be consistent wirj the higher rate of infection/less severe disease link.
“As for mechanism of causality…”
Isn’t there a strong proposed causal mechanism in the “significance” statement?
“Epidemiologic data consistently show strong protection for young children against severe COVID-19 illness…Parents of young children—who have high exposure to non-SARS-CoV-2 coronaviruses—may also benefit from potential cross-immunity. ”
As always in statistical analysis, the results are clear: it’s possible there might be a connection between lower incidence of severe COVID-19 and frequent exposure to young children. This confirms that what we thought might be true still really might be true. Or it could be that people without children go to bars and restaurants more often, where extended exposure to an infected person may cause more severe illness.
Chipmunk –
> Isn’t there a strong proposed causal mechanism in the “significance” statement.
>> : it’s possible there might be a connection between lower incidence of severe COVID-19 and frequent exposure to young children.
Intersting. You see that as a mechanism of causality, I see it as a statement of association.
I’m not sure what a non-subjective line would be to draw between correlation and mechanism of causality.
RE: “I’m not sure what a non-subjective line would be to draw between correlation and mechanism of causality.”
That non-subjective line would consist of objective evidence. Which is ultimately an empirical question, no?
Is there evidence of the association? If so, does that evidence more closely fit with a specified, biologically and physically plausible causal mechanism than it does with non-causal associations due to confounding, etc.?
Theories, associations and speculative causal pathways we have in abundance. Evidence that is clean enough to admit convincing statistical analysis remains thin on the ground 2-1/2 years in and counting.
There might be some protective effect from getting your kids’ colds every year, but as a parent of young children, I would probably chalk this up to parents getting tested more, so the data is capturing more of the milder cases in the parent group to begin with. I’ve been able to get tested (even when not experiencing symptoms) at the same time as my kids when we’ve taken them to their pediatrician’s office to get tested, but if it’s just me getting tested I’ve found that going to a testing kiosk to be more convenient than getting an appointment with my doctor. Many of those without kids going to testing kiosks and getting mild cases discovered won’t necessarily show up in the data set. Examining the testing behavior of each group would help solve the issue of whether the results are being driven by testing behavior.
My hunch is that it might be hard to be social (to meet a partner) and then get pregnant while having some kind of significant health issues, especially immune system-related ones. I wonder if there is data on that.