Are many of us immune to Covid now?

[tassojunior]

https://www.nejm.org/doi/full/10.1056/NEJMp2026913

Not to downplay the expected huge increase in covid cases starting in November, but an interesting possibility has surfaced in a new paper in the New England Journal  of Medicine today (also noted in the NYT). I've long wondered about how "viral load" factors in both transmission and creation of immunity with covid and now there's a paper on it. Surprisingly though there is still zero research or statistics. This virus seems to be highly contagious and very long-term airborne and surface-borne. By masks cutting down on the viral load someone inhales, they may be de facto vaccine-makers.

The unproven idea is inspired by the age-old concept of variolation, the deliberate exposure to a pathogen to generate a protective immune response. First tried against smallpox, the risky practice eventually fell out of favor, but paved the way for the rise of modern vaccines.

Masked exposures are no substitute for a bona fide vaccine. But data from animals infected with the coronavirus, as well as insights gleaned from other diseases, suggest that masks, by cutting down on the number of viruses that encounter a person’s airway, might reduce the wearer’s chances of getting sick. And if a small number of pathogens still slip through, the researchers argue, these might prompt the body to produce immune cells that can remember the virus and stick around to fight it off again.

Some people have speculated up to 40% of people in areas that have had covid "surges" may already have immunity. If so, maybe the expected surge will not be as widespread as predicted, at least in already-hit areas, and new cheap fast antigen tests may show more immunity than expected. We should already know this question but don't seem to. For now Europe is the best predictor of the US since they went through their surge before us, have now completely re-opened everything, and are seeing rapidly rising cases. If it matches the 1st wave there, it would mean there is not as much immunity from casual exposure as hoped.  

Getting antigen tests to see if one's probably immune is important. Of course ultimately the vaccine is critical but if there is already a decent community immunity the vaccine's social goal of creating herd immunity is helped a lot. The first of the $5 instant antigen tests are supposed to be appearing this month. 

Interested in what @Riobard  thinks too.

[boiworship]

Anecdotally, there’s some reportage, especially from East Asia, that O blood types may also have some degree of immunity. The few people I know who’ve contracted Covid have A+ blood. I’m O+ and have never had the flu in my life. 

[Riobard]

When you say antigen I think you mean antibody, @tassojunior

Or maybe you mean a positive antigen test result predicts immunity, so for the sake of argument equals prospective immunity?

In contrast, antibody tests are a better marker of possible immunity as antibodies wrt CoV are detected longer. Detectable antigen has a short shelf life, high chance false negative, and not the ideal strategy for exposure surveillance relevant to magnitude of community immunity.

I’ll try to review the NEJM article shortly. 

Edited September 10, 2020 by Riobard

[Riobard]

22 hours ago, boiworship said:

Anecdotally, there’s some reportage, especially from East Asia, that O blood types may also have some degree of immunity. The few people I know who’ve contracted Covid have A+ blood. I’m O+ and have never had the flu in my life. 

Anecdotally means a small number of testimonials or observations, not a solid study finding.

The bit of research in Asia and Europe on either susceptibility to getting CoV or severity of Covid-19 disease, across variable blood alleles, is methodologically flawed. I have read the few related papers and have not been concerned about blood type as a relevant predictor. 

Besides, there are factors that do legitimately contribute to variance in catchable and morbidity, and that can be regulated far more easily than altering blood type (which is impossible outside of Transylvania).

Susceptibility to infection vs acquiring immunity, by the way, are not the same thing. Blood type does not confer immunity at the binary level: immunity yes/no. 

Edited September 10, 2020 by Riobard

[Riobard]

Canada has had a reasonably high dose of CoV, large abatement but new recent upticking, and now the curve trend puts us at 100th out of 198 globally. The evidence is convincingly accurate that the exposure to date here is in the range of .7% to 1.1%. This represents an ascertainment adjustment of 3- to 4- fold the reported case tallies, both to date and in terms of rolling average of new case incidence.

The event risk algorithm models in USA estimate ascertainment bias as greater than Canada and propose true case report tallies should be subject to adjustment metrics ranging from 5 times to 10 times the actual detected cases. 

Edited September 10, 2020 by Riobard

[Riobard]

Reported infection USA to date has reached 1.9% population. Ascertainment adjustment puts it at 9.5 - 19%. I doubt that any of the up to 80% remaining population is spared susceptibility, but presentation can be presymptomatic, asymptomatic, paucisymptomatic, or symptomatic on a severity gradient.

Edited September 10, 2020 by Riobard

[Riobard]

Louisiana, at reported 3.3%, is the highest state tally with a potential range of exposure up to 33%. Isolated on its own might be considered to be approaching 50% of the exposure prevalence needed to categorize as a reasonable threshold of herd immunity. But there is just the one partial flimsy wall west of it and it does not barricade state lines. 

[tassojunior]

19 minutes ago, Riobard said:

Louisiana, at reported 3.3%, is the highest state tally with a potential range of exposure up to 33%. Isolated on its own might be considered to be approaching 50% of the exposure prevalence needed to categorize as a reasonable threshold of herd immunity. But there is just the one partial flimsy wall west of it and it does not barricade state lines. 

But testing is not random but of the more likely so the # exposed and immune is a guess even with some rationale. We just dont have the mass testing yet to more precisely know how many have gained anti-bodies of those not obviously infected. The irony is many of those who will refuse a vaccine will become infected and either die or become part of the immune herd. A poor choice of risk that may or may not stop herd immunity. Interested in what you find. Thx.

[Riobard]

Oh you only need a reasonably cross-sectional representative sample, not mass testing, to infer antibody-verified exposure/immunity estimated within a tight enough confidence interval to be statistically valid. A quarter of a percent of the population should do it, 15,000,000? Statisticians can compute sample sizes needed to extrapolate to the overall population. Anyway a lot of this has been done already and I am not inclined to dig into or look for what I have already read.

Try searching ‘USA covid true prevalence’ or ‘USA Covid antibody prevalence’ or ‘USA Covid immunity estimate’

[Riobard]

A little off track, but BTW the recently published JAMA study on CoV and Vitamin D in Illinois is considerably methodologically flawed.

There is much misleading research that gets published in supposedly respectable periodicals and the spurious findings get subsequently imported into media releases and poured into gullible trusting brains. 

Edited September 10, 2020 by Riobard

[Riobard]

OK, the NEJM article is very convincing and well written. The hypothesis cannot be tested in a human trial, but retrospective and animal model studies can further investigate. 

It value-adds to the importance and legitimacy of face-covering. 

We have mortality:recovery ratio curves all over the world that could possibly be triangulated with the temporal introduction and application of more universal widespread masking. 

However, the main content of the piece does not match well with the title of this new forum  topic.

[Riobard]

2 hours ago, Riobard said:

Oh you only need a reasonably cross-sectional representative sample, not mass testing, to infer antibody-verified exposure/immunity estimated within a tight enough confidence interval to be statistically valid. A quarter of a percent of the population should do it, 15,000,000? Statisticians can compute sample sizes needed to extrapolate to the overall population. Anyway a lot of this has been done already and I am not inclined to dig into or look for what I have already read.

Try searching ‘USA covid true prevalence’ or ‘USA Covid antibody prevalence’ or ‘USA Covid immunity estimate’

Erratum, 8,000,000 not 15,000,000

Edited September 10, 2020 by Riobard

[AdamSmith]

55 minutes ago, Riobard said:

Erratum, 8,000,000 not 15,000,000

&! if a doctor made that error...

[Riobard]

56 minutes ago, AdamSmith said:

&! if a doctor made that error...

Hahaha ... doctors get more leeway to backpedal. we only get 60 minutes on this message board to revise before exile from the related post field. 

[AdamSmith]

1 hour ago, Riobard said:

Hahaha ... doctors get more leeway to backpedal. we only get 60 minutes on this message board to revise before exile from the related post field. 

Course that ‘leeway’ only comes from doctors having to pay huge liability-insurance premiums. 

[Riobard]

7 hours ago, AdamSmith said:

Course that ‘leeway’ only comes from doctors having to pay huge liability-insurance premiums. 

With more than 3 billion Americans, bound to be lawsuits. 

[Riobard]

It wiil

14 hours ago, Riobard said:

We have mortality:recovery ratio curves all over the world that could possibly be triangulated with the temporal introduction and application of more universal widespread masking. 

However, as I look at several countries’ graphs plotting mortality:recovery ratio across time it appears that many have no alteration at all at any point in time you would estimate if and when masks introduced. Mask impact may also get buried within overall strategic mitigation effects.

The real story is the enormous difference in death rates comparing countries. Reporting is not standardized and is a ‘shit show’. Mortality rates depend on the differences among nations in incorporating recorded true deaths from CoV, presumed (how much CoV tipped the comorbidity ones to death), and what proportion of temporally contextualized excess deaths may be attributable to CoV.

[Riobard]

1 hour ago, Riobard said:

With more than 3 billion Americans, bound to be lawsuits. 

Because my estimated figure for antibody assay sample size should be 800,000 ... is everyone asleep or stuck on their ‘socials’ in this lecture?

 

[stevenkesslar]

8 hours ago, Riobard said:

The real story is the enormous difference in death rates comparing countries. Reporting is not standardized and is a ‘shit show’. Mortality rates depend on the differences among nations in incorporating recorded true deaths from CoV, presumed (how much CoV tipped the comorbidity ones to death), and what proportion of temporally contextualized excess deaths may be attributable to CoV.

What's particularly interesting is that the second wave spreading across Europe is a lot less deadly.  Nobody seems to know why, for sure.  I've read articles that speculate about different strains of virus and virus mutations that may make it less fatal.  My take away so far is it is mostly about the demographics of who gets sick.  This CNN article sums up what appear to be the facts pretty well.

Young people are driving a second, less-deadly surge of Covid-19 cases in Europe

That article is about a month old.  So we know more now about the fatality rates of this second wave.  The contrast in Spain is particularly striking.  The pattern is the same across Europe, but I picked Spain because their second wave looks a lot like their first in terms of case loads.  They peaked at about 10,000 cases a day, with the two peaks roughly five months apart.

When you look at the fatalities, there's no comparison.  With the March caseload peak in Spain, fatalities peaked a few weeks later, as would be expected.  Spain had just under 1000 deaths a day at the peak.   The second wave in Spain appears to have peaked in late August.  So the maximum deaths should be hitting right about now. The recent one week moving average in Spain is about 60 deaths a day.  On the face of it, the virus appears to be about 90 % less deadly. 

You can look at France or Italy or Germany and the basic pattern - lower fatality - is the same.  France already has blown past it's Spring peak on number of cases.  In the Spring, they had up to 1500 deaths a day.  Now it's more like 30.

I'm assuming a big part of this is that they're catching a lot more of the asymptomatic or minor cases now than they were in the Spring, due to mass testing.  And some of it may be due to knowing more about treatments.  Even so, the contrast in fatalities is striking.

This article is only a few days old and gives anecdotal pictures of what's causing the surge in various European countries:

'Not a game': Europe pleads with young people to halt Covid-19 spread

While it's not a game, it's also obviously not really bad news for Europe.  And for any country that can get its shit together on a national strategy for managing the risk. 

Part of what's surprising to me is that young Europeans who are getting COVID-19 don't appear to be spreading it to older Europeans - so far.  Other than a few articles, I haven't bothered to try to learn what may be driving this.  But I suspect mask wearing and social distancing protocols are probably a big part of it.  Meaning the Spanish 20-somethings may be partying without masks in bars, or in college dorms now that they are going back to school.  But older adults aren't going into those bars.  And they are wearing masks when they are around the 20-somethings.  Whatever is driving it, this has been going on for well over a month.  And the fatality rates are not spiking like they did in the Spring.

I think we know already that colleges that are open for in-person classes in the US are already seeing spikes.  The thing that I keep reading that makes sense to me is that young adults on college campuses should stay there.  If they bring COVID-19 home for Thanksgiving or Christmas it could be a huge national shit show.

When a vaccine does appear, it's going to be very interesting to see how it plays out.  I'll leave politics out of this, other than to say that trust in the efficacy of any vaccine has already been compromised in the US.  And we don't really know what natural immunity means for people who had COVID-19, or some exposure to COVID-19, or past exposure to some other type of Coronavirus.  We certainly don't know what artificial immunity means for people once we get a vaccine.  Nor do we know what herd immunity means due to some combination of the two.

I'm with Fauci and his common sense approach.  He keeps saying that the good news is that we know when we actually try to manage a spike by following certain protocols - like masks and social distancing - we can drive it down.  So what Europe is showing, even if it is haphazardly, is that there is a way to manage this so that young adults can do what they want to do without killing thousands of older adults every day.  I'm in no way encouraging it.  Spain is obviously correct to be telling young adults this is not a game.  But it is a fact that, at least so far, what's playing out in Spain and France right now is nowhere near as awful as what played out this Spring.

The silver lining in the cloud of this pandemic is that it is not the Spanish flu.  The second wave of that one was by far the deadliest.  And it was particularly deadly to young adults.  

I think most people could really care less about the scientific nuances of this.  Including me.  They just want to know whether they are going to live or die.  Or go broke from hospital bills, or losing their job.  So the good news is that while this is not a game, we ought to be able to figure out better how to manage it so that it is also not a death sentence like it was for lots of older people in the Spring.

Edited September 11, 2020 by stevenkesslar

[stevenkesslar]

 

I'm posting this as an addendum to what I posted directly above.  Presumably Gottlieb knows as much about the 30,000 foot view as anyone.  And I'm in sync with what he says here.

It would be nice to think that past exposure to other Coronaviruses has built up some type of immunity in lots of people.  But we don't know that, as he says.

Month's ago on Daddy's I was setting my hair on fire whenever anyone spoke up about the idea that we could go for herd immunity and magically put the vulnerable - particularly people in nursing homes - in bubble wrap for a year or so.  My argument was what Gottlieb is saying here.  If you have broad community spread, it's only a matter of time until it seeps into places like nursing homes and jails.  That certainly describes what happened in the real world - all across Europe and the US - this Spring.

It's not quite working out that way in Europe now.  Some part of it has to be that in the Spring it crept in before anyone knew what was happening, or was prepared to prevent it from creeping in.  So now maybe Europe is better prepared.  Or maybe there is more natural immunity.  But Gottlieb could also be right that it just will take a few months, as opposed to a few weeks. 

That is sort of what happened in the Sunbelt.  At one point everybody breathed a sigh of relief that caseloads weren't spiking in Georgia or Florida or Texas.  And then they spiked.  They spiked pretty much the same in California, which did have a more sober public health message and was more cautious about reopening.. 

My sense is that part of it is simply human nature.   People don't take it seriously until it really hits home.  Once it hit home in these Sun Belt states, it seems like a lot of people changed their tune and were just more cautious. If young people are being less cautious, it's obviously because they just aren't seeing the direct health consequences to them or their peers.  If they go home for Thanksgiving and their Granny is dead by Christmas, that will be a game changer.  Hopefully, we don't need to kill Granny to learn our lessons.  

[Riobard]

The kicker about Spain is that in neighbouring Portugal (proximity perhaps a good comparator basis) the ratio of daily deaths to new cases has been fairly constant, mortality keeping pace proportionally with infection incidence.

(time out ... I need to attend to something please do not comment yet as I will come back soon)

I decided to not elaborate other than to say there are a few reasons why case incidence to death ratio is a poor metric for the nature of the virus. Had a few paragraphs more but lost them.

Edited September 11, 2020 by Riobard

[Riobard]

 @tassojunior, what did I ask? You just knocked out a few complex points as I was adding them, by entering an emoticon reaction during the time I was editing.

Edited September 11, 2020 by Riobard

[Riobard]

However, I will add a thought about national specificity that was not snipped out by accident above because I had not yet written it.

Globally there is so far no evidence of waxing/waning/tides/waves. Granted, the temporal variations in case incidence onset and in death incidence for the collective of nations get smoothed out in the global metrics.

Death escalated quickly early but globally appears to keep pace with incidence. That pattern suggests considerable variation in incidence:death ratio because, assuming Spain is not unique in lower death reports ... for the average global curve pattern there would need to be other countries with high death rates to offset Spain data, etc. Then the 64-dollar question is accounting for why the virus is selectively death-sparing by country. 

If Spain is the main driver for the minor downturn in death globally, it remains unconvincing that age is a major factor because age distribution globally is likely similar at this point. Spain, like many nations, does not provide data on recovery:death ratio and that missing metric compromises interpretation. [plus I add a graph next post]

Stratifying patterns across 198 nations may be somewhat futile. 

Edited September 11, 2020 by Riobard

[Riobard]

Spain has tended to report mortality in latent data dumping fashion. 

[tassojunior]

18 hours ago, Riobard said:

 @tassojunior, what did I ask? You just knocked out a few complex points as I was adding them, by entering an emoticon reaction during the time I was editing.

Very sorry, I didn't realize an emoticon would wipe out most of your saved draft. As a matter of fact when I just reopened this comment draft all but the first sentence had disappeared so it may have something to do with new software. I appreciate your insight. 

So what do you make of New York and Germany? New York had the surge of all surges but has now re-opened with no surge while Germany has never had a surge and just re-opened over a month ago, including all schools, bars, etc with no new surge. (Lombardy could be substituted for NY for an all-Europe comparison). In the US the "hotspots" of a month ago are having rapidly declining cases while the new surge areas are rural areas previously clear of Covid. 

When Covid first started many people , including Merkle, said it was so contagious that eventually 70% of us would be infected whatever we did, and it was a matter of locking down to maintain steady hospital space and give time for better treatments (and eventually vaccines).  It does seem as if both from mortality and infection rates, the places with previous surges are more immune while previously unaffected areas eventually get their surge. But Germany seems to disprove this. I 'm concerned because Czech, especially southern areas, has never really been hit by Covid much and had a severe lockdown but is now wide-open suddenly. 

Similarly, here's a NY v. CA comparison:  https://www.mercurynews.com/2020/08/16/new-york-corralled-coronavirus-whats-californias-excuse-for-case-surge/