Vaccines - Let’s Not Give Them To Everybody

Oh god, not another post about The Plague. Ug. I’m so sick of it! But I have to squeak my little thoughts yet again because some of what I see in the news is so maddening! I’m going to throw a quick disclaimer that I’m not necessarily solving the world’s pandemic problem — that’s not my job — but I can point out some things to think carefully about.

First on the list is vaccines! Wooo! A magical cure. Perfect. But there are problems. Over on Marginal Revolution they’re really pushing the idea that distribution plans for these two dose systems are badly conceived. Let’s take pharma at face value for a second and assume that after two injections you are completely immune from The Plague. But after only one injection? Let’s be pessimistic and say it offers 70% of the protection. (Let’s not even worry about what that exactly means because, god knows, articles on the topic don’t.) The question can be best thought about on an island of 1000 people who get a shipment of 800 vaccines. Should 400 people become essentially immune? Or should 800 people become almost immune? Well, I’m not going to do the math, but I will point out that some math should be done.

Don’t forget to throw in the adverse effects being widely reported. It seems that spending a couple of days in bed is not uncommon after the second shot. At least according to the FDA.

Interestingly there are rumors that side effects seem concentrated on the first injection if you’ve already had Covid19.

It doesn’t stop with a single dose strategy though. Sensational articles are making white people FOMO at the mouth because minorities may be prioritized over cranky old people of no color. That is a big sensational distraction but it is also subtle. Here are the CDC’s distribution priorities. They correctly get the first three right:

But in my opinion the final one is wrong.

I get it — disadvantaged people have enough disadvantages. But this really should be a subordinate priority. We don’t give a flying shit who gets an injection; what we care about is ending the pandemic. We know the CDC can be so coldly heartless and focused on such large scale pragmatic utilitarianism. We know this because all of their advice "for" victims of Covid 19 is essentially tips for how the victim can limit their danger to others.

Are you following me here? What I’m trying to say is that if you want that latinx hotel maid to not die of Covid19, you must prioritize ending the pandemic. It may seem that arbitrary/random/equitable allocation of vaccines is as sensible as anything but that is a failure of imagination. A more sensible plan is to find strategic points where the pandemic is spread and focus on that. Check out this random person’s blog post from 2017 which summarizes Season 2 Episode 5 of the great Canadian TV series "Travelers". In the show people from the future come back in time to sort out problems. In this episode they’re curing a plague. Check this plot out:

Note that some of their medical practices, as shown above, are probably too advanced; in the show, these are people from the future with perfect knowledge. But in reality they are Canadian people writing a TV show script and they’re showing more good sense than the CDC! They’ve realized that a few key people are the biggest problems — super-spreaders — and need to be targeted to most effectively save <insert your favorite demographic group here>. Flight attendants (men/women, gay/straight, any color) probably should be vaccinated! After a year of this virus do we really still wonder who is spreading it? Moralizing should not be a part of the model — vaccinate the homeless, drug dealers/addicts, prostitutes, clergy, politicians, etc. The details need to be worked out, but is working out some details too much to ask with the biggest calamity of our lives?

And now we come to the main story that’s been driving me especially crazy. A couple of weeks ago I posted about the new flood of rough articles attempting to address the critical question of how long does immunity last? For people who asymptomatically tested positive? For people who had serious symptoms? And, importantly, for people who have been vaccinated?

Now that more people are awake to the issue I’m getting the feeling that a huge amount of speculating is coming from one particular article.

Do antibody positive healthcare workers have lower SARS-CoV-2 infection rates than antibody negative healthcare workers? Large multi-centre prospective cohort study (the SIREN study), England: June to November 2020

This was the "medical" article that the "news" articles I cited last time cited. Basically it seems that a lot of people skimmed the abstract and walked away with this quirky — likely wrong — fact.

I went ahead and read the whole thing — it’s written in regular enough English and not really that long. If anyone can tell me where that 83% number comes from, I’d really appreciate it because I could neither find nor calculate anything close to that in the body of the article.

What I did find was this.

(There is confusion about what "83% lower risk" and "99% lower odds" means; consider a 20% off sale made 50% better; is that 30% off or 70% off? But…) No matter how you interpret that, it would seem that having been infected previously is very helpful in not getting infected again.

This study is pretty good given the chaos of the last year, but it leaves much to be desired. For example, it doesn’t account for severity of disease presentation. Stronger symptoms may imply a stronger immune response — remember those cytokine storms?

The cohort in this paper was 84% female! While I understand why a bunch of health care workers might skew this way, that doesn’t exactly improve the clarity of the study.

I am also skeptical of reported odds of "re-infection" because of the large amount of information pouring in from people who have "long haul" Covid. Are symptoms of the same infection flaring up? Were they never fully "recovered"? These tests seem to be antibody tests and if there’s one thing we know about antibodies, it is that the immune system is complicated. I don’t have answers but that is a genuine complication. Here’s a direct quote from the paper in question, "…some of these potential reinfections were likely residual RNA detection at low population prevalence rather than true reinfections…" This is interesting too, "Investigations have been restricted by the limited availability of data and samples from historic infections, with most swabs discarded without sequencing, preventing the genomic comparison between infection episodes required to confirm a reinfection." Or how about "…use a range of testing platforms and assays, there is the possibility of misclassification bias." Or "Some of those PCR results, especially early in the epidemic, may have been false positives or laboratory contamination episodes…" At least they know they don’t know.

The citations provide a long list of interesting reinfection reports. I’m still predicting that the vaccines will not be radically better than previous live symptomatic exposure. Yet even if vaccines are somehow better, there will still inevitably be issues with efficacy.

My reading of the actual numbers is that out of the 6614 people they identified as having some kind of test profile suggesting infection only 2 of them "probably" were reinfected (13 had suggestive symptoms and 44 total is the maximum that could possibly have had "it" by any loose definition). That’s a reinfection rate of 0.03% (to 0.67% assuming the loosest standards of reinfection). Compare that to an infection rate of 318 among 14173 unprotected subjects, 2.24% (to 2.91% if you add antibody seroconversions).

The results do seem unambiguous and they say as much.

I’m going to let that be the final message. If you will be comfortable hanging around with someone who was vaccinated a year previously, you should be safe with me today.

UPDATE 2021-02-20

There’s another dimension to the topic of vaccinations and that is whether an immunized person can spread the virus while being at low risk of developing any symptoms themselves.

This article in Smithsonian Magazine frames the problem like so.

The article also brings up a very interesting possibility that I don’t hear much discussion about.

Too bad the link to that study is broken for me. But this is hardly supportive of the idea that vaccination protects society far better than a naturally acquired immunity attained by surviving the disease.