COVID Transmissions for 2-24-2021
Vaccines WILL impact transmission
Good morning and welcome to COVID Transmissions.
It has been 464 days since the first documented human case of COVID-19. In the year 464, the Roman General Syagrius rose to power in Northern France. He would become the last Roman ruler of that region, losing his seat in battle to the Frankish King Clovis I, the founder of the Merovingian dynasty. This was a major event in the fall of the Western Roman Empire.
Today in 2021, I have some news stories for you—evidence that the Pfizer vaccine really does impact transmission, and a comment on a scary LA Times article about a new variant.
As usual, bolded terms are linked to the running newsletter glossary.
Keep COVID Transmissions growing by sharing it! Share the newsletter, not the virus. I love talking about science and explaining important concepts in human health, but I rely on all of you to grow the audience for this, which you can do by using this button here:
Now, let’s talk COVID.
Good evidence that the Pfizer-BioNTech prevents infection, not just disease
Further results from the SIREN study, a large cohort of UK healthcare workers, have given us a picture of what the Pfizer-BioNTech vaccine can do for prevention of the spread of SARS-CoV-2.
Specifically, the SIREN study collects data on this healthcare worker cohort that include “fortnightly” (it’s a UK study, after all) COVID-19 PCR tests, whether there are symptoms or not. This allowed the researchers to determine not just if there is a difference in the appearance of disease in vaccinated people, but also whether there is a reduction in infection.
Now, I’ve said before that infection and transmission aren’t the same thing, but there is a dependency relationship here. Someone who is never infected can’t transmit, so if we see a difference in infection rates, that means we’re going to see a reduction in the spread of disease. This isn’t the same thing as seeing lower PCR-based amounts of virus in infected people, which some other studies have shown. This is much more reliable, because an uninfected person definitely isn’t transmitting virus.
Now to the results, which are also found in this preprint: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3790399
A single dose of BNT162b2 vaccine demonstrated vaccine effectiveness of 72% (95% CI 58-86) 21 days after first dose and 86% (95% CI 76-97) seven days after two doses in the antibody negative cohort.
Look at that! In the period between the first and second doses, there was an apparent 72% reduction in the risk of infection (but the 95% confidence interval ranges a bit wider, from 58% to 86%, meaning we’re 95% sure that even with experimental error, the “real” value lies somewhere in that range). From 7 days after the second dose, the protection against infection—again, not talking about disease here!—was 86%, with a 95% CI from 76% to 97%. Wow!
For reference—but not comparison, as I’ll mention—in the Pfizer Phase 3 clinical trial, the efficacy against disease was around 95% after two doses and was 52% in the 21 days between first and second dose. But we can’t exactly compare these results to the ones from SIREN, because SIREN was doing prescheduled sampling of any infection, while Pfizer’s trial was testing for COVID-19 only when patients were symptomatic, for the results I mentioned. There is a little bit of bias in either approach so we can’t really cross-compare here. Not to mention that these are different cohorts. Still, it’s a good point of reference, because it tells us that the SIREN results for protection against infection are in the same general neighborhood we expect for protection against disease.
This suggests to me that the vaccines can reduce the odds of transmission substantially. I wouldn’t say that they can totally eliminate transmission, though, because of, again, a design limitation from the sampling in SIREN. The tests performed here were done fortnightly, as mentioned. It is possible to have a full clinical course of COVID-19 in less than 14 days, and I’m sure it’s also possible to have an asymptomatic infection that resolves in less than 14 days. Some of the “negatives” may be people who were infected with SARS-CoV-2 in between tests, but who cleared the infection before the test detected it. This means there may have been a period between tests when some of these “negative” patients were capable of transmitting virus.
However, we have an experiment and we have a control here. The control group was much smaller than the experimental (vaccinated) group, but if anything this would make me expect to see more infections among those who got the vaccine, since there are more potential chances to detect infections. Both groups were sampled identically. This means that people who did not get the vaccine were more likely to turn up positive for infection on this fortnightly schedule that people who did get the vaccine.
At the very minimum, that implies that the duration of SARS-CoV-2 infections is probably limited by the vaccine. For certain, fewer people who got the Pfizer-BioNTech vaccine had infections that lasted for two weeks or more than those who did not get the vaccine. The longer someone is infected, the more time they are producing virus, and the more opportunities there are for that virus to be transmitted.
This is similar to Dr. Erin Bromage’s formula for how likely an exposure event is to lead to infection: “Successful Infection = Exposure to Virus x Time.” Here we see a vaccine effect that at minimum impacts the amount of time someone may be able to transmit to others, and may also impact how much they’re able to expose others to the virus overall. That means we reduce the chances of encounters that produce new successful infections.
Let’s say I’m a susceptible person (I am, I think), and the non-vaccinated patient has an average course of infection lasting 17 days while the vaccinated patient has an average course of infection lasting 10 days (warning: numbers made up for illustrative purposes!). Let’s then say that by chance, I meet one of these two people on day 14 of their infection. I’d rather meet the vaccinated person, who on average has already cleared SARS-CoV-2 from their system. Wouldn’t you?
Iterate this process over every susceptible person in a whole population, and then ask yourself if you think there will be fewer infections with or without a high rate of vaccination. Based on these data, I’m pretty convinced the answer is going to be yes. In fact, I think there will be a great many fewer infections with good vaccine uptake, based on these data.
I may as well make the subtitle of every issue of this newsletter “more research is needed,” and surely more research IS needed, but this is very compelling to me. I already felt that the vaccines probably do restrict transmission. I was pretty convinced that the AstraZeneca-Oxford vaccine does it for sure, which means I’m somewhat confident the Johnson and Johnson vaccine does it, too. Now am I pretty convinced the Pfizer-BioNTech vaccine does it too, which means I’m somewhat convinced that the Moderna vaccine also does it. That leaves me feeling rather confident (to put a number on it, I’m maybe 80% sure) that the available vaccine arsenal will substantially restrict transmission of SARS-CoV-2.
It feels pretty fantastic to be able to say that.
Now, let’s turn to some other items of interest from the paper—the vaccine uptake:
Vaccine coverage was 89% on 5/2/2021. Significantly lower coverage was associated with 44 prior infection (aOR 0.59 95% confidence interval [CI] 0.54-0.64), female (aOR 0.72, 95% CI 45 0.63-0.82), aged under 35 years, being from minority ethnic groups (especially Black, aOR 46 0.26, 95% CI 0.21-0.32), porters/security guards (aOR 0.61, 95% CI 0.42-0.90),or midwife 47 (aOR 0.74, 95% CI 0.57-0.97), and living in more deprived neighbourhoods (IMD 1 (most) vs. 48 5 (least) (aOR 0.75, 95% CI 0.65-0.87).
There are a lot of numbers here, but I’m encouraged by the 89% coverage by 5th February (UK date format!). What I’m less encouraged by is buried within the technical part of the above block quote, which is expanded on in the results section:
…having a prior infection, gender, age, ethnicity, IMD score and staff group remained significantly associated with vaccine coverage.
OK, so here we see people who have been infected with COVID-19 were less likely to have gotten the vaccine. This doesn’t seem to have protected the unvaccinated group much, so I think this is weak evidence that if you’ve had COVID-19, you should still get vaccinated when it’s your turn. That’s interesting.
However, the really disappointing part is where gender, age, ethnicity, IMD (“index of multiple deprivation", a measure of socioeconomic status/poverty) and staff group also impacted likelihood of getting the vaccine. Basically, this tells me that societal inequality, as well as potentially a false impression of invincibility among younger people, led to disparities in vaccine uptake.
The societal inequality piece has been a theme in the pandemic. Racial biases and economic biases, sometimes manifested in what job a person has access to, has affected whether they’re likely to get COVID-19 and potentially whether they will die of it. Now it is affecting whether or not they can protect themselves from COVID-19—in this and in other datasets I have seen. That’s disappointing news and I think we need to take a hard look at those facts and sit with them.
Anyway, the vaccine effect results are good. So that’s something to feel good about.
It’s not time to worry yet about the California variants
Today, the LA Times ran an article about a variant of SARS-CoV-2 that emerged in California in May, which they blamed for the surge in cases in California that occurred this fall and early winter.
The problem is, the news story portrays this variant as new—May isn’t new—and acutely threatening. The spike in cases this variant is said to have caused is already subsiding.
Worse, while I have gotten used to papers making it into the world without peer review, I haven’t gotten used to what this article does: report on a story before the preprint is even available! Instead it tells us that the preprint will be available soon.
Let’s get real with this already. Newspapers are not the place that scientists communicate with each other to help find fact. Scientific journals may not be that either anymore, but we don’t put these things in the newspaper to create a public debate before things are figured out in the literature. This LA Times article is causing a panic without the perspective of the scientific paper, with its restrained voice, to reign that panic in.
To put it mildly, I don’t think that’s cool at all.
When the reprint appears, I’ll cover it here. Until then, the facts of the case appear to suggest that this variant’s detection explains events that have already happened, but do not give us too much to worry about for the future.
What am I doing to cope with the pandemic? This:
Watching: The Muppet Show
OK, pretty much everyone knows about The Muppet Show. Many of you will also know that it recently found its way onto Disney+ in its entirety. My wife is very into the Muppets, and I also think they’re great, so we’ve been watching through.
The thing is, I am not of a generation that had variety shows on TV much when I was growing up (well, unless you think of Saturday Night Live as a variety show). I hadn’t seen shows like Laugh In until very recently, when that show became available on Amazon Prime. This is the first time I’ve watched through The Muppet Show with a good sense of what else was on TV at the same time that it aired, and I think that has really made a difference in how I process it. Instead of seeming like some off-the-wall thing that we were lucky enough to have on TV, I now see how The Muppet Show actually “worked” from an entertainment programming point of view. The show exists in conversation with these other programs that were around in the 1970s, and it builds on a language that they created.
I’m sure to anyone who saw these other shows I’m not saying anything groundbreaking here, but if this sounds like new information to you, it might be worth looking up some other classic variety television and seeing how The Muppet Show draws on it and builds something new.
Reader Ellen Kushner, a fantastic author and neighbor of mine, gave me some advice on Twitter today:
Ellen, I hope you don’t mind my sharing this here, but since it’s publicly available on Twitter, I figured it would be OK.
This is good advice, and I’m going to take it to heart and tweet about the newsletter more. But you all can help with this too—if something from the newsletter strikes you, please quote it and share a link to the full issue along with the quote. I’d like to think I’m doing something that can help people, and you can help me help people.
You might have some questions or comments! Send them in. As several folks have figured out, you can also email me if you have a comment that you don’t want to share with the whole group.
Join the conversation, and what you say will impact what I talk about in the next issue.
Also, let me know any other thoughts you might have about the newsletter. I’d like to make sure you’re getting what you want out of this.
Part of science is identifying and correcting errors. If you find a mistake, please tell me about it.
Though I can’t correct the emailed version after it has been sent, I do update the online post of the newsletter every time a mistake is brought to my attention.
No corrections since last issue.
See you all next time.
Always,
JS
"The tests performed here were done fortnightly, as mentioned. It is possible to have a full clinical course of COVID-19 in less than 14 days, and I’m sure it’s also possible to have an asymptomatic infection that resolves in less than 14 days."
I tend to read CT early in the morning, while finishing my first cup of coffee. As a result, reading this I thought, "Why didn't they do antibody tests to see if someone was infected?"
And then, "Oh, right, vaccinated."
'However, the really disappointing part is where gender, age, ethnicity, IMD (“index of multiple deprivation", a measure of socioeconomic status/poverty) and staff group also impacted likelihood of getting the vaccine. Basically, this tells me that societal inequality, as well as potentially a false impression of invincibility among younger people, led to disparities in vaccine uptake.'
There's also a frequently-written-about phenomenon that members of some ethnic groups simply distrust the medical system. I have no idea if the UK experiences this to the same extent as the US, with our atrocious (in the literal sense--these were atrocities) history of things like the Tuskeegee Experiment and the Doctor J. Marion Sims inventing gynecology by experimenting on slaves without anesthetic.
Another wrinkle here: everyone in this study (if I read your summary correctly) was employed by the NHS. They certainly varied by SES, but that would inevitably correlate with educational level--in the healthcare business, physicians get paid more than nurses get paid more than nurses' assistants, to name three jobs. Did they correct for that?