Thursday Throughput: COVID Vaccine Side Effects Edition
[ThTh1] The big news this week is that several European countries have suspended distribution of the AstraZeneca COVID vaccine. The purported cause is a rash of blood clots that occurred after getting the vaccine. However, it’s really not clear that these things are vaccine side effects related at all:
Call it luck, chance or fate – it’s difficult to incorporate this into our thinking. So, when the European Medicines Agency says there have been 30 “thromboembolic events” after around 5m vaccinations, the crucial question to ask is: how many would be expected anyway, in the normal run of things?
We can try a quick back-of-the-envelope calculation. Deep vein thromboses (DVTs) happen to around one person per 1,000 each year, and probably more in the older population being vaccinated. Working on the basis of these figures, out of 5 million people getting vaccinated, we would expect significantly more than 5,000 DVTs a year, or at least 100 every week. So, it is not at all surprising that there have been 30 reports.
In the Phase 3 trials, about 38% of the vaccine recipients complained about side effects. But so did 28% of the control arm. Serious side effects were about 1% of the trial participants, but those incidents were actually slightly higher in the control arm. The UK has now distributed 10 million vaccines with less than 1% reporting side effects.
Those of us who write about science knew something like this was going to happen. If you gave 15 million people a glass of water, some of them would drop dead the next day. This doesn’t mean that water is killing them. It means that, in a large enough group of people, bad things are going to some people just by random chance. That’s just the nature of statistics. But some anti-vaccine advocates and … give me a phrase other than pro-virus advocates … don’t care about data. They care about stoking panic. And they will jump on anything to make their case.
I’m hearing 30 or 40 people have had their dishwasher break down after having the AstraZeneca vaccine
— Matt Leys (@mattleys) March 16, 2021
Now, to be fair, the symptoms people are worried about are unusual, as described here. And some scientists think it is worth looking into. But the gripping hand here is that even if the AstraZeneca vaccine does cause blood clots — and there’s no evidence that it does — the rate is very low; 30 events at most. You have to compare that to the dangers of 5 million people being exposed to COVID, which would leave about 20,000 people dead and several hundred thousand with long term health issues. This effect was brilliantly demonstrated by Penn and Teller in a video I’ve probably shown before (language warning).
Now imagine that demonstration with a bowling balls and you’ll get the idea of what we’re talking about here. This isn’t the sniffles. This is the deadliest disease we have faced in a century. I know people who’ve had significant side effects from the vaccine, described it as feeling like they were run over by a truck. I also know people who’ve had COVID who’ve described it as one of the worst experiences of their life. This disease has slaughtered at least 50,000 people worldwide in the last week alone (that we know of). Now is not the time to cater to panic. By suspending the vaccine, these governments are not only endangering people, but they are also giving legitimacy to the anti-vax disinformation campaign. And it will only get worse until they stop conceding the statistical high ground.
The simple fact is that these vaccines have been extensively tested and are as safe as our testing limits can tell. Yes, I think it’s prudent to look at possible long-term vaccine side effects. But keep in mind that those vaccine side effects won’t just magically appear after ten years; they should start showing up in a portion of the population very soon. Yes, I think it’s prudent to not vaccinate kids at this stage. But with new variants exploding into the European population, to stop distribution of a vaccine based on a side effect that is rare, unproven and statistically insignificant in the 15 million doses that have been given out so far is simply irresponsible.
And not just for today; but for the long term. Because this won’t be the last pandemic we ever face. And the infrastructure we’ve built in the last year will be critical to containing the next one.
[ThTh2] Over the last year, there has been intense debate over the measures used to contain COVID. This debate, like seemingly all debates, has gone to the extremes with lockdown advocates claiming their opponents are literally killing people and lockdown opponents claiming their adversaries are implementing fascism.
But it doesn’t have to be so black-and-white. Charles Cooke makes a reasonable case that much-derided Florida Governor Ron DeSantis has done just a good a job of containing the virus without burning his economy as much-praised Gavin Newsome has. The two states have similar death rates, but Florida has a much strong economy right now because DeSantis didn’t institute the kind of restrictions Newsome did.
While I appreciate the perspective (and the NYT article linked within), I don’t these comparisons are quite appropriate. California was hit very early in the pandemic when the disease was much more lethal than it is now. I have little doubt that those early shutdowns saved the state from suffering the fates that New York and New Jersey did.
It’s also not clear that government mandates and private behavior coincide. Cell phone mobility data has repeatedly shown that many states without shutdown orders saw just as big declines in activity as state that had them. You can order people to stay at home all you like. Or you can tell them to go out and party. But how they will act tends to be a more individual decision.
Even more importantly are huge variations in local conditions. COVID spread is affected by weather, populations density, variants, when the disease hit and peaked and a variety of other factors we probably haven’t even considered.
We have spent the last year blundering in a dimly lit room trying to contain this thing. And while we’ve learned some things — outdoor activities are safer than indoor ones; masks work; crowds are bad — it will probably be years before we know who did better on this. If we ever do.
Every few months, we go through this cycle of trying to determine who did a better job of containing the pandemic. Which would be great except that the pandemic is still going on. Conditions can change. Areas where the virus was contained suddenly become hotspots. Hotspots calm down. States that we’re doing well suddenly aren’t. Just look at Europe. For months, we were told they’d done a better job than we had. But the US is now rolling out the vaccine three times faster than the EU.
The only countries that have really done well — so far — are either islands like Australia and Taiwan or Pacific Rim countries that have significant cultural differences. For most of the world, it seems like societal changes are hard while technological challenges are easy (for reasons we discussed on Twitter here). And the results have been periodic and almost random explosions of the disease.
As we enter what I hope is the final stages of the pandemic, it’s important to keep in mind that there is no winning in this game. There are only degrees of losing. Hopefully, we will get to the point where we have the luxury of looking back, figuring out what worked and planning appropriately for the future. In the meantime, it would behoove us to remember that no one wants to kill grandma, and no one is establishing the Fourth Reich through stay-at-home orders. There are politicians, pundits and writers who are grifting on this by spreading disinformation and hysteria, yes. But most of us are just fumbling our way through an unprecedented situation.
[ThTh3] People were alarmed when recent data from Spain showed a shocking rise in COVID deaths among young children. Turns out, it was a data entry error. The system was leaving off the 1 on victims who were over 100.
[ThTh4] The case of the missing eclipse.
[ThTh5] Yes. This is an actual thing that exists:
Fordite (also known as Detroit agate) is old automobile paint which has hardened sufficiently to be cut and polished. pic.twitter.com/wqBtsgBJJC
— Diane Doniol-Valcroze (@ddoniolvalcroze) March 10, 2021
[ThTh6] Want to hear a robot fire a laser on Mars? Sure you do.
[ThTh7] The Cow and the Camel sounds like an Aesop fable. In this case, it may be a black hole devouring a star from the inside out.
ThTh5: And it’s a limited resource, because the auto companies have changed how they coat cars! (Also: there are a number of craftspeople who have stock of Fordite and make jewelry out of it. I have a Fordite pendant I ordered after seeing the original stuff online because it’s just a cool thing)Report
Does it have to be from the auto plants? Seems like something that is easy enough to make at home, if you have the time and the paint. If it’s valuable enough, I could see building a bench top robot that would make some for you.Report
yeah, but part of the coolness factor of the pendant is that it was from a plant that manufactured Corvettes; a homemade simulacrum wouldn’t have the same history factor
At one time also, my dad had a piece of “fused” asphalt from when lightning struct a road; apparently he saw the strike and picked it up later because he wanted to thin section it and look at it (it had gone glassy). Now I don’t know what happened to it, if he even kept it…Report
I can see that.
The software that I work on is often used for modeling automotive paint application, specifically to make sure that when they dip the whole body into the pain tub, air bubbles aren’t trapped and causing voids in the paint application.Report
Our very own Clare Briggs had a vaccination cartoon.
The Spain thing was nuts. How did this take this long to catch?Report
About four hours, apparently; the story I heard was pediatricians saw a report of a huge surge in infant and young-child deaths, said “wtf”, and figured out what was going on almost as soon as the report was published.Report
I was worried that it was months. 4 hours is an admirable amount of time to catch this.Report
With my old manager hat on, sort of my stock question for this kind of situation: “If other people immediately notice that the data is badly out of whack, and you can figure out that problem in four hours, why the f*ck did you put me in the position of looking stupid? Why didn’t you see that the data was badly out of whack?”Report
“The computer said that that was what the number was.”Report
I am old, and perhaps out of touch with contemporary practice, but this was not an excuse. Even if it was just software, something asked the question of whether the result was obviously suspect.Report
Hey, I’m just telling you what *I* have been told by testers.Report
Serious side effects were about 1% of the trial participants, but those incidents were actually slightly higher in the control arm.
I believe that the term you should use here is adverse event. Whether it’s actually a side effect of the drug is uncertain, especially when it occurs at higher rates in the control arm.Report
We’ve got a very small n here but we can’t rule out that the Moderna vaccine increases the chances of a lightning event.
Covid vaccine volunteer, 72, struck by lightning after getting Moderna injection.Report
Good health makes people more attractive.Report
I got my second Pfizer shot on Friday afternoon. The side effects of the second dose were very mild. Nothing on Friday afternoon, On Saturday I had a SLIGHT flu-like discomfort (some joint pain, like when you have a cold, I had some headache, very mild, and no fever). On Sunday I woke up feeling perfectly fine and went biking for three hours.
Please note that on Friday night I had a work related celebration dinner, including two Old Fashioned, wine, and a vodka shot (the latter ordered behind my back, I’ll point out).
Any relationship between the Old Fashioned and the mild Saturday malaise is entirely Pfizer’s fault.Report
This is what happens when you hold aloft the syringe like Mjolnir during a lightning storm.Report
I’d get the vaccine without a moment’s hesitation. But let’s be honest. It’s a vaccine with a nearly zero chance of killing me for a disease with a nearly zero chance of killing me. I’d take it for the same reason I’d let women and children have a seat on the bus or on a life raft, but this is much closer to a bus situation than a life raft.Report
Even if Covid doesn’t kill you, it can cause serious damage to your lungs, heart, and other vital organs. I can’t find definitive numbers, but that seems to affect 5-10% of Covid victims. The vaccine is a much better bet.Report
Of more pertinance than your own nearly zero chance of being killed, is the fact that you can pass this on to someone else with a very much more than nearly zero chance of dying from it.
So ja, vaccination is the Right Thing to Do.Report
Not that much more likely, but yes, that’s why I plan on getting vaccinated.Report
Well, this is somewhat better than “Solar Roads”.Report
Good commentary. Here are some thoughts, add-ons and disagreements.
“I don’t [sic] these comparisons are quite appropriate. California was hit very early in the pandemic…”
And Florida has an older population. Like all comparisons they are useful, but can be taken too far.
“It’s also not clear that government mandates and private behavior coincide…. But how they will act tends to be a more individual decision.”
The governor of Florida didn’t say people COULDN’T wear masks, or that they HAD to go to crowded beaches and bars. The point is that one governor used the state to prohibit behavior more than the other and, assuming all else equal (admittedly not a valid assumption), the less restrictive state had better economic outcomes and similar medical outcomes. This is an extremely useful, albeit limited, data point. Data is good.
“We have spent the last year blundering in a dimly lit room trying to contain this thing.”
And as the fog begins to clear we can start to see early returns on our actions. Perhaps we can make a hypothesis that warnings and information can control the virus as effectively as edicts and lockdowns in some cases. Reasonable, albeit still unproven, hypothesis.
“As we enter what I hope is the final stages of the pandemic, it’s important to keep in mind that there is no winning in this game. There are only degrees of losing.”
That is pretty much what the NR author said: “Coronavirus has been a disaster and a challenge for everyone. Nobody has got it completely right, because it was not possible to get it completely right. ”
The question has always been how to respond to the threat in the most judicious manner considering that there are costs not just from the disease but from our cures and controls. I believe the term in medicine is “iatrogenic harm” — causing harm while trying to help. The optimal solution would balance the human health results with the economic, educational, and psychological iatrogenic harms.
“Hopefully, we will get to the point where we have the luxury of looking back, figuring out what worked and planning appropriately for the future.”
Why would we want to wait until afterwards if we can start to learn earlier? Isn’t this deciding what we should have done with the barn door after all our cows are gone?
I am not arguing that Florida or California handled it better. But the major media made their opinion pretty clear every single day, passing judgment for one side and against the other. A lot of people on this site did the same. It is pretty much expected that the party that the major media despises would respond with a counter opinion/judgment/cheer in their own little conservative media space.
The only thing I despise more than democrats are republicans (and vice versa). But beyond all the political tribalism, there is some evidence starting to emerge on the costs and benefits and pros and cons of various responses to the viral threat. Maybe some places will learn from it and handle it better moving forward. Others will just play to their tribal base. Lives will be saved and lost, jobs will be preserved and destroyed.
But, considering everything, did politicizing this issue help us or hurt us?Report
One of the things that we will probably never know is what the actual contact rates looked like in any of the states. For example, we know now that in New York, during the supposed lock down, there were religious funeral observations with hundreds of unmasked people crowded shoulder-to-shoulder. Outside my area of expertise, but it seems more likely that we could use the data to work backwards from the known case rates to get an actual contact rate, and then use that to judge the effectiveness of policies like closures.
One of the things I hope the experts learned in my state is that locking down elderly care facilities is not as effective as anyone thought it would be because of the sizeable number of staff that work multiple jobs at multiple facilities. We closed the front doors but not the back doors.Report
My contact rate (15 minutes indoors closer than 6 feet) plummeted and my contact circle narrowed considerably. Personally it would not have been much different for me or any family member on contact rate if my gov (Newsom) had made recommendations and best practices rather than edicts.
Indeed, I suspect tens of millions of conservatives and Christians intentionally stopped wearing masks and increased contact rates just to defy the orders which they honestly resented. The heavy hand of the California state may have backfired. Now they distrust everything the state says as they have mobilized a resistance.
Again, I am not arguing for Florida vs California, just suggesting that we are getting useful feedback on stringent lockdowns vs public service messages and minimal regulatory adjustments. The final results will include secondary and unintended effects, perversely including that a firmer hand may actually lead to more rather than less contact.Report