Thread by @SalvMattera

It is my belief that, eventually, many people - maybe most people, maybe all people - will develop severe long COVID. I don't know how long it will take, but it is only a matter of time. Here's a thread on why I think this, and why I might be wrong.

It only requires that you accept two premises: (1) that people will catch COVID many times in their lives, and (2) that every COVID infection carries with it a material risk of developing long COVID.

The first premise used to be controversial but now, heading into 2024, I think it's fair to say that it is self-evident. Many people - vaccinated or not - have had many COVID infections. If you yourself haven't had 2, 3, 4+ infections, you almost certainly know someone who has

The second premise is less certain, I think, but the evidence for it continue to pile up. Back in 2020 and 2021 many people doubted long COVID even existed - almost no one makes that argument anymore, even skeptics. Everyone knows that long COVID is real and a problem.

There are many studies, now, showing that it can happen following reinfection, and anecdotal reports of people getting it after third 3rd/4th/5th infection despite recovering from their previous infections just fine.

This is broadly acknowledged by most major medical authorities, including the CDC and WHO. It's basically the consensus at this point. What's not clear is how the odds of long COVID change with reinfection - perhaps the odds decline over time

But, anecdotally, I know of plenty of people who developed long COVID after their 3rd+ infection. So the odds can't be that bad.

So it becomes quite simple: if every infection is a dice roll, given enough infections, everyone will eventually develop long COVID. But how might this be wrong?

I have shopped this argument around in various forms to COVID skeptics, and setting aside the obvious nonsense (like the idea that long COVID isn't real, or it's just a vaxx injury) the responses more or less fall into one of two buckets.

The first bucket is what I would call no response. Statements like: "Obviously, that can't be true." Or, "You can't just extrapolate like that" or "that's a conspiracy theory". These are not real arguments, and the fact that serious people make them actually disturbs me

The second is to point to survey data, most commonly the CDC Pulse Survey or the NHS "Prevalence of ongoing symptoms following coronavirus" survey and note that the number of respondents to these surveys reporting long COVID has been flat or maybe slightly down over time

I have to admit that this is definitely evidence that long COVID isn't becoming more prevalent over time. After all, if it was, wouldn't these surveys show it? The thing is, I just don't find this evidence to be strong, especially given what is at stake here.

There are many problems with these surveys. To start: they're just surveys. They are literally just asking people if they have symptoms. The problems with this approach are obvious just from the data itself - for example, the CDC Pulse survey suggests that only 57% of adults have

had COVID. In reality, based on blood tests, this number is 80%+. So there are a lot of people who don't think they ever had COVID that we know actually did (or maybe they just won't admit it). If they don't think they ever had COVID, how will they know that their health problems

are long COVID? Other issues include the fact that long COVID sometimes takes weeks following infection to set in, and that symptoms vary so much, that many people - including MDs - don't know what it is, or what to look for

Long COVID tends to wax and wane over time. So someone might think they're getting better for a while, only to have their symptoms return later. It's not clear how this is reflected in these surveys.

Many doctors who specialize in treating long COVID have also noticed a bias where people think they're getting better when in reality, they're actually not - they just adjust their lifestyle to their new baseline. Again, unclear how this is reflected in surveys

There's also the fact that the surveys themselves have only been around for a little while, and in fact, the NHS discontinued their survey back in the spring, so at best, the UK has been flying blind.

I don't think these surveys should be completely ignored - just that they should be put in context and handicapped as being relatively weak evidence. Certainly not strong enough to base policy on.

What would be strong evidence? I can think of two sets of data: (1) take a large sample of people who have had 5 or 10 confirmed COVID infections. What % of them have long COVID? If the answer is "not many" then it's unlikely most people will develop long COVID

(2) Devise a predictive model that would allow you to predict, with a high degree of accuracy, who will develop long COVID and who won't. Some models have been discovered, but they're based on things like gut bacteria or prior exposure to other viruses which means

They're not actually of any true predictive value in a grand sense, as everyone's gut bacteria changes and people are constantly getting exposed to different viruses in the wild.

So we don't have that data, and since testing has been scaled back, governments around the world have made it almost impossible to obtain. It seems the best we can do is wait and see.

But I'm an optimist. And I think that eventually someone will solve this problem. The best way to find a solution is to try to get people to see what's at stake.

The risk is asymmetric to the downside. In other words, if I'm right, and many or most people develop long COVID, the consequences would be disastrous. If I'm wrong, then it doesn't really matter one way or the other.

And since it's entirely plausible that many or most people will eventually develop long COVID, and the consequences of that being true are so severe, everyone should just assume that it is true, and act accordingly to try to stop it until it's proven otherwise.

Adding to this thread my threads analyzing different pieces of evidence: