Except the CDC decided to hang their hat (https://www.cdc.gov/media/releases/2021/s0806-vaccination-pr...) on the one published article that claimed to show vaccination was dramatically more efficacious than previous infection, despite that study contradicting multiple, better studies, and despite the study itself admitting in a footnote that the data was exceptionally suspect. Also note how the CDC hedges their bets in the press release, saying that vaccination is better than prior infection alone, which only begs the question.

Here's the rub: is previous infection as good as Moderna? Probably not, especially given the new variants. (Though last year studies were showing infection providing 90%+ immunity.) But nothing is as good as Moderna. (Except maybe Novavax?) Should we require J&J vaccine recipients to be re-vaccinated with Moderna? Nobody is demanding that, though very roughly speaking J&J-level immunity may be about where things stand now for the previously infected.

This seems to be another unforced err by the CDC. Not as bad as some others, but unforced and very politically costly. It also complicates things going forward because it seems we're quickly approaching the point where we'll need to begin requiring a 2nd round of vaccinations. The CDC will need every last shred of credibility and good will left for that initiative. It could have been in a position where it could say, "hey, we tolerated the previously infected skipping vaccination before, but now the situation has changed." Instead, they'll have much less capital to work with.

I'm curious why you think previous infection is not better than Moderna. If we grant that Moderna is better than the vaccine used in the study in the OP, can it possibly be a factor of 6 better? a factor of 12? Those are the kinds of numbers they're seeing in this study. Admittedly it's not yet peer-reviewed and maybe those factors get watered down or thrown out, but if they don't, it would look like the difference between vaccine efficacies are down in the noise compared to the difference between any vaccine and natural immunity.

> I'm curious why you think previous infection is not better than Moderna.

I was trying to be conservative. I don't have any hard opinion on this one way or another. It seems like a degree of precision we'd be lucky to achieve in the best of times--i.e. when the virus and demographics aren't constantly evolving. And it's a degree of precision that is unnecessary to establish in this particular debate, even if we could. It feels like bike shedding.

The question isn't which is better--infection or vaccination. The question is, do we demand that the infected be vaccinated even though the substantial and consistent weight of scientific evidence gives us little reason to believe the previously infected put the population at more risk than the vaccinated, especially relative to vaccines that we affirmatively accept as sufficiently efficacious? The only scientifically justified answer is an unqualified, "No".

There are innumerable other questions people can debate, but from a pandemic policy perspective there's no need to venture there. And, frankly, it seems imprudent considering the level of uncertainty surrounding everything--not just the quality and relevance of data, but even if you're technically correct one moment, things can go sideways the next moment, necessitating a change in policy.

I think you're stuck on the word "previously" infected. If the government says you must be vaccinated or pass an antibody test by Sept 15 to keep your job, what are people going to do?

Well, given how many people have already dropped out of the labor market, I suspect some people will use this as another justification not to work.

Thanks, good points.

You misread the study. It does not contradict any study that I am aware of and agrees with the study that this discussion is attached to in particular.

The CDC study you're pointing to says that among people who have been previously infected, those who have been given vaccines are less likely to be reinfected than those who have not.

There is no error from the CDC here, nor any lost credibility.

The CDC study doesn’t control for behavioral differences between the two studied populations, though, which weakens their conjecture that there is a clear causal relationship that confers more immunity to those who’ve had Covid+vaccine compared to those having had just Covid. It seems entirely plausible that people who’ve refrained from getting vaccinated after having Covid are probably not especially careful about their exposure, which can lead to higher rates of infection.

One could just as easily suggest the opposite: that people who have gotten the vaccine, feeling even safer, will engage in more risk-taking behavior because they have the added safety.

With the absence of evidence in either direction, it is unwise to draw either conclusion.

I didn't say anything about the methodology of this CDC study or the study that this comment section is attached to, only that they both say the same thing. They both say that getting the Pfizer vaccine after infection confers additional protection against reinfection. There is no disagreement between these studies' conclusions.

An alternate conclusion to your own is that people who get the vaccine did so in order to take advantage of the reduced restrictions on vaccinated people at the time.