Hacker News Reader

Comment by gfodor | Hacker News Reader

I brought up the theory in other threads that the vaccine greatly reduces the chance of developing an infection but, conditional on infection, outcomes are the same. Got massively downvoted for being anti-vax, of course, despite the point being that if true we should be aggressively vaccinating but stop telling vaccinated people they get "less sick" and if they develop any symptoms to take it as seriously as they would if they were unvaccinated.

We're pretty screwed: tribalism has eaten through things enough that it's basically impossible to discuss these things.

View on HN

> conditional on infection, outcomes are the same

We know this isn't true, though. Repeatedly, since the data from the first EUA for the Pfizer vaccine.

https://www.cidrap.umn.edu/news-perspective/2021/09/covid-19...

> The study authors note that unvaccinated people in the Delta-dominant era are more than 10 times likely to be hospitalized from COVID-19 or to die from it than vaccinated people are.

https://www.reuters.com/article/factcheck-vaccine-reduce/fac...

> Meanwhile, most recent COVID-19 deaths have been among the unvaccinated. In Italy, 99% of COVID deaths since February were not fully vaccinated. Indonesian officials found unvaccinated people more than three times more likely to die of COVID-19 than those vaccinated.

> In the U.S., unvaccinated people represent nearly 97% of severe cases, according to the White House COVID-19 Response Team.

"Conditional on infection" - thanks for demonstrating the error.

https://drrollergator.substack.com/p/damned-lies-and-vaccine...

That's a quote from your comment.

You skimmed that article, didn't you?

> The death rate if infected was always going to be higher in the vaccinated groups if most of the vaccinated were those likely to die in the first place.

and in the comments, from the author, further explanation:

> BUT. When you vaccinate everyone, you wind up reducing infections in the people who were never going to die. The young. That means the IFR in the whole population of vaccinated people is higher.

The author is demonstrating issues with looking at IFR as a metric. The fact that a country (the US) with 65% vaccinated has 97% of severe cases in the unvaccinated speaks pretty solidly to efficacy at controlling severe disease.

It's kind of interesting you're accusing me of skimming the article, when based on the timestamps of these posts I can prove you literally skimmed the article.

One of the main points of the article is that the early Pfizer data is not consistent with the hypothesis that the vaccine provides additional protection after a person has been successfully infected.

Do you actually understand the argument?

If you can refute this, I welcome it. I am not adopting the position of the article. I am frustrated that every time it is surfaced I get the same garbage responses and vitriol, because the proper response to the theory, if true, is to be more conservative in how we think about the protection of the vaccine and stop telling people it turns COVID into a cold.

I've seen the article before; you'll find comments at the bottom pointing out that antivaxxers have been misreading it, and the author acknowledging that but opting not to revise.

Well I'm not an anti-vaxxer. I actually helped review this article before it was published. Can you provide anything constructive here or are you going to keep responding the way you have so far?

"vaccine reduces the chance of developing an infection but, conditional on infection, outcomes are the same"

That is a spectacularly wrong interpretation according to all the data.

I would like to update my priors if possible that this theory has been refuted. Can you provide counter evidence?

Note that the theory does not influence the degree to which one should seek to get vaccinated. If you think that it does, you don't understand it.

They do get less sick though. There's a clear difference in outcomes even if the viral load at a certain point in time is the same.

I'm not sure that adds up - if that were the case, vaccinated people with breakthrough would have the same outcomes as infected unvaccinated people.

But that does not seem so - the outcomes of vaccinated people are way better even when they end up in hospital.

This paper would just show that viral load reduction is not the mechanism how this happens - which I don't know what mechanism would that be then - but the outcomes are still different.

> But that does not seem so - the outcomes of vaccinated people are way better even when they end up in hospital.

I'd like to update my priors so can you provide any info on this?

A smallish study out of India on hospitalized-only patients focusing on mortality specifically:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8313045/

> A total of 332 (28.4%) patients died. Among fully vaccinated, 12.5% (23/184) died while it was 31.4% (309/984) among unvaccinated. Being fully vaccinated was associated with lower mortality than either unvaccinated (Unadjusted OR 0.3, 95% CI 0.2 to 0.5, p < 0.0001) or partially vaccinated (Unadjusted OR 0.3, 95% CI 0.2 to 0.5, p < 0.0001) status. There was no significant difference between unvaccinated vs. partially vaccinated (Unadjusted OR 1.2, 95% CI 0.9 to 1.6, P = 0.3).

Helpful, thank you!

(Its worth noticing exactly what happened here, and your previous responses a) citing information which was irrelevant to the theory I raised and b) you jumping to conclusions about my intent/agenda.)

It's also worth noticing this took about 30 seconds to find via a quick Googling.

Well, I said it was helpful, I didn't say it was conclusive. I too have done such quick googles. I was trying to be kind. Thanks for throwing it back in my face though.

It is extremely worth pointing out that you have been complaining about potentially inaccurate communication on covid risks statistics while promoting a theory that you haven't put the most basic research into and that has a large amount of contradictory data.

If you truely are trying to engage in good faith communication regarding covid, pointing this out should cause you to re-evaluate your approach.

You were probably downvoted for being incorrect rather than being antivax. Either way I'm not certain how it relates to the article. Viral load being the same doesn't mean that hospitalization rates or mortality rates are.

The science very clearly shows that your chances of severe illness / death are much lower for vacinated covid patients. This has been widely shown to be true for quite a while.

I would like a citation or which shows that breakthrough cases that lead to hospitalization for vaccinated people have better outcomes from that point forward than unvaccinated who are hospitalized.

That's an entirely different claim than the one you made (and the one you're replying to), which was predicated on infection, not hospitalization.

I didn't make any claims.

I am vaccinated + boosted.

Stop presuming I'm anti-vax and demonstrating the exact dynamic I mentioned. If you can provide useful information around either conditional probability functions (positive test, or hospitalization) I welcome it. But you haven't cited literally anything, just responded to try to score points, I guess.

They said:

> The science very clearly shows that your chances of severe illness / death are much lower for vacinated covid patients.

You replied (emphasis mine):

> I would like a citation or which shows that breakthrough cases that lead to hospitalization for vaccinated people have better outcomes from that point forward than unvaccinated who are hospitalized.

These are substantially different scenarios.

Correct. As acknowledged, if you read my responses.

In both cases, the public messaging is that vaccination provides substantial benefits throughout the entire funnel.

The point is that the public messaging is damaging if it turns out that the conditionals are misleading. I don't know how many times I have to say it, but I would like to see compelling evidence to dial in my priors. The links you have shared so far do not talk about this problem.

Edit: Maybe before replying again aggressively, you take a stab at repeating the overarching argument back to me, and the consequences of if it were proven true. Hint: if it were proven true, the result would not be to reduce the incentives of vaccination whatsoever.

You aren't doing any better job of communicating than the people who you are trying to correct and the message you are spreading is more easily used to prop up harmful misinformation.

I have seen very few if any claims that say "vaccinations reduce your chances of death if you are hospitalized."

I have provided a citation showing that vaccinations reduce your chances of hospitalization if infected. This is also easy to infer from the wide number of studies that show higher (per time) rates of protection against hospitalization than against infection.

The effects may disappear in any numbers (such as those discussed in the article you linked) from places / times were vaccine availability is limited and resticted to vulnerable groups. This is due to a combination of the selection bias and the small numbers of infections among the vaccinated.

Well, you've moved your goal posts from post infection benefits to post hospitalization benefits. I don't have any data on that specific question, but I do have data relating to your original one:

> Among 43,127 reported SARS-CoV-2 infections in Los Angeles County residents aged ≥16 years, 10,895 (25.3%) were in fully vaccinated persons, 1,431 (3.3%) were in partially vaccinated persons, and 30,801 (71.4%) were in unvaccinated persons. Much lower percentages of fully vaccinated persons infected with SARS-CoV-2 were hospitalized (3.2%), were admitted to an intensive care unit (0.5%), and required mechanical ventilation (0.2%) compared with partially vaccinated persons (6.2%, 1.0%, and 0.3%, respectively) and unvaccinated persons (7.6%, 1.5%, and 0.5%,

https://www.cdc.gov/mmwr/volumes/70/wr/mm7034e5.htm

Thank you. I didn't move the goalposts. You used the term "patients", which implies admission, so it sounded like you were making claims about post hospitalization benefits.

It seems like it may be the opposite of what you're saying, though. Vaccinated are just as likely to be infected, but are infected for a shorter time and do tend to get less sick. Natural immunity from a previous infection still reigns supreme, though.

Does it mean that a vaccinated person who caught the virus and developed minor symptoms has a better chance to develop longer lasting immunity than those only being vaccinated and without the dangers of no vaccine?

This study doesn't say anything about that, and I haven't seen any studies that look at that question specifically. I have seen studies that look at people who get vaccinated after having recovered from COVID, and they do tend to have the strongest antibodies. But nothing about antibodies from an infection after being vaccinated. I think that would be interesting to see.

> Natural immunity from a previous infection still reigns supreme

Except for the part where you had to get sick for the first time, risking death and long-covid.

I also remember talks that the previous infection needs to be a serious one - asymptomatic case will not give strong enough immunity. But I'm not sure if that got confirmed.

Well, a very large number of people got infected before they were even eligible to be vaccinated. So it is what it is.

I have also heard that the severity of infection could determine the strength of immunity, but the studies done since then haven't found that to be the case. It may become more clear on a longer timeframe, but as of now who knows.

> Vaccinated are just as likely to be infected

This is definitely false.

This seems wrong, if there's two models that could be consistent with immunization it's:

- Immunization reduces infection risk and severity of breakthrough

- Immunization reduces infection risk but, once infected, severity risk is the same as an unvaccinated person

Both mean you should get vaccinated but the latter means you should be more risk-on if you are vaccinated. The people downvoting me and arguing with me in this thread don't realize that the latter theory, if true, is actually one which leads to more conservative risk taking.

It's only impossible to discuss them if your argument is based on a strawman that's been repeatedly debunked by actual science. People are exhausted with all the hypotheticals that are predicated on the things that are not true being true.

Tribalism may be the cause, but the cure (perhaps) is to approach these things in good faith. There are so many bad faith actors out there whose primary purpose is to evangelize, intentionally sow division, or both.

You were probably downvoted because the “theory” is just wrong according to every piece of data we have.

How did you even come up with this theory? Was it based on some evidence or just something that felt right to you?

Pfizer data

https://drrollergator.substack.com/p/damned-lies-and-vaccine...

So you read one article about a early study of vaccine efficacy that clearly states it's limitations and then decided that was good enough and stopped following any of the data that has come out since then.

You keep asking for "data to update you priors", but it is easy to find such data if you were actually interested.

Reply on news.ycombinator.com