Hacker News Reader

Comment by tacomonstrous | Hacker News Reader

>the risk of developing symptomatic COVID-19 was 27 times higher among the vaccinated, and the risk of hospitalization eight times higher

This sounds convincing, but then in the next graf:

>For instance, the higher hospitalization rate in the 32,000-person analysis was based on just eight hospitalizations in a vaccinated group and one in a previously infected group. And the 13-fold increased risk of infection in the same analysis was based on just 238 infections in the vaccinated population, less than 1.5% of the more than 16,000 people, versus 19 reinfections among a similar number of people who once had SARS-CoV-2.

View on HN

>the risk of developing symptomatic COVID-19 was 27 times higher among the vaccinated, and the risk of hospitalization eight times higher

The denominator in those odds ratios seem to fishy to me. Comparing people who were previously infected with COVID and vaccine recipients does some weird conditioning on getting COVID, so it should hide risks of developing symptomatic COVID and hospitalization for people who got COVID previously. It seems to me that the risks should be "cumulative" for people who were previously infected (i.e., the denominator should include symptomatic infections and hospitalizations for previous COVID variants and not just delta) rather than "zeroing" those risks for this group and just focus on delta re-infection. The decision to only focus on delta infections will bias up the OR estimates for vaccine (in)effectiveness, so we should read the results of this paper as the possible upper bound of the ORs.

Also, the study compares the risk of delta infection, meaning that people cannot really just say "I won't get vaccinated because natural immunity is better" as the variants that provided that natural immunity are not "dominant" anymore, meaning that the opportunity to acquire natural immunity against delta is probably close to zero at this time.

Also also, isn't it something that is known that natural immunity is better than vaccine-acquired immunity? The trade-off is that natural immunity has the "side effect" of actually having to fight off and survive the illness?

> Also also, isn't it something that is known that natural immunity is better than vaccine-acquired immunity?

Yes, I pointed this out elsewhere in the thread, but it has somehow been verboten until very recently to state it in regards to Covid. My brother-in-law was even told by the people administering his vaccine that having had a prior Covid case put him at more risk and made it even more essential that he get vaccinated.

It does not seem unreasonable to me to extrapolate that people who caught COVID-19 once are more likely, as a group, to be exposed to it again in the future. That would be a matter of their environment rather than their individual resistance to the virus. I haven't seen any studies on that issue, though.

In any case the vaccine is safe, effective, and way better than actually catching COVID-19. And in the event that you did already recover from COVID-19, getting the vaccine doesn't hurt and will boost your immunity even further. So—unless you have a specific medical condition which dramatically increases your changes of an adverse reaction–you should just get the vaccine.

You raise a good point that it would be useful to report the "cumulative" risk for people previously infected.

However the goal of this research is specifically evaluating risk of reinfection, not the cumulative risk of SARS-COV-2 infection. The fact is that a vast majority of people survive initial SARS-COV-2 infection, so the pure risks associated with reinfection are also equally useful to know.

The decision to focus on delta infections does impact the OR estimates for vaccine effectiveness as you mention. But again this was one of the primary research goals of the paper - to evaluate the risk of reinfection by comparing natural immunity to vaccine-induced immunity in the context of a variant with mutations conferring immune escape.

> the variants that provided that natural immunity are not "dominant" anymore

This was controlled for in the study by censoring data based on the date of previous infection.

> The trade-off is that natural immunity has the "side effect" of actually having to fight off and survive the illness?

Yes, and people need to stop interpreting the title and cited paper as a manifesto against vaccination. Rather, the study provides valuable data on the efficacy of natural and vaccine induced immunity in the context of evolving variants, which will be useful in guiding public health policy and also formulating next generation vaccines that improve upon the current formulations.

To develop natural immunity, you would need to get infected with the virus, no? That's considerably more risky than getting the vaccine.

If you do get infected with the virus after vaccination, wouldn't you also develop natural immunity?

I think the worry with natural infection is if the timing is such to provide durable immunity. But it’s not like the vaccines have had 5 years of trials to determine the best wait between the 2 and 3 doses so kind of a moot point.

Why 5 years? Why not 1? Why not 50? What's so special to people about 5 years? I keep hearing people say "it hasn't been tested long enough" followed by "I'd take it in about 5 to 10 years" which is a HUGE difference. Is it just that "time has passed", because a year has passed since the Pfizer-BioNTech vaccine was first given to test subjects. But I guess it's up to people - their body, their choice.

Even though we live in what is commonly referred to as a "society". We agree to live by certain rules and norms that allow us to coexist. There is a faction of fools out there who resist being a part of society, but claim all of its privileges. These are sub-human individuals who just seek some form of control over others. That has to stop.

>Why 5 years? Why not 1? Why not 50? What's so special to people about 5 years?

Clinical trials are conducted over many years (rather than just 1) for historical safety reasons[1]

>There is a faction of fools out there who resist being a part of society, but claim all of its privileges. These are sub-human individuals

This is hate speech, especially considering the disproportionate rates of black and hispanic people who are unvaccinated.

[1]https://www.brightfocus.org/clinical-trials/how-clinical-tri...

> This is hate speech, especially considering the disproportionate rates of black and hispanic people who are unvaccinated.

It is not. Black & hispanic people are vaccinated at lower rates than white people mostly due to access issues, not refusal to get the vaccine. One of those access issues is, in fact, that they are disproportionately members of a part of society we've demanded to take a lot of risk: essential workers.

https://www.npr.org/sections/health-shots/2021/04/26/9899620...

That's right, vaccine doesn't prevent you from getting the virus so if you get it you'd then amount an immune response of some sort which is on top of the one provided by the vaccine.

Or that goes to "fill up" again the immunity lost due to the time which went by since the vaccination.

If you got Covid before vaccines were available that risk assessment is irrelevant.

If you get COVID after vaccination, your immune system may not learn how to fight COVID-19 as well, because the vaccine is present and doing something.

The vaccine isn't doing anything in your body to fight COVID-19 infection actively. Your immune system is doing the fighting regardless.

That is not what I said.

“ your immune system may not learn how to fight COVID-19 as well, because the vaccine is present and doing something.”

But the vaccine being present isn't "doing something". The thing it did has already happened.

If you develop covid-19 symptomatic disease after vaccination, you will likely still get natural immunity.

The immune system "doing something" is why you feel sick.

The vaccine fighting the virus doesn't make you feel sick because it's just antibodies clumping on virus particles, preventing them from infecting cells.

There are also inactivated-virus covid-19 vaccines [0] in trials that one could expect to impart more broad-based immunity than the mrna vaccines which are designed to target only one part of the virus.

[0] https://en.wikipedia.org/wiki/Sinovac#COVID-19_vaccine_devel...

But that seems intuitive? If you survive the first Covid infection, you’re biased towards less severe symptoms for the 2nd infection (ignoring even any immune response).

Basically all the folks who died from Covid are excluded from this population.

The bigger problem is that the risk of hospitalization post-vaccine appears to be lower than the risk of death pre-vaccine. If seven of those eight hospitalized post-vaccine would have died when naively infected, then the difference disappears. Seven out of 32000 is 0.02%, while the risk of death from COVID-19 is 0.4%.

And if you remove hospitalization as an endpoint, you're basically dependent on self-reports. That's not to say I don't believe the conclusions -- they should survive this critique -- but there is reason to expect a little inflation of the ORs.

Are you saying 32,000 people isn't a large enough study to be meaningful?

For low-prevalence events the number of events that happened is just as important for statistical power as the big 32,000.

Especially when doing relative risk, the confidence interval on that "27 times" could be extremely large.

True, but we aren't comparing a 1 to 0 or 1 to 2. Its 238 to 19. sqrt(238)=+/-15 sqrt(19)=+/-4 (if the numbers are to be believed) . 238/19=12.5 (238-15)/(19+4)=9.7 so it becomes 20% less on the lower end of the fluctuation which is only nominal.

It's more about the small effect sizes of 8 people vs 1. Still a lot of room for noise in the result.

True, but where's a better study at the moment? I think it would be easier to gather more studies and do a cross-study meta analysis rather than recruiting a million people into a single study to eliminate the effects of noise.

The "27 times" is factually their finding in one study, which someone might read as being given too much weight in the article but doesn't change the data.

Have you ever looked at the numbers on the vaccine trials? Especially the ones in children under 18?

Reply on news.ycombinator.com