Hacker News Reader

Comment by dahfizz | Hacker News Reader

> Because the data shows they don't matter.

Source? Because the CDC says otherwise.

> In the first full week of October, vaccinated New Yorkers with a prior Covid-19 case were 19.8 times less likely to catch the virus than their unvaccinated and uninfected peers, whereas people who were unvaccinated but previously infected were 14.7 times less likely, and vaccinated but uninfected New Yorkers were just 4.5 times less likely.

In other words, natural immunity alone can be up to 5X more protective than the vaccine alone, according to the CDC.

https://www.forbes.com/sites/joewalsh/2022/01/19/cdc-prior-c...

View on HN

The person you are replying to is talking about hospitalizations for severe ilness. The study you are quoting is talking about covid infections in general. These two are not the same and the vaccine protects from the former (severe illness and death).

No, he's also talking about severe illness and death. He's saying that prior infection is a good substitute for the vaccine in preventing severe illness and death. The corollary being that we should consider "vaxed or prior infection" and "immunologically naive" as the two relevant groups (for mandates etc), not "vaxed" and "unvaxed".

Prior infection is a good substitute for the vaccine because the ones that die from that "prior infection" aren't going to be in the statistic that shows how effective a prior infection is compared to a vaccine for a subsequent covid infections

You're responding to the wrong comment. I'm clarifying what the other commenter said, not championing his point.

That being said, your comment still doesn't make any sense. The people who would be substituting prior infection for a vaxpass would not be the dead ones, because dead people are not known for going to comedy shows, bars, or restaurants.

Ie, the claim is that "conditioned on being infected [and alive], a vaccine is superfluous". The [and alive] that you clarified is about as helpful as "and non-fictional"

Here is the actual study: https://www.cdc.gov/mmwr/volumes/71/wr/mm7104e1.htm?s_cid=mm...

Importantly it doesn’t account for the effect of boosters and looks at pre-omicron data.

That noted the best protection seems to be had by a combination of vaccination and prior infection, which is perhaps unsurprising.

Also other CDC studies suggest that natural immunity declines quicker than vaccine induced immunity (https://www.cdc.gov/mmwr/volumes/70/wr/mm7044e1.htm?s_cid=mm...).

Given these results it would seem like generally a good idea to get vaccinated, even if you have had prior infection.

> Importantly it doesn’t account for the effect of boosters and looks at pre-omicron data.

Don't just read the summary. Look at figure 1:

https://www.cdc.gov/mmwr/volumes/71/wr/mm7104e1.htm#F1_down

There are unvaccinated, uninfected people...and then there is everyone else. Once you have been infected or vaccinated, the difference between is so small as to be academic. And as the authors note, the importance of natural infection increased over time, as the virus escaped the vaccines:

"Importantly, infection-derived protection was higher after the Delta variant became predominant, a time when vaccine-induced immunity for many persons declined because of immune evasion and immunologic waning."

To your other claim:

> other CDC studies suggest that natural immunity declines quicker than vaccine induced immunity

No. That's not what that study said. First, it followed symptomatic infection, not severe illness. As we're seeing now, Omicron pretty handily bypasses vaccine-induced immunity to cause minor illness.

Also, methodologically, the study you're citing was bad. It only gave credit for natural immunity to those people with confirmed infections, which was/is an underestimate, and made vaccines look more effective than they are. A substantial fraction of the "only vaccinated" group almost certainly had undiagnosed covid infection.

> There are unvaccinated, uninfected people...and then there is everyone else.

This is shifting the goal. The disclaimers I cited come directly from the study and are highly relevant when comparing the effect of natural immunity vs vaccination.

> No. That's not what that study said. First, it followed symptomatic infection, not severe illness.

This doesn’t contradict what I said. The study gives evidence that natural immunity (to symptomatic infection) declines more quickly than vaccine induced immunity.

> Also, methodologically, the study you're citing was bad. It only gave credit for natural immunity to those people with confirmed infections, which was/is an underestimate, and made vaccines look more effective than they are.

One, the same effect would also have been present in the natural immunity group, as people may have been reinfected since their initial confirmed infection.

Two, this is largely irrelevant to the point I was making.

The statistical evidence suggests that even if you have had a previous confirmed infection, you’re still better off getting vaccinated.

>The statistical evidence suggests that even if you have had a previous confirmed infection, you’re still better off getting vaccinated.

Sure. And you'll continue to be better protected every time you get a booster, every 6 months, for the rest of your life. But all of us are going to stop getting boosters at some point. We're going to decide we're "immune enough", and go on with our lives, even though there will be some minute chance of getting Covid, from whichever variant is then around.

Many people with prior infection may feel they are "immune enough", and you can argue over the significance of the remaining risk, but ultimately that's a decision they make for themselves, just as all of us do in every facet of our lives every day.

Regardless of the minutia of the numbers, the recommendation to get vaccinated regardless of whether you have had COVID in the past or not is based on the fact that a very high number of people in the US and worldwide "think" they probably had covid but aren't entirely sure because they never got tested. The vaccine for everyone is easier and cheaper from a public health standpoint than to have people taking serology tests that may or may not be accurate and then waiting for the results in order to find out if you should get the vaccine or not. When the fact is, even if you were previously infected, you will be even better off getting vaccinated.

Apparently a lot of people without prior infection or the vaccine feel they are immune enough. And they are the ones clogging the ICU beds and otherwise taxing our health care system. I think "immune enough" in some cases, like a pandemic or potential pandemic, has to be decided from a public health standpoint and not on an individual basis.

> This is shifting the goal. The disclaimers I cited come directly from the study and are highly relevant when comparing the effect of natural immunity vs vaccination.

No, it isn't. It's clarifying that there's a great big forest here, lest we get lost in the details of the spots on the bark beetles in one particular tree. The immunity granted by prior infection is at least on par with that of vaccination.

Moreover, I quoted directly from the study to explain why short-term squiggles in the lines at the bottom of the graph aren't worth dwelling upon: the vaccines are getting worse over time (against symptomatic infection), and the effect of natural immunity is gaining importance. This is exactly as you would expect for a virus that has evolved to partially escape the vaccine.

Regardless, these are minor differences. You can't reasonably argue that vaccination has a definitive advantage over natural infection. We should be counting both.

From a recent editorial from a Johns Hopkins professor addressing the CDC trying to whitewash their own failings in the study:

https://www.wsj.com/articles/the-high-cost-of-disparaging-na...

"Yet the CDC spun the report to fit its narrative, bannering the conclusion “vaccination remains the safest strategy.” It based this conclusion on the finding that hybrid immunity—the combination of prior infection and vaccination—was associated with a slightly lower risk of testing positive for Covid. But those with hybrid immunity had a similar low rate of hospitalization (3 per 10,000) to those with natural immunity alone. In other words, vaccinating people who had already had Covid didn’t significantly reduce the risk of hospitalization."

"Similarly, the National Institutes of Health repeatedly has dismissed natural immunity by arguing that its duration is unknown—then failing to conduct studies to answer the question. Because of the NIH’s inaction, my Johns Hopkins colleagues and I conducted the study. We found that among 295 unvaccinated people who previously had Covid, antibodies were present in 99% of them up to nearly two years after infection. We also found that natural immunity developed from prior variants reduced the risk of infection with the Omicron variant."

Classic NIH: Say there isn't enough data when challenged on a policy, then deliberately not study it. They did this for DECADES with cannabis.

I don't think your math is correct there and that data is pre-Omicron.

https://www.imperial.ac.uk/news/232698/omicron-largely-evade... estimates that prior infection offered 85% protection against Delta but only 19% against Omicron.

The vaccines also offer very little protection against infection by Omicron, and less with the B.2 variant.

Definitely not saying vaccines aren't important - they keep you out of hospital! But they don't prevent infection post Delta.

I don’t know why you think this. Covid cases — people merely testing positive — are overwhelmingly among the unvaccinated in every state. Vaccine cuts your risk by 3x to 5x depending where you look.

Look at this chart: https://covid.cdc.gov/covid-data-tracker/#rates-by-vaccine-s...

If you read to the bottom of the previous study I posted they addressed vaccine effectiveness too based on UK data. They work! Just make sure to get your booster.

> It cuts your risk by 3x to 5x depending where you look.

In other words, it's 70-80% effective, which is pretty poor performance for a vaccine. So, can we please get an update that encodes the Omicron spike instead of trying to boost antibodies for a virus that's now extinct?

They’re working on it. But the original target effectiveness during vaccine development was 75%. We got lucky it was so much better against earlier strains. It’s still pretty good — about the same as an average year’s flu shot.

It’s a far cry from the apparently very popular myth that the vaccines no longer work at all.

The flu shots are some of the least effective vaccines out there.

The original target effectiveness was set low because anything to nudge the R value down was better than nothing.

We now know that we can do much better.

Flu shots are about 40% effective.

What would be interesting, would be to see flu shot efficiency and COVID shot efficiency over time.

Preferably including Omicron data as well.

Pfizer is planning to release an "Omicron + previous variants" version of their vaccine in March. I guess even with the accelerated approval scheme it takes some time. Then we will wait for the manufacturing.

There is also the "what will happen to the previous batches" question as no one will be willing to get the old vaccine.

Quick question, if I forced you to stand in a groin-kicking machine that will:

A) Once in every.. I dunno, 200 times (0.5%) it kicks you so hard that, without protection you're 100% going to die

B) Once every 10 times (10%) it kicks you so hard you're sent to the hospital

C) Just to spitball, let's say 3 out of 10 times it will kick you hard enough that you're out of commission for a couple days but able to recover without hospitalization

D) 3 out of 10 times it will lightly kick you where you hurt a bit but it's just a minor inconvenience

E) The rest of the time it will let you leave without being kicked.

Ok, so you're strapped into this machine. Now, before I turn it on, I offer you a cup (protection). 70%-80% of the time this cup will offer you complete protection against whatever the machine throws at you. But remember that it's still 100% effective when the machine does nothing, so we only need to look at the 70.5% of the time the machine will actually kick you.

So with 70%-80% protection call it 75% effective, we drop our percentages for each scenario:

A) 0.5% -> 0.125% B) 10% -> 2.5% C) 30% -> 7.5% D) 30% -> 7.5%

This is just doing the simple math. The vaccines drastically lower the chance of severe COVID compared to no vaccine at a higher rate than their overall effectiveness.

So again, you going in there wearing the cup or no? Because that's the choice you have when it comes to getting vaccinated.

I think what they are suggesting is that natural immunity + vaccine is also more protective than just natural immunity

But natural immunity doesn't work if you die or end up having severe loss of health, so vaccine + natural immunity would be best way.

CDC said last October that 146 million people (~40%) in the US have already recovered from Covid.

There are also interventions other than vaccines, e.g. early treatment, https://c19early.com

yes, vaccine + natural immunity gives you super immunity by a large factor, they also found that it doesn't matter if you get the virus, then get vaccinated, or get vaccinated then get the virus, you'll end up with super immunity.

Super immunity is not a scientific concept, it would be nice to have some numbers or studies to show the difference.

Natural immunity as a personal choice should have been possible. But that has not been an option, for someone who wants to be part of society.

There is a behavioral component.

Is that because people who have caught it are more likely to take precautions against catching it again?

Those who were vaccinated but previously uninfected may have had a “it can’t happen to me and if it does, I’m vaccinated” attitude.

I’d expect unvaccinated people to be in the bucket with covid deniers and anti-maskers, which would have the highest base rate of COVID.

I doubt it. I caught it in Aug/September during the delta wave and am way less careful now than I was before. Not saying this is true or advocating a position here, but I feel like the pandemic is over for me, I don't have to worry about any current variants, etc. The people I talk to who have also developed natural immunity have the same sort of feeling.

Meanwhile… my sister and her daughter had a Covid sandwich: caught it in summer 2020, got vaccinated as soon as feasible, then caught it again a few days ago.

Anecdotally, my sister is one of those people who “catches everything” so I can’t say I’m surprised, but clearly there is no such thing as absolute sterilizing immunity (natural or vaccine-induced) for this particular coronavirus.

It's not specific to this one, it's true of all coronaviruses.

This article explains why it's so hard to make a vaccine for coronaviruses:

https://www.abc.net.au/news/health/2020-04-17/coronavirus-va...

Of course, this was written a few weeks before "Project Warp Speed" when everyone and their dog announced that had found a vaccine for this brand new coronavirus.

You are confusing the effect on individual people with the effect on the whole community. I do not debate the effect of natural immunity on individuals who have had COVID in the past, but it is not widespread enough to cause a dent in the infection or to alleviate the stress on the healthcare system.

The effect exists but the data shows it does not matter.

> The effect exists but the data shows it does not matter.

You may have been right at a time where almost nobody had a previous infection. However, estimates by reputable sources https://covidestim.org/us show that a large percentage of people have been infected by covid at least once. The data are by state, I see ~75% for MA, 80% for CA and NY, and you can pick your favorite state.

At this point it is more correct to say that the vaccine does not matter, and immunity by prior infection is the dominant effect. Either that, or the Yale school of public health, the Harvard school of public health, and the Stanford Medicine institutes who run the covidestim.org site are full of shit.

If the vaccine does not matter, why do we still see such a prevalence of unvaccinated people among patients Who required hospitalization?

Because many of these are older individuals with contraindication. The elderly with unknown vaccine status. The homeless. People in nursing homes trying to live out the last years of their life without the discomfort of vaccine side effects.

Not age adjusting the vaccinated and unvaccinated cohort is disingenuous. Not breaking out ‘unknown’ vaccine status — which only becomes relevant for the stats after death - is disingenuous.

You are reading propaganda, and reading it uncritically.

The elderly (more susceptible) are overwhelmingly more vaccinated than the younger, not less.

Medical records are not well distributed. In most of these statistics, unknown vaccine status is considered to be the same bucket as unvaccinated. My hypothesis is that most of this group is either elderly or homeless.

Similarly, there is a selection problem for the elderly who choose to remain unvaccinated. You're saying that they think vaccines are ineffective -- and this is definitely a cohort of people! But there is a separate cohort of people, who are more likely to be hospitalized, who know they have severe existing health problems.

Some anecdotes:

I got COVID before there were vaccines and I've received all three shots. While all of my friends caught Omicron, I have not - and I'm the only one among them who got sick beforehand. On the other hand, I have an unvaccinated coworker who has now gotten Delta and Omicron. I know that doesn't prove anything, but it also doesn't make these studies I see any more illuminating either.

As an aside, I'm not even sure I want things to go back to "normal" other than lifted travel restrictions and the anger/frustration/callout culture to go away. Restaurants taking over the streets, working from home, less cars, etc have all been a boon for the way I live my life.

Reply on news.ycombinator.com