Hacker News Reader

Comment by cmrdporcupine | Hacker News Reader

" We can easily see that in many graphs, but my favorite is below from New York City, showing a clear distinction in hospitalizations among vaccinated compared to unvaccinated people"

I don't see this as much here in Ontario. Among the vaccinated as of today there is a 10.10/100k hospitalization status vs 16/100k for the unvaccinated (and 11.9 for one-dose). The hospitalization rate among the vaccinated is climbing quickly. Yes, there's still a gap, but yikes.

ICU is better, though, with a 0.76/100k rate for the vaccinated vs 4.74/100k for the unvaccinated.

This may have to do with different hospitalization and testing criterion, I don't know. But hospitalization here is in full exponential growth and not just among the unvaccinated.

Obviously being vaccinated is still far preferable. I have had 3 doses (1 AstraZeneca and 2 Pfizer) FWIW.

View on HN

Simpson's Paradox applies here: once you slice by age, the data tells a different story. Even when vaccines are highly effective, the confounding factor that gets lost when you aggregate across ages is that the most vulnerable (the elderly) are also the most vaccinated.

I highly recommend anybody concerned about hospitalization among the vaccinated read this post. It's about Delta hospitalizations in Israel 2021, but we should expect the same effect from Omicron and across countries: https://www.covid-datascience.com/post/israeli-data-how-can-...

You have to normalize by age.

Unvaccinated people in Canada are overwhelmingly very very young, in Quebec most of them are too young to be vaccinated.

That's why they don't get hospitalized as much :)

Lies, damned lies, and statistics!

I want to lookup data on the average age of unvaxed patients currently in the ICU but so far have not found any.

That's not the metric you should look for (at least you don't need to). You should look at the average age of unvaccinated people. That way you can estimate the prior probability of infection.

You can go here: https://health-infobase.canada.ca/covid-19/vaccination-cover...

From that, you can figure out that there are more people unvaccinated under 12 years old, than in the entire rest of the population, by a large margins - out of 7.6 million people without full vaccination, 2.9 million are between 5 and 11 years old, and 1.9 million people are under 5 years old.

So out of the 7.6 million people who aren't fully vaccinated, 4.8 million are under 12 years old!

And from 2.8 million remaining, 430k of those are under 18, and 1.0 million are under 30.

From this, we can figure out that the pre-vaccine hospitalization risk for the unvaccinated population is much, much lower than for the normal population.

This is without taking into account health status, of course!

Quebec publishes age-ajusted hospitalization rates per vaccine status, the latest number is a 7.7x odds ratio, which would be even worse for the ICU : https://mobile.twitter.com/sante_qc/status/14791225681667481...

Its good to see that its mostly young people who are at lower risk that are of the unvaxed, however the majority of ICU cases in Canada are still of the unvaxed, so that's why I am curious what is the average age of the unvaxed in the ICU. If 92-99% of people over 60 are vaxed, but its the 60+ demographic of unvaxed taking up ICU spaces, what more can be demanded by the government? Forcing younger people who are unvaxed might not reduce ICU cases very much.

This data from NYC has been supplanted and updated for all of NYS: https://coronavirus.health.ny.gov/covid-19-breakthrough-data

This still shows a greater than 10x difference in outcomes. Additionally, the "unvacccinated" population includes people naturally vaccinated by virtue of having a previous case.

It could very well evolve into a form that is mild enough and different enough from the vaccine spike that brings the unvaccinated and vaccinated hospitalization rates to convergence. Respiratory illnesses in general harm the elderly and immunocompromised disproportionately. But if that becomes the end game where COVID is still circulating, vaccines are no longer as effective at preventing serious illness, because the general severity has gone down, that is also a good outcome as it is what we have always lived with.

Colds and flus have always killed 10s of thousands a year. COVID killed many more because it was particularly severe. Just as the flu of 1918 is still with us today in a less harmful form (H1N1) This virus has a global reservoir and will never run its course entirely before mutating again most likely. Just like colds and flus before it.

> "COVID killed many more because it was particularly severe."

no. covid is novel, not appreciably more severe, and that's where mortality rates are coming from. it's following the dynamics of the many cold and flu viruses that came before it. we just happen to be at the beginning of the dynamics rather than at relative steady-state (e.g., 'endemic'), which is where most other viruses are at.

this inability to reason cogently about steady-state vs. dynamic aspects of systems is absolutely rampant across media, politics, and casual conversation, and must be a named fallacy at this point, though i don't know that name offhand.

It is? Care to source some data on these 'dynamics'?

You are engaging on wishful thinking I'm afraid - wishful thinking that is even now still killing hundreds of thousands. Perhaps there is a name for that too.

The best available evidence indicates that another coronavirus HCoV-OC43 probably caused a worldwide pandemic and killed a lot of people starting in 1889. The same virus is still endemic today and the only reason it doesn't take a huge death toll is that most of us get infected when we're young and the resulting natural immunity protects us as we age. But it and other endemic coronaviruses can still be quite deadly to frail and immunocompromised patients.

https://www.nature.com/articles/s41598-021-83987-3

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7252012/

https://sfamjournals.onlinelibrary.wiley.com/doi/full/10.111...

"Additionally, the "unvacccinated" population includes people naturally vaccinated by virtue of having a previous case."

Only if it's recent. How recent isn't clear yet. Somewhere in the > 90 days, < 1 year range, it seems.[1] As of last summer, about 1% of COVID cases were known re-infections.[1] Most repeat cases seem to involve two different variants.[2] Whether people who had previous variants are showing up infected with the omicron variant doesn't appear to be published yet.

[1] https://publichealth.jhu.edu/2021/why-covid-19-vaccines-offe...

[2] https://www.ncbi.nlm.nih.gov/labs/pmc/articles/PMC8604456/

First version and Delta re-infection is fairly common and Omicron re-infection will be more common still, even if vaccinated.

This virus has a global reservoir and will never run its course entirely before mutating again most likely. Just like colds and flus before it.

Maybe not. A much broader vaccine is coming along well.[1] That's from the US Army's Walter Reed Medical Center, and it just passed phase I (safety) clinical testing. That one is supposed to protect against all COVID & SARS variants.

There are 93 more COVID vaccines in development at the moment. Some are pills. Some are nasal sprays. Something is going to work. What we have now is the minimum viable product, and even that is pretty good. Early on, there was concern that a 50% effective vaccine was all that would result. Which is about where the original Sputnik vaccine, and the original SinoVac, are. Instead, we got 95% effective vaccines in round 1.

[1] https://www.defenseone.com/technology/2021/12/us-army-create...

Is the defense department going to vaccinate the reservoir of deer and cats and hippos, and do that worldwide?

This will be with us, and mutating, approximately forever.

The Ontario science table dashboard has the hospitalization ratio for unvaccinated to vaccinated still at 5:1. Where are you getting that these rates have become nearly equal?

edit: sorry had that written backwards

Link, for those who are looking for the (science table) source: https://covid19-sciencetable.ca/ontario-dashboard/

> Ontario science table dashboard has the hospitalization ratio for vaccinated to unvaccinated still at 5:1. Where are you getting that these rates have become nearly equal?

If the rates are the same, and 5/6 people are vaccinated, we would expect 5/6 people admitted to the hospital to be vaccinated. In fact, 5/6 people are vaccinated in the 12-40 age cohort, with older (and more vulnerable) populations having at least 9/10 people vaccinated and quickly going to 19+/20. Therefore, vaccines prevent hospitalizations, as the average of all adults is greater than 5/6.

You're misunderstanding. The dashboard is per Capita (so you don't need to account for the stuff you're describing), and shows 5x lower rate of vaccinated people hospitalized (and 10x lower ventilated). A far cry from the ~1.5x rate mentioned by ggp

Edit: It was a series of errors. Please check out https://news.ycombinator.com/item?id=29830131

Corrections: My above post was written in response to the unedited version of the GP post, which stated the 5:1 ratio was vaccinated:unvaccinated (and implied it was based on total numbers). Apparently, since I posted it was edited (and corrected) to say it was 5:1 ratio of unvaccinated to vaccinated (and based on per capita.) This makes far more sense and also makes my reply look insane.

I take GP at their word that there were just errors in writing their post. But it does leave a weird artifact out there that I cannot fix.

Yes, reading your post made me realize I typed it wrong -- sorry again!

ontario still conflates anyone hospitalized for anything testing positive for covid, with anyone hospitalized from covid; this alone could explain the discrepancy between a hospital bed and icu. lies damned lies and statistics...

It’s not that simple: yes, if someone gets in a car crash it’s probably not COVID but there are many cases where someone had a manageable condition that was disrupted by COVID — not 100% fault but not 0%, and a big problem for hospital capacity.

Counterpoint; My mum nearly died of Covid, six weeks in hospital, four of which were in ICU. On leaving we discovered she no longer had epilepsy.

If this is true, I hope she is in contact with an epilepsy researcher or three.

She's not, no.

This is the sort of extraordinary claim that requires extraordinary evidence to support it, and there are many potential confounding factors.

What sort of evidence would satisfy you?

A formal case report by her physician would be satisfactory.

I'm probably being stupid, but are the down votes due to scepticism about the veracity of my comment?

My guess is because it was not a counter-point to what the previous poster wrote. Although what they wrote does not exactly address the comment being replied to either.

Uh...What?

Comparing the rates only makes sense if one believes the risk is distributed equally among the two populations. However, the most vulnerable are the most likely to be vaccinated, and the least vulnerable are the most likely to be unvaccinated, so the vaccine could still be very effective even if the rates of hospitalization are equal, or even worse for the vaccinated population.

Alberta does a nice job breaking vaxxed VS Unvaxxed down by age group:

https://imgur.com/a/Actj5Tf

https://www.alberta.ca/stats/covid-19-alberta-statistics.htm...

My guess is that this speaks more to the cheapness of hospitals in Canada and the greater likelihood to go to a hospital with a lesser case of COVID. For instance, you quote a 7.6% rate of the vaccinated moving from the hospital to the ICU and a 29.6% rate of the unvaccinated moving from the hospital to the ICU. Meanwhile, the US blended average seems to be around 30% (eyeballing the data.) That could be because the unvaccinated are an overwhelming percentage of the admitted (it's the same rate), or it could be because there is a higher bar to go to the hospital at all for either class.

Mind sharing the data you're looking at from Ontario?

Yes, surprisingly few outlets show this, but it would have been crucial in showing vaccines work.

Reply on news.ycombinator.com