Hacker News Reader

Comment by dukeofdoom | Hacker News Reader

How can this possibly be true, when we now have real world data from Ontario. The case count of the vaccinated has now surpassed the unvaccinated and seems to be growing exponentially. 3rd graph.

32.39 cases / 100,000 people double vaccinated vs 31.93 / 100k of the unvaccinated.

https://covid-19.ontario.ca/data?fbclid=IwAR2pRUq9GN9EEoDTm0...

The vaccines don't seem to confer any lasting difference.

View on HN

That's interesting data. It's not in agreement with other jurisdictions, so far. Elsewhere, including South Africa, we see (poor quality) protection from vaccination.

One concern with observational studies: people who choose vaccination don't behave exactly the same as people who don't. They may assume more risks out of perceived safety. They may also be more likely to test or seek medical care. This is why, at a minimum, you need case control to start to believe the conclusions.

Unlikely, in Ontario the unvaccinated as a group have to test 2 or 3 times a week keep their jobs. If anyone is more likely to report a case it's the vaccinated.

OK. We'll see. I think there's lots of confounds-- e.g. vastly differing vaccination rates among children vs. the population at large.

In any case, we have actual case control studies showing mRNA vaccine efficacy against illness and particularly against severe illness from Omicron. They disagree on the amounts, and in some cases the numbers are not thrilling (30%), but there's no studies so far implying no efficacy.

The study we're discussing here concludes its abstract with: "These data support the notion that, provided high neutralization capacity is elicited by vaccination/boosting approaches, reasonable effectiveness against Omicron may be maintained."

Ok.We'll see.

I think you could give out 2 doses of a strong laxative and show it statistically protects sick people against covid too. If you take out the weakest, the rest will seem to do better.

https://twitter.com/MartinNeil9/status/1466814347762671628

You're citing a ridiculed preprint from a CS Ph.D?

That's just plainly not true, as a matter of government policy. There's a soft mandate in some settings like hospitals. Otherwise it's up to the employer, and there are tons of employers without policies.

Way too many confounders to draw that conclusion from the data. Vaccinated in Ontario are allowed to participate in a whole host of risky activities like going to gyms, dining at restaurants etc. that unvaccinated technically can't do. And the people left unvaccinated at this point are mostly ideologically opposed; do you think they're actually testing and reporting their cases? We're also just way behind on tests in general.

The unvaxxed have to submit test results to do anything though, even attend work. So you might argue their testing rate is higher.

At some workplaces they have to use a rapid antigen test. You need a positive PCR test to be in the official numbers.

There is something odd with the stats. If we look at individual age groups listed, the cases for unvaccinated is multiple x higher than fully vaccinated. Then if one looks back at "all ages", the cases per 100,000 people are almost the same. Is there a bug with the charts?

The data broken down by age is less recent for some reason. Only goes to Oct 24.

Edit: Disclaimer at the top of the page says technical difficulty.

I suggest reading up on Simpson's paradox, which keeps showing up in these kinds of data evaluations regarding covid.

https://en.wikipedia.org/wiki/Simpson%27s_paradox

In this case it's not relevant, because it couldn't explain these similar magnitude trend lines adding up to something so tremendously different.

The effect is caused by the age data ending in mid-October, before the spike we're looking at in mid-December. So, the age-segregated data couldn't show that spike at all.

“The vaccines don't seem to confer any lasting difference.“

Why would we expect vaccine derived coronavirus-19 immunity to be substantially longer than natural other-coronavirus immunity, which we know to be months, not years.

Vaccine-derived immunity often outperforms immunity from natural infection. For COVID-19, it looks like this is true, but not by nearly as much as we'd like.

How is that actually possible and not just completely variable across the board based on each individual's health? 'Often outperforms' leaves the question 'how often?'

> How is that actually possible and not just completely variable across the board based on each individual's health? 'Often outperforms' leaves the question 'how often?'

That is, we know of many vaccines where the protection from vaccination is greater than the protection from prior infection. It looks like that is true of COVID vaccines, both by observational / case control data and from data comparing antibody neutralizing titers between convalescent plasma and vaccinated individuals.. e.g. https://cdn.jamanetwork.com/ama/content_public/journal/jama/...

For any given patient--- who knows. There could be one person somewhere who'd get no benefit from vaccination and a huge benefit from prior infection. We can only know about the superiority of vaccination to prior infection in aggregate, but the effect is real.

The first problem with this hot take is, you're completely ignoring the severity of cases, and death, in favor of just a raw case count.

Reply on news.ycombinator.com