Hacker News Reader

Comment by moate | Hacker News Reader

As someone who works in medical publishing, you have no idea how infuriating posts like the one you were responding to make me.

This is HN, a group I would expect to have above average understanding of statistics and statistical analysis and yet the idea of causal relationships and that old adage of "correlation is not causation" STILL seems to slip past far too often.

To the larger world: There's a reason every single drug commercial/ad you've ever engaged with has a list of "possible side effect" and not "known things that will/did happen". The "possible" doesn't even always mean "this is a side effect we conclusively know this drug causes, and could possibly happen to you" but sometimes just "we couldn't prove this drug didn't cause someone to shit blood since we had one guy who did one time during our months of trials".

View on HN

To be fair I never said causation nor meant to imply any. Perhaps correlation would have been a better word than linked.

My original point meaning being FDA probably does not want these records released because it could shine a unflattering light on how our pharmaceutical industry operates.

You more than implied.

You straight up said deaths linked to the vaccine

Just like states continually acknowledge widespread statistical errors after getting called out where they attribute deaths and hospitalizations to COVID and then someone points out that person was shot in the head or had a car accident or the kid went to ER with a broken bone?

This is called "whataboutery". It's also disingenuous.

In general, there's a fast statistic about COVID deaths, which is something like "deaths within N days of a positive PCR". This will catch some people who died for another reason, though if you think that's a serious problem, you'd need to argue that so many people could be expected to die for some other reason within N days that this would significantly bias the stat.

There's a slow stat, where COVID is a contributing or underlying cause on a death certificate. In the UK, these are in rough agreement. Notably, at the start of the pandemic, when testing wasn't available, the fast stat was an underestimate. https://www.nuffieldtrust.org.uk/news-item/measuring-mortali... has some links to the ONS and places like that.

As someone who understands basic logic and ethics, you have no idea how infuriating your defending the indefensible makes me!

We're talking about a completely new type of "vaccine" that works in completely new and different ways than ANY previously, AND which has never been used in humans before (after all previous animal trials failed, AND is intended to be given under high urgency (even duress), to a large fraction of the population of the planet.

DAMN STRAIGHT that under those unprecedented high-risk conditions the default assumption MUST BE that correlation IS causation - UNTIL PROVEN OTHERWISE.

Anything else is gross and callous negligence and lack of consideration for your fellow man, not to mention the active shredding of medical ethics, including the Hippocratic oath and the Nuremberg code.

This is all the more true now that, a year on, data is clearly showing that the danger of the virus itself was grossly exaggerated, and that we are now facing vaccine injuries and deaths that are at least of the same order of magnitude as the virus itself, and may yet prove to be considerably higher.

>>DAMN STRAIGHT that under those unprecedented high-risk conditions the default assumption MUST BE that correlation IS causation - UNTIL PROVEN OTHERWISE.

I'm glad you're not in charge of Science, since this is literally the opposite of science. No matter how many caps you use. The default assumption is there is no default assumption, even if there's a gun to your head.

I'm not defending "the indefensible" I'm explaining common terms that are seemingly misunderstood by large groups of people.

Reply on news.ycombinator.com