Hacker News new | ask | show | jobs
by beebmam 1630 days ago
When you look at individuals, you'll never be able to truly say any treatment "worked" with certainty for virtually anything. It's same with nearly all medical treatments. Humans, very often, get better on their own with time, and time can't ever be rewinded to test the alternative.

The way we know a medical treatment works is by looking at large scale data, controlling for variables, understanding mechanisms of action, and by approaching this science in good faith and not cherry picking data to fit one's own conclusions. There's a concept known as "Number Needed to Treat" that is probably relevant to your concerns here.

Vaccinations have historically been extremely effective for many diseases, and COVID is no exception. It's accurate to say (most) "COVID vaccines are highly effective at preventing severe disease caused by the SARS-CoV-2 viruses". Omicron is so new that it's hard to be certain of anything regarding it right now, but we'll have a better picture in a few weeks. I do know that we're seeing many hospitals get overwhelmed across the US currently, and many medical workers are out sick with a COVID infection.

1 comments

I have some skeptical family members who like to talk about 3rd party anecdotes. "A nurse told me she saw blah blah blah happen". As you say, stories like this are not a way to know a thing, but I can't get them to understand.

It's like we rolled a die once, and it came up 6, and now we're all forming our own special opinions on whether or not the die is fair. We just can't know. Of course, if someone rolled the dice 30,000 times and reported what they found, that would be great (wink wink).

Instead of using Calc 1 as a filter course, we should use Statistics instead.