Post

Or how about “well-conducted study, or bust”?

The first especially is not a reliable source, and we have no idea whether it truly worked or it is dumb luck and "AI psychosis", but it seem fair to say that this sort of stuff would be hard to get by RCTs

touche

Ah sorry, yes I agree on that, I meant to say compared to other prevention drugs. GLP1s is one example but tbf the incentive to participate in that trial would've been higher since it's also a treatment.

I think observational studies make many more assumptions than RCTs do, and it's harder to tell what can potentially cause issues. Generally, for observational studies on the effects of medicines and vaccines, differential selection into treatment is an important one.

The different time points would need to be very different--years apart? I don't think that's ethical.

Counterpoint: the absence of RCTs has been weaponized especially in gender affirming care for minors and is the load-bearing rationale for the forced detransition of thousands of kids. RCT is best, I agree, but not a necessity to show that a given therapy has benefits > harms.

Yes but a trial would have been very be hard here? So going to England after wales seems like a good move, ont you can do while you get a trial going anyway? That welsh effect size would have shown up in the English data (and the fact it doesn’t is very telling)..

why don’t you simply go and read the ethical rules governing medical trials instead of writing pointless posts and then argue when it is pointed out to you it is pointless? The trials you ask for are unethical.

Sounds sane to me! RCT for confirmation, observational (plus, and I know this sounds crazy, theory) for exploration of potential RCTs to design. Caveat: I am not a real scientist.

I agree, I don’t think you can ethically randomize people to no treatment here. And so if you insist on an RCT that means you never know the answer. Ironically the antivaxxers love taking this position for bad faith reasons

You know there is a benefit and you are withholding that benefit. Equipoise?

If not universally recommended but researcher knows there is an established effect on shingles,is it really ethical to give a placebo without informing the participants they could benefit had they chosen to take the vaccine? I don't think so But you can't really randomise if you do provide this info

but I have worked on cardiovascular effects studies. I would imagine it would be similar in terms of the number of people and how long folks would need to be followed. This is a lot of money. And I think protective effects might not result in enough sales to justify it.

Give me a year or so, I might write a ditty about him/her.

I weight RCTs very heavily, too But wouldn’t you agree that most medical choices & decisions do not have RCT evidence (much as we’d like)? In many cases, we just have to do our best with bad evidence

This chart from Duggan and Evans is from an observational study that clearly showed the causal impact of haart on mortality years prior to any RCT being conducted.

"RCT or bust" is indeed very unpopular with this account. But in your case you probably have a point. 😆

I think I'd agree that for vaccines, RCTs are crucial. I'd also want some evidence that there is a biologically plausible metchanism - but you don't get to do RCTs of vaccines without that. I don't agree that RCTs are the best method for every research qn.

Ditto the effects of diet.

I’m an epidemiologist and I mostly agree with this - the majority of observational research you see doesn’t even have a question that’s asked clearly enough for the answer to be meaningful However - sometimes you can’t do the RCT, sometimes you can do good observational research, and -rarely- both

Especially when it comes to dementia / cognitive decline, I always think there are obvious confounders that would make it less likely to receive such a vaccine AND more likely to get dementia. Same btw for hearing aids