Comment by hprotagonist
3 years ago
relatedly, the only good reason to stop a trial early is that it becomes unethical not to treat the control group because the effect size in the treatment group is so huge. And it does happen, sometimes.
3 years ago
relatedly, the only good reason to stop a trial early is that it becomes unethical not to treat the control group because the effect size in the treatment group is so huge. And it does happen, sometimes.
I sometimes wonder if the exhilaration of such a result comes with a twinge of regret that the result could not have been foreseen before the science reached it, and more people given the lifesaving treatment immediately.
But that is the human condition, I guess. Scientific progress and learning brings regrets, often very momentous ones in retrospect.
While I totally get what you mean, I’d guess for scientists, the answer is generally, “no”. Expected outcomes for trials like this are a whole lot less certain to the people doing the work, that than it seems on the outside, so i think it wouldn’t even occur to the scientists that the downside of “withholding” treatment from the tiny (relative to the population) control group comes close to the upside.
Yeah, it's just a wistful thing, and most of us laymen don't even experience the part where you want to know something but don't know it yet.
Regret, sure, but since we can't change the past that regret should motivate us to work harder to make the present and future better. We're a young species, and part of growing up is looking back with chagrin at how foolish we seem in the light of our new growth and learning.
The number of trials that don't work in humans when it worked in every pre-clinical trial up to that point is enough that it makes sense to be extra cautious.
It’s true, but there’s no alternative. 99% of things don’t work.
generally speaking, “holy shit it worked?!!!” drowns out a lot. It’s not so much exhilaration as it is a kind of astonished joy.
The other reason is futility: the treatment has no benefit or any improvement is grossly outweighed by side effects.
I mean, treat everybody in the world with that condition at that point, right?
The treatment costs ~$100k over half a year.
If this is as effective and reliable as is suggested, this the expense is a cost saving effort. Cancer is an exceptionally expensive disease, at least in the US.
That's peanuts in the context of cancer treatment.