4 ms·

I don’t believe causation was mentioned. There was a study done that said people who took the drug were slightly better off. No idea how valid the study was.

• gumby 3y ago
  > There seems to be a lot of questionable statistics done. I doubt it in the sense of "finger on the scale". For some indications (common cold) the statistics are pretty clear (large n, unambiguous signals) even if the therapies are poor (thus there is basically no treatment for the common cold). Something like Alzheimers is at the other extreme: the disease is very poorly understood, it's very serious, and even measuring progress is difficult. So if you can find something that appears to show some net benefit to a nontrivial proportion of patients it's considered ethical to get it out there. Sure, the gold standard would be a multi-arm trial with controls, lasting 30-40 years and starting with subjects without any symptoms at all. That's probably the only way to get a reasonable n. But apart from the fat that nobody would, or probably even could, do such a trial, I doubt it would be ethically acceptable. This is just reality.