News for hackers. Stuff that matters.
Re:Alzheimer’s Drug Sharply Slowed Cognitive Decline by acveilleux on Friday March 20, 2015 @11:05PM (#9241732)
Lack of statistical power. Phase I enrol too few subjects for efficacy. The reason they do a Phase I is to determine it would be safe to do a larger trial.

Source: I used to work for the company that is most likely to have done any MRI analysis for this study. We did Phase I/II/III and they are very different beasts.

Re:Alzheimer’s Drug Sharply Slowed Cognitive Decline by kirsebaer on Saturday March 21, 2015 @07:45AM (#9242477)
A small study will lack statistical power to detect small effect sizes, but a small study could still detect a large effect.
Re:Alzheimer’s Drug Sharply Slowed Cognitive Decline by pcrh on Saturday March 21, 2015 @08:45AM (#9242600)
The point of statistical tests is to assess the accuracy of a prediction. These trials were not designed to predict/test the efficacy of the drug, therefore no conclusion regarding efficacy can be drawn.

It is of course possible that there was a real effect, and that this is reflected in the data gathered, however to determine if that is the case another trial has to be performed.