Doesn't giving the drug to everyone in the trial at least add some statistical information (i.e. what is the likelyhood on it working for everyone vs. it being

Doesn't giving the drug to everyone in the trial at least add some statistical information (i.e. what is the likelyhood on it working for everyone vs. it being a spurious event)?