I think you are on to something. When science became so big that it first needed then eventually was run by "managers" I bet it lost much of its effectiveness. But how do we change this? For starters there is a lot of money involved for the institutions conducting this (apparently shoddy) research.