This is a very interesting question. I suspect that many, many, many tests requested by the ED (A & E) could be categorized as "unnecessary" during analysis in the quiet time after "The Storm", i.e., the results had no relevance to the patients' treatment. But, in the moment (especially, during trauma), the medics have very little idea of what clinical data is important at that time. A "shotgun" approach seems to be appropriate. I'm not sure there is a way to meter or control this process (other than having extremely savvy ED staff). It seems to be a necessary evil - the need for rapid turn around overweighs the concern of "over-ordering" tests.
And, as Malcolm says, a little extra time for the BB to do its work is always appreciated.