When John Thompson and I first began talking about finding a way to measure and cost the output of hospitals in the 1960s. But it seemed important to us to begin to understand what up until then seemed unexplainable - the rather strange cost behavior of hospitals.