Then all you get is data about the effectiveness of XY. Would it be any better if only X was tested, but it turned out that that one drug 'X' contained two separate active ingredients, each of which the pharmaceutical company thought would probably be an effective treatment?