Commentary

Abstract

As technology progresses new and additional diagnostic tests become available for various conditions.  In imaging the newer tests are evaluated and validated by comparing their performance against an “established” test or a reference standard. Obuchowski et al argue that this alone may be insufficient to establish their interchangeability.  They suggest formal evaluation of this before replacing one test with another.  This is an important concept as we as radiologists often pay little attention to this aspect of technological evaluation.  The key point to remember is that just because a test has similar diagnostic performance may not necessarily make it directly interchangeable.