A test that means the difference between knowing a diagnosis and not knowing a diagnosis is often management-changing. If this test gives you the diagnosis, even if there is no therapeutic option for that disorder, then it means that your search for an answer is through (ending the need for continued testing, anxious uncertainty, etc).

In other posts I offer examples of tests that would not change management, and surely experienced clinicians could offer myriad other examples. (A lot of it depends on the clinical context.)

But thank you for bringing this up; I agree that a diagnosis is, in itself, valuable for patients (and care providers).