0
$\begingroup$

I have two different medical diagnostic tests, both test the same condition (binary outcome). The condition in question is rather vaguely defined, so they don't always agree. What would be the best way to quantify the "degree of concordance" between the two tests?

$\endgroup$
  • $\begingroup$ If the condition is vaguely defined, the issue then is not about the agreement between the tests, but what they are calibrated to detect. Poor eGFR and poor creatinine clearance are two indicators of CKD, both indicate related but not identical ways in which the kidneys are failing. Which test is "right" or "better" depends greatly upon the intended purpose. $\endgroup$ – AdamO Jul 31 '14 at 20:21
1
$\begingroup$

I think that Cohen's Kappa is a well known measure with good properties that assesses the interrater agreement of two medical tests of a binary outcome.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.