Mayo study review

Burke Harry harry.burke at GMAIL.COM
Thu Apr 6 17:36:10 UTC 2017


van Such M, Lohr R, Beckman T. Extent of diagnostic agreement among medical referrals. J Eval Clin Pract 2017;1:1-5.

It is not clear what the authors meant by “Few patients presented with specific diagnoses.” (p. 3)

The “different diagnoses” group had higher utilization rates, which suggests that they received many tests and procedures. It could be that some patients were sent to the Mayo for just that purpose, to be more fully worked up, and that the diagnosis they carried with them to the Mayo was simply a dummy variable necessary for the referral.

Medical referrals tend to not represent common, easily diagnosed cases.  Rather, they tend to be ambiguous cases; either because the symptoms are nonspecific, the diagnosis is itself diffuse, or because it is a rare disease. This means that there is an inherent diagnostic bias toward divergent diagnoses in referral cases.

It is not clear whether the Mayo diagnosis was related to the referral condition, or if it was a different condition that was discovered at the Mayo.

There was no standard metric for determining the correct diagnosis, other than autopsy. The Mayo physicians had to agree on a diagnosis, but what was their Cohen’s kappa?

The study mixed various types of clinicians but it did not have the power to detect differences between the types of clinicians.


Harry B. Burke, MD, PhD





Moderator: David Meyers, Board Member, Society to Improve Diagnosis in Medicine




More information about the Test mailing list