The average age of physicians in the US is increasing, and there is ample evidence that cognitive impairment increases with age, as well as do adverse patient outcomes. It has been suggested that physicians of a certain age undergo periodic screenings for cognitive decline.

However, Primary Care Physicians (PCP), as well as many highly trained individuals, do better on Mini Mental Status Exams (MMSE) than their non-professional peers. I believe that's in part because PCPs use one of the many mini-mental status exams frequently enough to have the answers and the mechanics of the questions memorized.

It is a test that has become one of the most widely used internationally for the diagnosis and clinical prognosis of cognitive impairment, mainly in elderly patients.

Yale requires cognitive testing in physicians 70 years old and above, but the (secret) test is only 16 items long. The MMSE and the Montreal Cognitive Assessment (MoCA) exams each have 30 items. Regardless of which one physicians use, they are basically the same. Having administered these tests for at least 5 years in practice, these tests were committed iron clad in my memory. I believe a physician would need to be severely impaired by the time they fail either of these tests. Extensive neuropsychiatric exams are just more of the same.

The best paper I could find had suggestions but no details.

Is there any exam that measures cognitive impairment in physicians sensitively?

Sorry about the tag; neither cognition nor competency is a tag here. Also, please know that I am totally in favor of mandatory testing. But it should be sensitive, i.e. detect cognitive impairment before patients start to suffer. There's a difference between malpractice caused by hubris and cognitive impairment.

  • 1
    Would this be better off in Psych&Neuro?
    – Carey Gregory
    Sep 3, 2023 at 18:23
  • @CareyGregory - I thought about that, but didn't know. it's about both medicine and neuropsychiatry. Should I delete and post there? Sep 3, 2023 at 21:10
  • 1
    @anongoodnurse a sort of "quis custodiet ipsos custodes", though more like "quis custodiet ipsos doctores" in this case. Very interesting question and I suspect one that hasn't been covered thoroughly by the literature.
    – bob1
    Sep 3, 2023 at 21:18
  • 1
    Incidentally, a couple of years ago I participated in a clinical trial that required me to take the MoCA several times per year for two years. It got boring and repetitive, so I started using the proper Latin names for the animals you have to name. It amused me knowing that somebody probably had to use google to find out if I was right.
    – Carey Gregory
    Sep 3, 2023 at 22:04
  • 1
    That's hilarious! I would have loved it! Sep 4, 2023 at 15:46

1 Answer 1


I had a bit of a search and found 3 papers of relevance. Only one of these indicate the tests used, but the others do indicate that there are tests available for use, though I suspect mostly in-house systems.

The Cooney et al. paper indicates that Yale uses a battery of tests (all secret) to administer to their physicians. The tests seem to test more than just memory, though mostly focussing on higher cognitive function:

The screening battery used consisted of 16 brief tests including rudimentary information processing (2 tests), visual scanning and psychomotor efficiency (2 tests), processing speed and accuracy under decision load (1 test), concentration and working memory (1 test), visual analysis and reasoning (2 tests), verbal fluency (2 tests), memory (1 visual test and 1 verbal test), “prefrontal” self-regulation (1 test), and executive functioning (3 tests).

The Armstrong paper is an editorial on the Cooney one, but has some very relevant links, including the following review, indicating that machine learning might be the objective way forward:

Finally, the Del Bene and Brandt paper indicates that they used the following tests:

tests used

I make no commentary on these tests, as it isn't my specialty at all, but they seem appropriate to me and to have similar aims as the Yale tests.

  • 1
    Thanks for this! It's exciting that Del Bene and Brandt are using other physicians as the control group rather than the general population, and use testing unfamiliar to most physicians. Sep 4, 2023 at 16:00

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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