Discussion Board

Expand all | Collapse all

Lists of Common Diagnostic Errors

  • 1.  Lists of Common Diagnostic Errors

    Posted 04-30-2020 00:35
    Dear all,

    I have wondered if there are lists collected by specialists or others of
    the most commonly seen Errors in Diagnosis.

    Is such data available/used anywhere - perhaps in another country?

    If not available would such data be of value to physicians?

    Or would it confuse?

    Could such lists alert physicians to colleague mistakes and decrease diagnostic error rates?

    Rob Bell, M.D.


  • 2.  RE: Lists of Common Diagnostic Errors

    Posted 04-30-2020 16:43
    Recommend we first define error:
    It is not an error if the second party relies on information ruled in/out by the first.
    This tendency to find error in retrospect is in and of itself a most grievous error.
    What is the name of this error and can you ID it in yourself when you use it? If not, that's an error.

    If I can get permission, I'll wear a wire for a few patient interviews and publish a word for word transcript, no other data.  That would be so cool.....

    The people to whom I refer of course refine my diagnoses. that's called a system. When refined, my initial diagnosis is an error.

    It is rare indeed that my admitting diagnosis is the final diagnosis. Are these errors or would the error be sending home the patient needing further diagnostics?

    tom benzoni





  • 3.  RE: Lists of Common Diagnostic Errors

    Posted 05-01-2020 06:45
    Dr Benzoni

    You have hit the nail on the head !!!

    A prelim dx or working dx can NOT be an error by definition (unless the clinical reasoning was "eggretiously wrong" however we want to define that... lol)

    it's a "working hypothesis" that is either confirmed or denied ("refined") during further evaluation and observation
    (which can occur during hrs or days or weeks or months or yrs depending on the severity of the physiologic derangement)

    Until we fully acknowledge that fact, we will continue to struggle with the whole underlying conceptual framework and dx errors will erroneously be counted as "failures" rather than as "successes" that evolved as more data was accumulated over time

    the real issue that we haven't figured out (besides effectively communicating with the pt) is how to build rescue strategies into the care plan that surveils the pt for worsening symptoms/derangement and "rescues" the pt prior to harm occurring... the failure to alert and failure to react/rescue are the true dx errors in healthcare not the mere fact that we didnt recognize that a specific dx was present in some arbitrary and artificial time frame

    just my two cents.....

    Respectfully

    Tom Westover MD









  • 4.  RE: Lists of Common Diagnostic Errors

    Posted 05-01-2020 09:30

     

     

    May 1, 2020

    8:36 AM

    Dr. Westover

     

    As to " the real issue that we haven't figured out"; the aviation industry tackled this years ago when aircraft crashes became unacceptable.  It is done through a concept called Accommodation.  For each critical fault (those that could lead to a catastrophic event), there must be fault detection and a secondary control scheme in standby, so that if the fault occurs the standby method automatically implements to avoid a catastrophic scenario.  Believe it or not, these type of faults occur routinely in daily flights.  You normally do not see the catastrophic effect because the backup handles the situaton until on the ground.

     

    Requiring a DDX that is carried through to full issue confirmation, is an attempt to have a secondary strategy.  This should cause some critical thinking on the part of the clinician.  However, it hasn't been figured out how to "automatically" implement an observation or care plan in the event the patient takes a significant downward turn (except perhaps in the hospital setting).

     

    Even with this excellent strategy in aviation, there is still no remedy for human mistakes that do not thoroughly investigate the fault possibilities - aka B737-800 MAX.

     

    One way to improve the observation is to clearly identify to the patient that they are a partner in the care and must be on heightened alert.  For those that have been harmed by medical mistakes or have lived a long time without the benefit of a confirmed diagnosis, I think they know the need to fully participate; however for the general population, they still depend on the clinician to fix the situation.

     

       Nelson Toussaint

     

    TAMARAC LLC

    860-844-0199

    ntoussaint@tamarac.com

     






  • 5.  RE: Lists of Common Diagnostic Errors

    Posted 05-01-2020 10:00
    I agree 100%

    But it's even worse than you think ... because our current EMR's "force" us (enable us) to carry our  incorrect / incomplete dx forward in time ..

    The so called "sloppy and paste"..

    WITHOUT a built in safety mechanism that requires us to confirm, deny or refine our original hypothesis 

    Tom





  • 6.  RE: Lists of Common Diagnostic Errors

    Posted 05-01-2020 10:46
    Re back up systems to prevent catastrophic errors.  I note yesterday's issue of JAMIA (see below) found that even for the most dire CDS warnings, the override rate is 90%.  Of course we know some of the warnings are off target. And we know the usual override rate of CDS is from 50% to 99% (not a mistype). But 90% for what the good folks at Harvard considered to be the highest priority...

    High-priority drug-drug interaction clinical decision support overrides in a newly implemented commercial computerized provider order-entry system: Override appropriateness and adverse drug events

    Journal of the American Medical Informatics Association, ocaa034, https://doi.org/10.1093/jamia/ocaa034
    Published:
     
    26 April 2020



    Ross Koppel, Ph.D. FACMI, UNIV. OF PENNSYLVANIA

    Prof. of Biomedical Informatics, Perelman Sch of Medicine.

    Senior Fellow, Wharton's Leonard Davis Institute of Healthcare Economics;    

    Senior Fellow, Center for Public Health Initiatives, Perelman Sch of Medicine; 

    Adjunct Professor (full) Sociology Department;    

    Affil Prof of Medicine, Perelman Sch of Medicine;  

    Prof. of Biomedical Informatics, SUNY@Buffalo    

    rkoppel@sas.upenn.edu