Diagnostic Error as Major Issue in Healthcare reported in ECRI and ASHRM Today

Jain, Bimal P.,M.D. BJAIN at PARTNERS.ORG
Fri Aug 22 13:14:57 UTC 2014


Hi Frank,

I completely agree with all your comments which are right on target. I am soon going to look at all your work in papers mentioned by you.


1.       The typicality of presentations of a given disease is distributed normally in different patients as you mention. This is a characteristic distribution of any trait e.g height, intelligence which depends upon a number of independent factors; symptoms, age, sex, risk factors etc. in case of typicality of a presentation.

2.       It is extremely important, I believe, for physicians to be aware of this wide range if they are to diagnose a disease when its presentation is atypical. Experienced physicians who have encountered  a large number of patients with a given disease over a long career may be aware of this wide range. Novice physicians (students, trainees) could be given second hand experience of this wide range, so to speak,by displaying presentations of a large number of patients with a given disease on a computer file. Your experimental study with midi cal students seems to support the premise that exposure to a wide range improves diagnostic ability specially in those patients with atypical presentations.

3.     Another important point to note is  that a Bayesian approach may be a hindrance to making a correct  diagnosis when the presentation is highly atypical. In such a patient, the pretest probability would be extremely low indicating very low plausibility or pretest evidence which may lead a physician to rule out a disease outright without further testing. We need to look more closely, I suggest at how diagnosis is performed in such patients in actual practice.

With warm regards,

Bimal


4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.  IsuggestFrom: Papa, Frank [mailto:Frank.Papa at unthsc.edu]
Sent: Thursday, August 21, 2014 3:08 PM
To: Society to Improve Diagnosis in Medicine; Jain, Bimal P.,M.D.
Subject: RE: [IMPROVEDX] Diagnostic Error as Major Issue in Healthcare reported in ECRI and ASHRM Today

The issues brought via Dr Ely's presentation invoke four important considerations in efforts to better understand the factors underlying diagnostic accuracy and error. First, is the notion that at the bedside, the vast majority of diseases are 'ill-defined' in that they lack an unambiguous and specific set of necessary and/or sufficient signs and symptoms by which that disease can be 'always' be correctly diagnosed at the bedside. Second, given that most patient presentations of a given ill-defined disease are neither classically prototypical (in that the case presents with all the classical signs and symptoms of the disease at hand), nor extremely atypical (the case portrayal contains very few and only the less frequently encountered of all the classical signs and symptoms associated with the disease at hand), one could safely assume that the vast majority of case portrayals of an ill-defined disease represent 'permutations' of the disease's associated signs and symptoms. Third, if one placed each specific portrayal of all these possible permutations along a line, the result would be the creation of a 'gradient' of case portrays ranging from the prototypical through very atypical. Fourth, when considering the large number of resulting permutations (case portrayals of specific sets of sign and symptoms) along this gradient, most patient presentations of a given ill-defined disease are likely to lie along the middle and not at the extremes of classically prototypical or extremely atypical. Several of these issues are presented in the following: Papa FJ. Learning Sciences Principles that Inform New Approaches to Diagnostic Training for Future Health Care Providers. Diagnosis, 2014 1(1), 125 - 129)

Given these considerations (realities?), what do we know about diagnostic accuracy/error as a function of case portrayals varying along what has been referred to as a 'disease/case typicality gradient?'

Our work has demonstrated that diagnostic accuracy is a function of a case's typicality (i.e., accuracy is a function a degree to which a given case's set of signs and symptoms lies along this theoretical, disease/case typicality gradient). Papa FJ & Elieson W. Diagnostic Accuracy as a Function of Case Prototypicality. Academic Medicine. 69,10,S58-S60, 1993; and, Papa FJ, Stone RC & Aldrich DG. Further Evidence of the Relationship Between Case Typicality and Diagnostic Performance: Implications for Medical Education. Academic Medicine. 71,1, S10-S12, 1996.

Simply put, the closer the given case portrayal aligns towards the prototypical end of the gradient, the more likely it will be correctly diagnosed; the closer the given case portrayal aligns towards the atypical end of the gradient, the less likely it will be correctly diagnosed. Given that most cases lie along the middle of this gradient, we could assume that most errors involve cases around the midpoint of this gradient (thus, the point made by both Bimal and Art that errors associated with less than typical portrayals of common diseases are much more likely than fairly typical case presentations of rare diseases).

Given these considerations (realities?), what instructional activities might we undertake in efforts to improve the diagnostic accuracy of future health care providers?

Other than a few recent investigations wherein investigators have utilized Dual Processing Theory as a framework with which to design new approaches to diagnostic training, there is limited experimental evidence concerning how best to train students to perform differential diagnosis. A few years ago we compared 2 different methods for training 2nd-year medical students to perform differential diagnosis (DDX) of heart failure: a traditional classroom-based lecture involving the use of a single 'prototypical case portrayal for each of the common and important causes of heart failure (control group) versus a learning sciences-based approach to DDX instruction (largely, a System 1 based approach) implemented through a computer based tutor and involving training cases representing points along a typicality gradient for each of the same common and important heart failure differentials (treatment group).

The results demonstrated that the treatment group diagnosed correctly significantly more test cases than the control group (74% versus 60%, respectively; effect size (Cohen's d), 1.42). Further, the treatment group also diagnosed correctly significantly more cases at the extremes of the typicality gradient: 81% versus 65%, respectively, for the prototypical/easy cases; 65% versus 48%, respectively, for the most atypical/difficult cases. In our conclusion we suggested that the ability to perform differential diagnosis is enhanced by training based upon principles of learning sciences including the notions that diagnostic training should include the explicit representations of case portals representing the varying degress of 'typicality' with which any given disease might present. Papa FJ, Oglesby MW, Aldrich DG, Schaller F, Cipher DJ. Improving diagnostic capabilities of medical students via application of cognitive sciences-derived learning principles. Medical Education, 2007; 41: 419-425.

Frank


Frank J Papa, DO, PhD
Professor, Medical Education and Emergency Medicine
Director, TCOM Academy of Medical Educators
Associate Dean, Curricular Design and Faculty Development
University of North Texas Health Science Center




From: Jain, Bimal P.,M.D. [mailto:BJAIN at PARTNERS.ORG]
Sent: Thursday, August 21, 2014 9:25 AM
To: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>
Subject: Re: [IMPROVEDX] Diagnostic Error as Major Issue in Healthcare reported in ECRI and ASHRM Today

Art, your comment on variation in presentations of skin diseases is very enlightening. Is it possible for you to put presentations of a particular disease say herpes simplex in different patients in a computer file that I and others can easily access. We could then perhaps learn to diagnose it like you do. I wish somebody would do for diseases like pulmonary embolism, myocardial infarction, pneumonia what you have done for skin diseases. Thank you for your great comment.

Bimal

From: Art Papier [mailto:apapier at logicalimages.com]
Sent: Thursday, August 21, 2014 9:41 AM
To: 'Society to Improve Diagnosis in Medicine'; Jain, Bimal P.,M.D.
Subject: RE: [IMPROVEDX] Diagnostic Error as Major Issue in Healthcare reported in ECRI and ASHRM Today


Bimal, This is a great comment on variation.  We too often focus on the "rare birds" instead of diagnostic error due to variants of common disease.  We teach medical students about pemphigus vulgaris (see a new patient with pemphigus about once every 10 years) but for example how often do we insure that primary care physicians understand the variations of herpes simplex.  Practicing in a rural community I had a patient treated with Keflex about 6 times over 3-4 years for grouped vesicles on the palm.  The PCP thought it was bacterial.  The PCP was likely trained to look for herpes on the lips and genitals, and did not see the pattern of the grouped vesicles because of the thickness of the stratum corneum of the palm, making it difficult to see the lesions as vesicles.    Also had probably been taught that herpes does not occur anywhere which it can.  In the same community months later I had a hospital consult for an OB patient with grouped vesicles on the thigh in the same community, obvious herpes simplex, but the hospitalist and attending could not diagnose it.   This is anecdotal, but I can tell you that all dermatologists worldwide are gritting their teeth about the crazy number of "bilateral cellulitis" consults for obvious stasis dermatitis, and host of other repeatable diagnostic mistakes of common presentations.

In our work we have been systematically collecting cases of disease presenting with skin, eye, oral mucosal, hair and nail findings for 15 years.  We have over 100,000 images that are systematically catalogued and obtained from leading institutions and experts.  If you look through an expansive set of images for a disease, you can easily see variants that are simply not in the text books.  For instance a fixed drug eruption presents most typically with a single or few round patches on the genitals or face or perhaps an extremity.  A known variant is bullous fixed drug eruption.  The cases we have systematically collected show many fixed drug eruptions localizing to the web spaces of the hand. Having had the  I had a patient with a lesion in the web spaces treated with lotrimin and other antifungals and it was fixed drug.  We teach to the classic and then we expect clinicians to generalize to the variants.  It seems to be a flawed premise.  From a cost perspective, my hunch is that the volume of error around variants of common dwarfs the missed rare birds.  Art



[http://scomber:8080/resizeimage?ImagePath=/ImageArchive/81166-81176/81171.jpg&Fit=1&Height=600&Width=800&Quality=.75&ScaleMethod=SCALE_AREA_AVERAGING]



Art Papier, MD

Chief Executive Officer

585.272.2630 | apapier at logicalimages.com<mailto:apapier at logicalimages.com>

______________________________



www.visualdx.com<http://www.visualdx.com>





-----Original Message-----
From: Jain, Bimal P.,M.D. [mailto:BJAIN at PARTNERS.ORG]
Sent: Thursday, August 21, 2014 6:49 AM
To: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>
Subject: Re: [IMPROVEDX] Diagnostic Error as Major Issue in Healthcare reported in ECRI and ASHRM Today



I did not have a chance to comment during John Ely's interesting webinar yesterday as I joined in by telephone. Here are my thoughts about his important study.

1. A major reason for failure to broaden differential diagnosis may be our lack of awareness of the wide variation in clinical presentations of a given disease ranging from highly typical to highly atypical in different patients. This may be due to medical school and textbook teaching which focuses primarily on typical presentations of a disease. Awareness of variation could be increased, I suggest, by looking at summaries of presentations in about one hundred consecutive patients with a given disease seen at a large medical institution. These summaries could be stored in a computer file which could be made available to all interested physicians.



2. An atypical presentation is often interpreted as low pretest evidence due to low pretest probability of a disease which may be ruled out without further testing. I believe, a presentation should be looked upon as a clue to a disease and not pretest evidence which should be ruled in or out by further testing. This would avoid premature closure.



3. The rule of not neglecting base rate is a statistical rule which may not apply in a given, individual patient as we saw in the patient found to have neurosyphilis. I believe it is all right to break this rule by suspecting a rare disease if all other suspected diseases are found not to be present.



4. There is so lttle we know about how diagnosis is performed in actual practice or should be performed to minimize diagnostic errors. I would like to thank John Ely and other investigators who have started to look at diagnosis in actual practice and increase our understanding of this important process.



All the best,



Bimal



Bimal P Jain MD

Pulmonary-Critical Care

NorthShore Medical Center

Lynn MA 01904



-----Original Message-----

From: Ruth Ryan [mailto:rryan at LAMMICO.COM]

Sent: Tuesday, August 19, 2014 10:01 AM

To: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>

Subject: Re: [IMPROVEDX] Diagnostic Error as Major Issue in Healthcare reported in ECRI and ASHRM Today



Bravo and hats off to you, Jeffery!  If critical thinking skills can be built into the community college program training paramedics in Bossier Parish, North Louisiana, never let it be said such training is ethereal, too difficult, or relevant only to the ivory tower.



Ruth

Ruth Ryan RN, BSN, MSW, CPHRM

Senior Risk Management Education Specialist LAMMICO

1 Galleria Blvd., Suite 700

Metairie, LA 70001

E-Mail rryan at lammico.com<mailto:rryan at lammico.com>

Telephone (504) 841-2736

Fax (504) 841-5312





-----Original Message-----

From: Jeffery Anderson [mailto:janderson at BPCC.EDU]

Sent: Sunday, August 17, 2014 9:26 PM

Subject: Re: Diagnostic Error as Major Issue in Healthcare reported in ECRI and ASHRM Today



I am developing a module on critical thinking for my paramedic program.  Besides using Dr. Croskerry's articles and information from the Foundation for Critical Thinking, I am also using the textbook Patient Safety in Emergency Medicine by Croskerry and Emergency Medicine Decision Making by Weingart as references.  In addition I have been following the work of Dr. Kevin deLaplante of the Critical Thinker Academy.



Jeffery D. Anderson, NREMT-P

Paramedic Program Director

Bossier Parish Community College

Office Phone (318) 678-6403

To unsubscribe from the IMPROVEDX:

mail to:IMPROVEDX-SIGNOFF-REQUEST at LIST.IMPROVEDIAGNOSIS.ORG

or click the following link: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>



Address messages to: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>



For additional information and subscription commands, visit:

http://www.lsoft.com/resources/faq.asp#4A






Visit the searchable archives or adjust your subscription at:

http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?INDEX



Moderator: Lorri Zipperer Lorri at ZPM1.com<mailto:Lorri at ZPM1.com>, Communication co-chair, Society for Improving Diagnosis in Medicine



To unsubscribe from the IMPROVEDX list, click the following link:<br> <a href="http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?SUBED1=IMPROVEDX&A=1" target="_blank">http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?SUBED1=IMPROVEDX&A=1</a<http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?SUBED1=IMPROVEDX&A=1%3c/a>>

</p>



Confidentiality Notice: The information contained in this e-mail (and/or the documents accompanying it) is private and confidential information belonging to the sender.  The information is intended only for the use of the individual(s) identified above and others who have been specifically authorized to receive it. If you are not the intended recipient or believe that you have received this communication in error, do not print, copy, retransmit, disseminate, or otherwise use the information. If you have received the transmission in error, please alert the sender that you have received this email by reply e-mail and delete the copy you received. Thank you.



To unsubscribe from the IMPROVEDX:

mail to:IMPROVEDX-SIGNOFF-REQUEST at LIST.IMPROVEDIAGNOSIS.ORG

or click the following link: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>



Address messages to: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>



For additional information and subscription commands, visit:

http://www.lsoft.com/resources/faq.asp#4A






Visit the searchable archives or adjust your subscription at:

http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?INDEX



Moderator: Lorri Zipperer Lorri at ZPM1.com<mailto:Lorri at ZPM1.com>, Communication co-chair, Society for Improving Diagnosis in Medicine



To unsubscribe from the IMPROVEDX list, click the following link:<br> <a href="http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?SUBED1=IMPROVEDX&A=1" target="_blank">http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?SUBED1=IMPROVEDX&A=1</a<http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?SUBED1=IMPROVEDX&A=1%3c/a>>

</p>





The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.



To unsubscribe from the IMPROVEDX:

mail to:IMPROVEDX-SIGNOFF-REQUEST at LIST.IMPROVEDIAGNOSIS.ORG

or click the following link: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>



Address messages to: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>



For additional information and subscription commands, visit:

http://www.lsoft.com/resources/faq.asp#4A






Visit the searchable archives or adjust your subscription at:

http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?INDEX



Moderator: Lorri Zipperer Lorri at ZPM1.com<mailto:Lorri at ZPM1.com>, Communication co-chair, Society for Improving Diagnosis in Medicine



To unsubscribe from the IMPROVEDX list, click the following link:<br> <a href="http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?SUBED1=IMPROVEDX&A=1" target="_blank">http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?SUBED1=IMPROVEDX&A=1</a<http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?SUBED1=IMPROVEDX&A=1%3c/a>>

</p>

________________________________

Address messages to: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>

To unsubscribe from IMPROVEDX: click the following link:
http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?SUBED1=IMPROVEDX&A=1
or send email to: IMPROVEDX-SIGNOFF-REQUEST at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX-SIGNOFF-REQUEST at LIST.IMPROVEDIAGNOSIS.ORG>

Visit the searchable archives or adjust your subscription at: http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?INDEX

Moderator: Lorri Zipperer Lorri at ZPM1.com<mailto:Lorri at ZPM1.com>, Communication co-chair, Society for Improving Diagnosis in Medicine

To learn more about SIDM visit:
http://www.improvediagnosis.org/








HTML Version:
URL: <../attachments/20140822/8a4ae4d6/attachment.html> ATTACHMENT:
Name: image001.jpg Type: image/jpeg Size: 88941 bytes Desc: image001.jpg URL: <../attachments/20140822/8a4ae4d6/attachment.jpg>


More information about the Test mailing list