[External] Re: [IMPROVEDX] How to convey uncertainty

Edward Winslow edbjwinslow at GMAIL.COM
Mon Jan 15 17:57:25 UTC 2018


Great discussion about the diagnostic process:

John, I like your first suggestion about approach to a symptom, when you
tell the parent that A, B, of C could be the cause of your child's symptoms
- even though A is the most likely, we need to consider and test for (I
personally don't like the term "rule out") B & C. (you made that choice
because B & C are much more dangerous).
Then Art points out that we may also need to consider D & E, which we
didn't think of - they may also be common clinical conditions, with serious
consequences, but with atypical presentations.
I'm reminded of an aphorism that I was taught (a long time ago):
"Uncommon presentations of Common Diseases are more likely than Common
Presentations of Uncommon Diseases"
This was to encourage us to look for common conditions (that we shouldn't
miss, even if they weren't presenting "the usual way") at least as early as
we were looking for "Zebras".

I'm interested in the emphasis of not being able to remember "everything"
and not on "the complete H & P". I know that time constraints in "the
Clinic" seem to raise a barrier to the H&P that we did in Medical School. I
wonder whether the likelihood of diagnostic error is in fact related to the
truncation of the H&P for a "new" problem.

I am impressed that it seems that none of our Diagnostic Aids, from the IT
PoV, are aids to *getting* more complete information, but are aids to
organizing the information that we have. Is there a way to get the review
of symptoms or HPI done more effectively than by Dr. Falchuk, described in
Groopman, who took a thick chart, put it aside, and started again at the
beginning?
How can we help physicians "listen to the patient's story"?




On Mon, Jan 15, 2018 at 11:01 AM, Art Papier <apapier at visualdx.com> wrote:

> Hi John,
>
> Thanks for your thoughtful reply.  Here are some responses:
>
> 1.       Agree 100% and that is what we do, enter symptoms and other
> findings to develop a ddx, and simultaneously show which presentation is
> the best match for each Dx in the differential
>
> 2.       Yes every type of practice/specialty has a different severity
> and incidence of low or high acuity diseases.  A quick web search says
> there are 250,000 cases of appendicitis in the US/yr, and 1500 cases of
> acute intermittent porphyria.  Seems that if we knew the approx. variation
> rate of each disease we could calculate the likelihood of variant
> presentations vs rare diseases.  Approx 800,000 MI’s a year, 600,000
> strokes…just seems these numbers dwarf the rare diseases.
>
> 3.       You are correct, but the culture would change if we examined and
> credentialed doctors differently.  You create boards and exams and that is
> what students focus on.  We allow grade school students to use calculators
> in the exam, we train pilots on their instruments, but in medicine most
> still memorize the route instead of using the map and checklist.  We need
> to test and train students and residents in how to access information, so
> they continue to do that lifelong.  The good news practice is changing,
> even with medical education is slow to adapt.  There is a wide spectrum of
> behavior out there, with many clinicians using technology to aide
> decisions.  We have data to prove that.  Doctors and the culture of
> medicine are evolving use of information at the point of care will greatly
> accelerate as patients start using more and more information to guide their
> decisions.  The patients will drag many doctors by their ears into the 21
> st century, some will retire, and practice will change.  We are just late
> to the information revolution we are living through.
>
> Best
>
> Art
>
>
>
> *Art Papier MD*
>
> CEO
>
> *phone* 585-272-2630 <(585)%20272-2630>
>
> *mobile* 585-615-8245 <(585)%20615-8245>
>
> *address* 339 East Ave, Suite 410 Rochester, NY 14604
> <https://maps.google.com/?q=339+East+Ave,+Suite+410+Rochester,+NY+14604&entry=gmail&source=g>
>
> *web* visualdx.com
>
> *email* apapier at visualdx.com
>
>
>
> [image: https://newoldstamp.com/editor/images/f.jpg]
> <http://facebook.com/visualdx>  [image:
> https://newoldstamp.com/editor/images/tw.jpg]
> <http://www.twitter.com/visualdx>  [image:
> https://newoldstamp.com/editor/images/in.jpg]
> <http://linkedin.com/company/logical-images>  [image:
> https://newoldstamp.com/editor/images/instagram.jpg]
> <http://instagram.com/visualdximages>
>
> [image: cid:image005.png at 01D159E6.3E7218B0] <http://www.visualdx.com/>
>
>
>
>
>
> [image: cid:image006.jpg at 01D159E6.3E7218B0]
>
>
>
>
>
> *From:* Ely, John [mailto:john-ely at UIOWA.EDU]
> *Sent:* Monday, January 15, 2018 11:15 AM
>
> *To:* IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG
> *Subject:* Re: [IMPROVEDX] [External] Re: [IMPROVEDX] How to convey
> uncertainty
>
>
>
> All great points Art.  Some reactions:
>
>
>
> 1.  I totally agree that variants are important to include in any
> discussion of any disease, and that these variants are often ignored in
> information resources (VisualDX being a rare exception).  But it might be
> even more helpful to start with symptoms rather than diseases.  So instead
> of listing the different ways herpes can present, we could list the
> different causes of blisters in the palm, listed in order of prevalence,
> and with the distinguishing features of each.  What distinguishes herpes
> from bacterial infection, candida, dyshidrosis, and scabies?
>
>
>
> 2.  Also agree that missed diagnoses (MI, PE, cancer, sepsis, ruptured
> aneurysm and stroke) are not rare.  But they are relatively uncommon in
> primary care.  These diseases probably account for less than 1% of patients
> with the symptoms they typically present with (chest pain, abdominal pain,
> fever, etc.)  By far, the most common cause of all these symptoms will be
> some benign, spontaneously resolving disease that never gets definitively
> diagnosed.  Next most common will be benign diseases that do get diagnosed,
> like chest wall pain, abdominal wall pain, reflux esophagitis, benign viral
> infections, and so on.  Specialists are less likely to see these patients
> and of course they are never presented at M&M, so we get a distorted view
> of their prevalence.  The chance that a patient with an atypical
> presentation for MI will actually have an MI is probably even lower than 1%.
>
>
>
> 3.  Also agree that we are crazy to depend on the unaided human mind to
> memorize all the variants or even to reliably generate a comprehensive
> differential diagnosis.  But that craziness is deeply embedded in our
> culture as physicians.  Generating the differential diagnosis from memory
> is what we do.  Just look at any interaction between a teaching physician
> and a trainee -- whether it be one-to-one, or small group, or big lecture.
> We ask the trainee to generate the differential from memory.  And we do the
> same in real time with real patients.  It’s terribly unreliable and
> dangerous but it’s who we are.
>
>
>
> John
>
>
>
> John Ely, MD
>
> University of Iowa
>
>
>
> *From:* Art Papier [mailto:apapier at visualdx.com <apapier at visualdx.com>]
> *Sent:* Sunday, January 14, 2018 4:04 PM
> *To:* 'Society to Improve Diagnosis in Medicine'; Ely, John
> *Subject:* RE: [IMPROVEDX] [External] Re: [IMPROVEDX] How to convey
> uncertainty
>
>
>
> John, Let’s add to the discussion below,  A1, A2, A3…etc.. which I am
> using to symbolize the variants of common dignoses that are responsible for
> diagnostic error.  One of our realizations in working in developing CDS
> over the past 20 years is the need to account for variation of the common.
> Over the past 40 years many wonderful tools have been developed to expand
> the differential, but such tools need to be balanced in their design with
> bringing forward common variants in the DDx.  When you look at closed
> claims, the missed diagnoses are for the most part not rare
> diagnoses…clinicians are missing MI, PE, Cancer, Sepsis, ruptured aneurysm,
> Stroke etc……these are all common diagnoses, and certainly diagnoses we all
> know.  Yet they are all frequently missed.  If we put aside system issues
> as causes, and focus on cognitive mistakes,  shouldn’t we elevate
> representative bias in these discussions, and our solutions?  We have
> memorized illness scripts based on classic presentations, not “variant
> presentations”.   As long as medical education is exam and board exam
> results focused thereby encouraging extraordinary, non-human feats of
> memory and processing, we will continue promote the myth that we can
> evaluate simultatenously for variants of common and rare diagnoses,
> flawlessly,  and every 20 minutes in our clinics.  The unaided human mind
> just will not be able to see and memorize all these variants.   In my
> specialty, we see frequent diagnostic error in referral from generalists,
> and it almost always results in comments from the referring PCP saying “I
> didn’t know it could look like that” or “I didn’t know it could occur
> there”.  A very common one for example would be a recurring group of
> blisters or pustules invovling the palm, diagnosed as bacterial infection,
> not herpes.  The referring physicians just often don’t think of herpes in
> unusual locations.  Variation of the common needs to be part of our
> calculus.
>
>
>
> Best
>
> Art
>
>
>
> Art Papier MD
>
> CEO VisualDX
>
> Associate Professor of Dermatology and Medical Informatics
>
> University of Rochester College of Medicine
>
> *From:* Ely, John [mailto:john-ely at UIOWA.EDU <john-ely at UIOWA.EDU>]
> *Sent:* Friday, January 12, 2018 9:39 AM
> *To:* IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG
> *Subject:* Re: [IMPROVEDX] [External] Re: [IMPROVEDX] How to convey
> uncertainty
>
>
>
> Good point Charlie.  I think the key is that A is different from B and C.
> Typically, A is the common benign diagnosis, whereas B and C are the
> uncommon more serious diagnoses.  The problem is not that I have failed to
> rule out A.  The problem is that I have not even considered B and C.  And C
> is the correct diagnosis, but it’s not even on my radar screen.  Also I
> can’t rule out A because A is the common cold.  B is pneumonia and C is
> meningitis and they both can be ruled out – possibly by history and
> physical, possibly by testing.
>
>
>
> Studies by Mark Graber and Gordy Schiff found that the most common cause
> of diagnostic error is the failure to ever consider the correct diagnosis.
> It’s not that we consider it and reject it.  It’s that we never consider
> it.  That’s what the differential diagnosis checklist is for.  It’s a
> cognitive forcing strategy to make us consider the correct diagnosis.  We
> still might mistakenly reject it, but at least it gets us over the first
> hurdle.
>
>
>
> But as Atul Gawande said, “We don’t like checklists . . .  It somehow
> feels beneath us to use a checklist, an embarrassment.  It runs counter to
> deeply held beliefs about how the truly great among us – those we aspire to
> be – handle situations of high stakes and complexity.”  We consider
> checklists as crutches for the feeble minded.  Fortunately, the airline
> industry put a higher priority on reliability and safety than it did on the
> egos of pilots, who also initially rejected checklists.  One way to counter
> the feeble-minded concern in medicine is to encourage the physician to
> generate the differential diagnosis and primary diagnosis from their memory
> first and communicate these to the patient before using the checklist.  The
> checklist is just a method to make diagnosis more reliable.
>
>
>
> Thanks for your interest and comments.
>
>
>
> John
>
>
>
>
>
> *From:* Charlie Garland - The Innovation Outlet [mailto:cgarland@
> innovationoutlet.biz <cgarland at innovationoutlet.biz>]
> *Sent:* Thursday, January 11, 2018 10:24 AM
> *To:* Society to Improve Diagnosis in Medicine; Ely, John
> *Subject:* RE: [IMPROVEDX] [External] Re: [IMPROVEDX] How to convey
> uncertainty
>
>
>
> John, for the most part I agree with your response.  However -- as you've
> stated it -- might not what you've put forth be fundamentally biased?  If
> you see the "need" to rule out B and C (and not prepared to rule out A, *just
> as readily*), that seems to be an example of confirmation bias, at the
> very least.  At least that's how it strikes me.
>
>
>
> I don't think you're at all wrong for beginning with the *probability*
> that A is the correct Dx.  One needs to embrace a "both/and" (subjective
> and objective) orientation here.  I'm just curious whether the subtlety of
> how our logic is developed and presented ought to be at least a red flag
> for consciously considering the involvement of one or more biases.
>
>
>
> Charlie
>
>
>
> =================================================
>
>
>
> Charlie Garland
>
>
>
> Senior Fellow of HITLAB <http://www.hitlab.org/people/> (Healthcare
> Innovation & Technology Laboratory @ Columbia University Medical Center)
>
> Board Member @ The Creative Education Foundation
> <http://www.creativeeducationfoundation.org>
>
> Author of Upcoming Book: *Innovation Inspiration
> <http://www.InnovationInspiration.org>* (2018)
>
> Developer of Cognitive Buoyancy <http://www.cognitivebuoyancy.com>  ("The Trigger to Innovation")
>
> Developer of The Innovation Cube
> <http://portal.sliderocket.com/BIWIR/Cubie-TOTB> (a.k.a. CubieTM - a
> Critical Thinking & Creative Problem-Solving Tool)
>
> Certified in Advanced Polarity Thinking
> <http://www.polaritypartnerships.com/consultant-list-test/2016/5/31/charlie-garland?rq=garland>
> TM Conflict-Resolution Construct
>
>
>
> *LinkedIn:* http://www.linkedin.com/in/innovationoutlet
>
> *Twitter:* *@innovationator* <http://twitter.com/innovationator>
>
>
>
>
>
>
>
>
>
> -------- Original Message --------
> Subject: Re: [IMPROVEDX] [External] Re: [IMPROVEDX] How to convey
> uncertainty
> From: "Ely, John" <john-ely at UIOWA.EDU>
> Date: Thu, January 11, 2018 8:33 am
> To: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG
>
> The best option may be a combination of the two implicit strategies:  “You
> child’s belly pain is probably due to Disease A, but could also be due to
> Disease B or Disease C.  And this is what we need to do to rule out B and
> C.  And before you leave, I just want to run through this checklist of
> other possibilities to be sure I’m not forgetting anything else that we
> need to rule out.”
>
>
>
> John Ely, MD
>
> University of Iowa
>
>
>
> *From:* Bob Latino [mailto:blatino at RELIABILITY.COM
> <blatino at RELIABILITY.COM>]
> *Sent:* Thursday, January 11, 2018 5:23 AM
> *To:* IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG
> *Subject:* [External] Re: [IMPROVEDX] How to convey uncertainty
>
>
>
> Along these lines some friends of mine wrote this book a while back 'What
> Do I Say?' that addresses these uncomfortable conversations that need to
> take place between a physician and a patient.  This is targeted to
> obstetrics but the basic premise is about communicating information to
> people that otherwise would not want to hear it (bad news).
>
>
>
> https://www.amazon.com/Communicating-Intended-Unanticipated-Outcomes-
> Obstetrics/dp/0787966541
>
>
>
> *Robert J. Latino, CEO*
>
> Reliability Center, Inc.
>
> 1.800.457.0645 <(800)%20457-0645>
>
> blatino at reliability.com
>
> www.reliability.com
>
> [image: linkedin logo signature file]
> <https://www.linkedin.com/company/958495?trk=tyah&trkInfo=clickedVertical%3Acompany%2CclickedEntityId%3A958495%2Cidx%3A1-1-1%2CtarId%3A1464096807851%2Ctas%3Areliability%20center%2C%20inc.>
>
>
>
> *From:* Mark Graber [mailto:graber.mark at GMAIL.COM <graber.mark at GMAIL.COM>]
>
> *Sent:* Wednesday, January 10, 2018 10:56 AM
> *To:* IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG
> *Subject:* [IMPROVEDX] How to convey uncertainty
>
>
>
> Congratulations and thanks to Viraj Bhise and Hardeep Singh’s group for
> their novel study on the best way to convey uncertainty to patients about
> the diagnosis – advice we can use every day.
>
>
>
> https://academic.oup.com/intqhc/advance-article/doi/10.
> 1093/intqhc/mzx170/4791877
>
>
>
>
>
> Mark L Graber MD FACP
>
> President, SIDM
>
> Senior Fellow, RTI International
>
> Professor Emeritus, Stony Brook University
>
>
>
>
>
>
>
>
> ------------------------------
>
>
>
> 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
>
>
>
> Moderator:David Meyers, Board Member, Society for Improving Diagnosis in
> Medicine
>
> To learn more about SIDM visit:
> http://www.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
>
>
>
> Moderator:David Meyers, Board Member, Society for Improving Diagnosis in
> Medicine
>
> To learn more about SIDM visit:
> http://www.improvediagnosis.org/
>
>
> ------------------------------
>
> Notice: This UI Health Care e-mail (including attachments) is covered by
> the Electronic Communications Privacy Act, 18 U.S.C. 2510-2521 and is
> intended only for the use of the individual or entity to which it is
> addressed, and may contain information that is privileged, confidential,
> and exempt from disclosure under applicable law. If you are not the
> intended recipient, any dissemination, distribution or copying of this
> communication is strictly prohibited. If you have received this
> communication in error, please notify the sender immediately and delete or
> destroy all copies of the original message and attachments thereto. Email
> sent to or from UI Health Care may be retained as required by law or
> regulation. Thank you.
> ------------------------------
>
>
> ------------------------------
>
>
>
> 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
>
>
>
> Moderator:David Meyers, Board Member, Society for Improving Diagnosis in
> Medicine
>
> To learn more about SIDM visit:
> http://www.improvediagnosis.org/
>
>
> ------------------------------
>
> Notice: This UI Health Care e-mail (including attachments) is covered by
> the Electronic Communications Privacy Act, 18 U.S.C. 2510-2521 and is
> intended only for the use of the individual or entity to which it is
> addressed, and may contain information that is privileged, confidential,
> and exempt from disclosure under applicable law. If you are not the
> intended recipient, any dissemination, distribution or copying of this
> communication is strictly prohibited. If you have received this
> communication in error, please notify the sender immediately and delete or
> destroy all copies of the original message and attachments thereto. Email
> sent to or from UI Health Care may be retained as required by law or
> regulation. Thank you.
> ------------------------------
>
>
> ------------------------------
>
>
>
> 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
>
>
>
> Moderator:David Meyers, Board Member, Society for Improving Diagnosis in
> Medicine
>
> To learn more about SIDM visit:
> http://www.improvediagnosis.org/
>
>
> ------------------------------
>
> Notice: This UI Health Care e-mail (including attachments) is covered by
> the Electronic Communications Privacy Act, 18 U.S.C. 2510-2521 and is
> intended only for the use of the individual or entity to which it is
> addressed, and may contain information that is privileged, confidential,
> and exempt from disclosure under applicable law. If you are not the
> intended recipient, any dissemination, distribution or copying of this
> communication is strictly prohibited. If you have received this
> communication in error, please notify the sender immediately and delete or
> destroy all copies of the original message and attachments thereto. Email
> sent to or from UI Health Care may be retained as required by law or
> regulation. Thank you.
> ------------------------------
>
>
> ------------------------------
>
>
>
> 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
>
>
>
> Moderator:David Meyers, Board Member, Society for Improving Diagnosis in
> Medicine
>
> To learn more about SIDM visit:
> http://www.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@
> LIST.IMPROVEDIAGNOSIS.ORG
>
> Moderator:David Meyers, Board Member, Society for Improving Diagnosis in
> Medicine
>
> To learn more about SIDM visit:
> http://www.improvediagnosis.org/
>



-- 
*Edward B, J. Winslow, MD, MBA*
Home 847 256-2475; Mobile 847 508-1442
edbjwinslow at gmail.com
winslowmedical.com

"The only thing new in the world is the history that you don't know"
       Harry S. Truman, 33rd President of US (1945-1953)


"... it can be argued that underinvestment in assessing the past is likely
to
lead to faulty estimates and erroneous prescriptions for future action."
        Eli Ginzberg, 1997






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


HTML Version:
URL: <../attachments/20180115/b38c81d2/attachment.html> ATTACHMENT:
Name: image008.jpg Type: image/jpeg Size: 4137 bytes Desc: not available URL: <../attachments/20180115/b38c81d2/attachment.jpg> ATTACHMENT:
Name: image009.jpg Type: image/jpeg Size: 17486 bytes Desc: not available URL: <../attachments/20180115/b38c81d2/attachment-0001.jpg> ATTACHMENT:
Name: image004.jpg Type: image/jpeg Size: 716 bytes Desc: not available URL: <../attachments/20180115/b38c81d2/attachment-0002.jpg> ATTACHMENT:
Name: image006.jpg Type: image/jpeg Size: 814 bytes Desc: not available URL: <../attachments/20180115/b38c81d2/attachment-0003.jpg> ATTACHMENT:
Name: image010.png Type: image/png Size: 8433 bytes Desc: not available URL: <../attachments/20180115/b38c81d2/attachment.png> ATTACHMENT:
Name: image003.jpg Type: image/jpeg Size: 704 bytes Desc: not available URL: <../attachments/20180115/b38c81d2/attachment-0004.jpg> ATTACHMENT:
Name: image005.jpg Type: image/jpeg Size: 778 bytes Desc: not available URL: <../attachments/20180115/b38c81d2/attachment-0005.jpg> ATTACHMENT:
Name: image007.png Type: image/png Size: 6444 bytes Desc: not available URL: <../attachments/20180115/b38c81d2/attachment-0001.png>


More information about the Test mailing list