NYTimes: Her Various Symptoms Seemed Unrelated. Then One Doctor Put It All Together.

Follansbee, William follansbeewp at UPMC.EDU
Wed Feb 21 19:29:13 UTC 2018


Art,

I agree completely with the very thoughtful approach that you describe. As we all know, there is no one solution to diagnostic error. It will take initiatives from many different approaches to make real progress. Decision support tools clearly have an important role to play. When I came to the University of Pittsburgh in 1980, I heard many bold predictions from leaders in the field who happened to be there at the time about the great impact that computer assisted diagnosis would have, an impact that felt to be just on the horizon. But that impact has been very slow to come. It doesn’t mean that the concept is wrong but that it is an extraordinarily difficult challenge. I think you are on exactly the right track trying to figure out how to integrate augmented reasoning into existing workflow, because the reality is with today’s seriously overburdened physicians, if decision support it is not seamless integrated in to what they are already doing, they are very unlikely to use it in other than isolated cases. That is the lesson of the last 40 years. I am glad that extremely bright people such as yourself continue to work on it.

Best,
Bill


William P. Follansbee, M.D., FACC, FACP, FASNC, FAHA
The Master Clinician Professor of Cardiovascular Medicine
Director, The UPMC Clinical Center for Medical Decision Making
Suite A429 UPMC Presbyterian
200 Lothrop Street
Pittsburgh, PA 15213
Phone: 412-647-3437
Fax: 412-647-3873
Email: follansbeewp at upmc.edu<mailto:follansbeewp at upmc.edu>

[cid:image001.png at 01D1AF95.1368ACB0]

This email may contain confidential information of the sending organization. Any unauthorized or improper disclosure, copying, distribution, or use of the contents of this email and attached document(s) is prohibited. The information contained in this email and attached document(s) is intended only for the personal and confidential use of the recipient(s) named above. If you have received this communication in error, please notify the sender immediately by email and delete the original email and attached document(s).





From: Art Papier [mailto:apapier at visualdx.com]
Sent: Monday, February 19, 2018 1:41 PM
To: Follansbee, William <follansbeewp at upmc.edu>; 'Society to Improve Diagnosis in Medicine' <IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>
Subject: RE: [IMPROVEDX] NYTimes: Her Various Symptoms Seemed Unrelated. Then One Doctor Put It All Together.

Thanks Bill for continuing the conversation.  I agree that the logical competence implicit in the idea that “bilateral cellulitis” is a diagnostic oxymoron should be taught,  and we should as you describe be absolutely modeling the pathophysiologic thinking process for our students and residents.  That is mandatory.

With that said when our team sits at a table discussing and visioning the near future we are thinking in terms of how we are going to successfully augment the clinician brain.  We work from the assumption that care is extremely fast paced, highly variable and often delivered by green and junior MD’s, PA’s and NP’s and overconfident senior clinicians.  Our resident’s, and soon to graduate medical students will practice in an increasingly more complex and time critical, demanding environment.  By augmentation I mean we have to do more with checklists and decision support, to develop better tools and to teach how evidence is used elegantly at the point of care.  We need to innovate and create tools as augmented intelligence, not as artificial intelligence.  There is a way to build skills with point of care CDS, not replacing thinking, but aiding it.   We bring high reliability to aviation by augmenting pilots skills with checklists, instruments and software based systems.  We have yet to do the same consistently in medicine as had been done in aviation.  Probably because medicine has a lot more unknowns and ambiguity then aviation.
We have customers of our decision support system who are asking us to embed a “dot phrase” in Epic  (and similar techniques in other EHR’s) around cellulitis. The physicians we are working with want to embed a “flag or alert” when a clinician orders a patient admission or IV antibitoics for cellulitis.  If a thorough physical exam has been performed and documented, it is easy to imagine intelligence in the EHR prompting a bilateral leg redness differential within the EHR when the CDS detects bilateral redness has been documented and the physician has ordered an admission or IV antibiotics.  The newly widely used HL7 FHIR  standard for CDS interoperability in the EHR with “hooks” could be used to accomplish this.   Given the potential patient harm, I believe bilateral leg redness diagnosed as cellulitis deserves a differential diagnosis insertion in the EHR frame so the clinician is alerted and forced to review the differential.   Obviously that is a n=1 belief.  I believe more and more hospitals have clinical decision support governance teams which would handle a decision like this to make sure such a “rule” is implemented with thoughtfulness.  Hospital IT governance committees will hopefully start to include diagnosis in their thinking and work.  Thanks again for the conversation!
Regards,
Art
CEO VisualDx
Associate Professor of Dermatology and Medical Informatics
University of Rochester

From: Follansbee, William [mailto:follansbeewp at upmc.edu]
Sent: Monday, February 19, 2018 12:15 PM
To: Society to Improve Diagnosis in Medicine <IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>>; Art Papier <apapier at VISUALDX.COM<mailto:apapier at VISUALDX.COM>>
Subject: RE: [IMPROVEDX] NYTimes: Her Various Symptoms Seemed Unrelated. Then One Doctor Put It All Together.

Art,

I agree with your thoughtful comments. I would also add, however, that for a disease like cellulitis, which I agree is frequently over diagnosed and treated unnecessarily, the answer is not going to be in decision support tools. Clinicians are just not going to consult them for such a common diagnosis. It is also to teach them how to be a little more thoughtful and analytic in their bedside decision making.   We teach trainees to use small groups of common sense but not uncommonly overlooked questions at appropriate times in the diagnostic process in a systematic fashion. In this context, one question we emphasize that they should ask themselves when considering a diagnosis is, “is there any discordant data?” Cellulitis is rarely bilateral yet many patients admitted and treated for apparent cellulitis have red and swollen legs bilaterally, ie discordant findings.  If their illness script for cellulitis includes bilateral disease, then that is a knowledge problem that also has to be addressed.

Best,
Bill


William P. Follansbee, M.D., FACC, FACP, FASNC
The Master Clinician Professor of Cardiovascular Medicine
Director, The UPMC Clinical Center for Medical Decision Making
Suite A429 UPMC Presbyterian
200 Lothrop Street
Pittsburgh, PA 15213
Phone: 412-647-3437
Fax: 412-647-3873
Email: follansbeewp at upmc.edu<mailto:follansbeewp at upmc.edu>

[cid:image001.png at 01D1AF95.1368ACB0]

This email may contain confidential information of the sending organization. Any unauthorized or improper disclosure, copying, distribution, or use of the contents of this email and attached document(s) is prohibited. The information contained in this email and attached document(s) is intended only for the personal and confidential use of the recipient(s) named above. If you have received this communication in error, please notify the sender immediately by email and delete the original email and attached document(s).



From: Art Papier [mailto:apapier at VISUALDX.COM]
Sent: Monday, February 19, 2018 11:17 AM
To: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>
Subject: Re: [IMPROVEDX] NYTimes: Her Various Symptoms Seemed Unrelated. Then One Doctor Put It All Together.

Likewise VisualDx had Schnitzler’s at the top of the differential, but as much as I agree that all physicians need to understand and use point of care diagnostic decision support, we should recognize that relatively rare diseases like Schnitzler are uncommon and relatively easy for decision support to “pick up”.   The real need is to handle the cases when clinicians do not know they need help, but do need help.  How do you know what you don’t know?  Uncommon diseases are uncommon, and therefore variants of common are much more common that rare diseases.   Our real challenge in decision support is to provide tools that also provide useful and valuable content around the common, and more particularly with the variants of the common so clinicians have decision support top of mind.  80-20 rule:  If 80% of diagnoses are common, then it is reasonable to assume that variants of the 80% dwarf the super rare diseases in number.  It is also safe to assume that clinicians who are in a constant rush, and bogged down by mind-numbing EHR charting exercises, will question the efficiency of using these tools.  We are focused on variation in disease presentation in our work with the goal of expanding the use of decision support beyond use for seemingly rare presentations.  We belive that the memory based training and care delivery system creates self-fulfilling prophecies where clinicians ask questions around the “classic presentation disease” scripts they memorize, but do not know the questions to ask around the related variants.  As an example,  over 100,000 people are admitted to hospitals each year for cellulitis when they do not have cellulitits.  This is a boring “story” for decision because cellulitis is common, but there is so much harm happening just from error around this single diagnosis.  How do we bend this curve and reduce unnecessary admissions while recognizing all true positives?   By focusing on commn diseases and their variants we can expand the use of decision support.

Thanks to Lisa for another wonderfully written great case and prompting discussion at SIDM !
Art

Art Papier MD
CEO VisualDx
Associate Professor of Dermatology and Medical Informatics
University of Rochester
From: Edward Hoffer [mailto:ehoffer at GMAIL.COM]
Sent: Sunday, February 18, 2018 6:40 PM
To: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>
Subject: Re: [IMPROVEDX] NYTimes: Her Various Symptoms Seemed Unrelated. Then One Doctor Put It All Together.

This story makes a very good case for the use of computer-based diagnostic decision support systems. I entered the findings into the one with which I work, DXplain, and Schnitzler's came in ranked #1 I did not try Isabel, but would not be surprised if it also had the correct diagnosis near the top. Much easier than spending the reported "hours" in PubMed that the hero expended to arrive at the correct diagnosis.
Ed
Edward P Hoffer MD, FACC, FACP

On Sun, Feb 18, 2018 at 9:16 AM, Joe Graedon <jgraedon at gmail.com<mailto:jgraedon at gmail.com>> wrote:
https://www.nytimes.com/2018/02/14/magazine/her-various-symptoms-seemed-unrelated-then-one-doctor-put-it-all-together.html?smprod=nytcore-ipad&smid=nytcore-ipad-share

Chills, sweats, hives, achey bones — the older woman was sick for years before someone figured out the unusual disease that ailed her.


Sent from my iPad
To unsubscribe from the IMPROVEDX:
mail to:IMPROVEDX-SIGNOFF-REQUEST at LIST.IMPROVEDIAGNOSIS.ORG<mailto:to%3AIMPROVEDX-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

http://LIST.IMPROVEDIAGNOSIS.ORG/ (with your password)

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

Moderator: David Meyers, Board Member, Society to Improve 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>
</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:David Meyers, Board Member, Society for Improving Diagnosis in Medicine

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

________________________________

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:David Meyers, Board Member, Society for Improving Diagnosis in Medicine

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

To unsubscribe from the IMPROVEDX:
mail to:IMPROVEDX-SIGNOFF-REQUEST at LIST.IMPROVEDIAGNOSIS.ORG
or click the following link: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG

Address messages to: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG

For additional information and subscription commands, visit:
http://www.lsoft.com/resources/faq.asp#4A

http://LIST.IMPROVEDIAGNOSIS.ORG/ (with your password)

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

Moderator: David Meyers, Board Member, Society to Improve 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>
</p>

HTML Version:
URL: <../attachments/20180221/ef519590/attachment.html> ATTACHMENT:
Name: image001.gif Type: image/gif Size: 1410 bytes Desc: image001.gif URL: <../attachments/20180221/ef519590/attachment.gif>


More information about the Test mailing list