New Study of Dx Error in the Emergency Dept

Vipindas Chengat syncopesystem at GMAIL.COM
Thu Apr 21 17:08:43 UTC 2016


Dear Dr. Corcoran,

   We share the same dream. However, there are several practical problems
with many of these approaches.

1. Probabilistic approach: A precise calculation of probability is an
incredibly difficult task. Pre-test probability or base rate for various
conditions vary significantly among hospitals. So, data will be imperfect
to begin with.  Also, we will not be able to diagnose a rare condition for
the very first time in a hospital if we just go by previous data alone.

2. Big data approach : Data in EHRs are not adequately structured for such
a research project. There is a lot of noise in the data. Considering the
fact that 15% of all diagnoses are either delayed, missed or wrong, it is
fair to assume that the practice of Medicine has been imperfect. Capturing
those data might not solve the problems.

3. Evidence based approach : There are a few things to consider here.
Quality and methodological rigor used in the studies. Do we think that all
the studies done in this field are of the highest quality? Second point is,
whether a study can consider all the variables that could lead to a
diagnosis. Probably not.

4. CPC method: Agree that analytical methods do perform better than
intuitive approach. But, the deliberate and structured way of clinical
reasoning is slow and effortful. Since 80-85% of our daily cases are very
common presentations, can we reasonably assume that every practitioner will
be willing to adopt such a method for daily practice?

I would like to introduce a new approach -
Systematic use of technology to aid critical thinking in diagnosis. The
goal is to build optimized diagnostic algorithms based on expert feedback
and big data analytics.

I have been conducting research for the last 3 years using this particular
approach.

Critical thinking is defined as the careful application of reason in the
determination of whether a claim is true. We know that physicians vary
considerably among themselves in reaching certain conclusions. To prevent
cognitive errors, it is important to learn how does physicians reach in
certain conclusions.
It is also important to study if we can measure the amount of variability
among clinicians and then try to understand the reason behind it.


As we know that the validity of clinical reasoning depends on accuracy of
premises and its logical relation to the conclusion. This can be measured
in a systematic manner.

The interface that we use has the ability to provide differential diagnoses
based on any inputs and list them in the order of relevance. It generates
further questions to narrow them down. Based on expert feedback or
analytics from a big database, the data can be continuously optimized. It
gives reasoning for each of the decisions that it makes; in an inductive
and deductive format and engages with users to improve results.


With our limited experience, we have observed the following issues at each
steps in clinical diagnosis:

   1. *Data acquisition, formulation of initial hypotheses : *We see
   significant variability among medical students and residents. A mindless
   capture of data is not very helpful in the formulation of initial
   hypothesis. Computers can help by providing dynamic checklists.
   2. *Problem representation, semantic qualifiers: *To certain level, it
   is a knowledge based problem. Computer algorithms are useful in improving
   this aspect.
   3. *Comparison with illness scripts:* This has been a unique problem.
   This vary among physicians considerably. It is more of a data problem from
   a software perspective.
   4. *Reasoning and refining using Bayesian inference: *This varies
   significantly among clinicians based on expertise. For our computer
   algorithms, it took more than half a million lines of code to simulate the
   logic used by physicians. Big data approach might be effective if used on
   top of logic algorithms here. Our interface provides deductive and
   inductive arguments for each diagnosis that it has made. It allows users to
   interact with such heuristics and optimize and refine them.


Please contact me if you would like to learn more about our research.


Appfully Yours (Android
<https://play.google.com/store/apps/details?id=com.physiciancognition.xebrapro>
 | IOS <https://itunes.apple.com/us/app/xebra-pro/id1051676634>)

https://youtu.be/D_ZZPYbA6Q8

*Vipindas Chengat, MD FACP**  |  *Chairman, Physician Cognition, Inc.
  —————————————————————————————————
  Mobile: +1 (773) 575-3550
  Email: Vipin at PhysicianCognition.Com <Vipin at PhysicianCognition.com>
  Website: PhysicianCognition.Com <http://physiciancognition.com/>



On Thu, Apr 21, 2016 at 8:12 AM, DR WILLIAM CORCORAN <
williamcorcoran at sbcglobal.net> wrote:

>
> Look forward to the day when all data from a patient’s medical,
> occupational, locational, genetic, recreational, and family history is fed
> into a computer that already contains the integrated body of knowledge and
> the computer will render a probabilistic diagnosis with error and
> confidence and will suggest additional tests and other data to reduce the
> error and increase the confidence.
>
> The computer could also suggest interventions and their cost-benefit
> predictions.
>
> The physician’s added value would include supervising input data quality,
> assessing computer output, engaging the patient, facilitating
> interventions, follow-up, etc.
>
> Take care,
>
> Bill Corcoran
>
>
> William  R. Corcoran, Ph.D., P.E.
> 21 Broadleaf Circle
> Windsor, CT 06095-1634
> 860-285-8779
> William.R.Corcoran at 1959.USNA.com
> http://www.linkedin.com/in/williamcorcoranphdpe
> https://www.box.com/shared/kfxg1lt9dh
>
>
>
> On Thursday, April 21, 2016 7:46 AM, Jason Maude <
> jason.maude at ISABELHEALTHCARE.COM> wrote:
>
>
>
> Computers are actually very good at reminding clinicians of diseases with
> atypical presentations as they carry around in their ‘memory’ many more
> presentations than possible in the human memory coupled with instant recall
> 24/7.
>
> Regards
> Jason
>
> Jason Maude
> Founder and CEO Isabel Healthcare
>
>
> From: Mark Gusack <gusackm at COMCAST.NET>
> Reply-To: Society to Improve Diagnosis in Medicine <
> IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>, Mark Gusack <gusackm at COMCAST.NET>
> Date: Wednesday, 20 April 2016 02:48
> To: "IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG" <
> IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>
> Subject: Re: [IMPROVEDX] New Study of Dx Error in the Emergency Dept
>
> Good Evening Peggy:
>
> I agree.  In fact, isn’t it our job as physicians to be on the lookout for
> ‘atypical’ presentations of diseases?  After all, if it weren’t for these
> ‘atypical’ presentations then diagnostic criteria would be so easy to apply
> that an eighth grader could do it.  In fact, a computer could do it…
>
> Hoping to see you at the LA DEM conference.  I plan to submit a number of
> new posters that may be of interest to you.
>
> Mark Gusack
>
> *From:* Peggy Zuckerman [mailto:peggyzuckerman at GMAIL.COM
> <peggyzuckerman at GMAIL.COM>]
> *Sent:* Tuesday, April 19, 2016 5:57 PM
> *To:* IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG
> *Subject:* Re: [IMPROVEDX] New Study of Dx Error in the Emergency Dept
>
> Surprises me to see that 'atypical' presentations are thought to be
> non-remedial.  Why not expand the definition of what is typical, though
> rarer?
>
> The same kind of thinking that does not permit the 'atypical' presentation
> to be reviewed properly is where the "classic symptoms" are missing.
> Example; in renal cell carcinoma, the 'classic' symptoms are present in
> fewer than 10% of cases.
>
> Peggy Z
>
> Peggy Zuckerman
> www.peggyRCC.com <http://www.peggyrcc.com/>
>
> On Tue, Apr 19, 2016 at 2:19 PM, Ruth Ryan <ruthryan at cox.net> wrote:
>
> A new study by our own peeps.
>
> This is an ED study of 214 diagnostic errors involving most often the most
> common diagnoses (sepsis, MI, fractures, vascular events) related to the
> usual suspects of cognitive error and systems problems like high workload.
> Three quarters involved multiple factors.
>
> Nearly a third were due to patient factors, and a similar number to
> “atypical” presentations (when will we find a better term for something
> that happens so often?).
>
> Citation is: Okafor N, Payne VL, Singh H et al. Using voluntary reports
> from physicians to learn from diagnostic errors in emergency medicine.
> Emerg Med J. 2016 Apr;33(4):245-52. doi: 10.1136/emermed-2014-204604. Epub
> 2015 Nov 3.
>
> Link is:
> http://www.ncbi.nlm.nih.gov/pubmed/?term=Using+voluntary+reports+from+physicians+to+learn+from+diagnostic+errors+in+emergency+medicine
>
>
> *Ruth*
>
> Ruth Ryan RN, BSN, MSW, CPHRM
> Medical writer
> Risk management/patient safety
> Continuing medical education
> Telephone (504) 256-8797
> Email ruthryan at cox.net
> [image: canstockphoto14944494_revised]
>
>
>
> ------------------------------
>
>
> 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/
>
> ------------------------------
>
>
> 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/
>
>
>
> ------------------------------
>
>
> 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/
>






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


HTML Version:
URL: <../attachments/20160421/ba72a4f8/attachment.html> ATTACHMENT:
Name: PhysicianCognitionLogo_email.png Type: image/png Size: 3718 bytes Desc: not available URL: <../attachments/20160421/ba72a4f8/attachment.png> ATTACHMENT:
Name: image001.jpg Type: image/jpeg Size: 3997 bytes Desc: not available URL: <../attachments/20160421/ba72a4f8/attachment.jpg>


More information about the Test mailing list