?Datamining for pernicious anemia?

Alan Morris Alan.Morris at IMAIL.ORG
Mon Dec 9 16:06:19 UTC 2013


An iterative refinement knowledge engineering process could capture the way experts deal with the diagnostic challenge.  The challenge is to produce a reasonable protocol (not a "correct, perfect or right" protocol since these are not known to us).  The effort spent on knowledge engineering would produce a "stabilized diagnostic process" that could be further refined as data are acquired.  This wold require a distributed laboratory, and electronic database, and a refinement committee that would meet regularly to review performance and modify the protocol logic.  The extensive effort would be well-spent when compared to the cost of unnecessary variation in medicine.  However, as Robert Bell subsequently noted, I know of no readily available funding source- although I think this is clearly a route to improvement that should be taken.

Alan H. Morris, M.D.

From: <Hoffer>, "<Edward P.>", "M.D." <EHOFFER at MGH.HARVARD.EDU<mailto:EHOFFER at MGH.HARVARD.EDU>>
Reply-To: Society to Improve Diagnosis in Medicine <IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>>, "Hoffer, Edward P.,M.D." <EHOFFER at MGH.HARVARD.EDU<mailto:EHOFFER at MGH.HARVARD.EDU>>
Date: Monday, December 9, 2013 7:44 AM
To: "IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>" <IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>>
Subject: Re: [IMPROVEDX] ?Datamining for pernicious anemia?

The task is daunting but at least partially doable.  When we began putting together the DXplain database in the mid-1980's, we found that textbooks were of limited use.  Textbook descriptions of diseases tend to use "weasel words" like "often" or "may be present."  The problem is that if you ask a group of doctors to quantify "often" you will get answers ranging from 20% to 80%.  For many conditions, valid numeric data is available in published case series.  We search hard for these and use them when available.  Of course, there is always the objection that these may be biased by geographic location, by coming from a referral center, etc, but at least 100 cases from the Mayo Clinic with precise quantification of findings is a big step up from "often."
Edward Hoffer MD, MGH Lab of Computer Science

________________________________
From: robert bell [mailto:rmsbell at ESEDONA.NET]
Sent: Sunday, December 08, 2013 9:33 PM
To: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>
Subject: Re: [IMPROVEDX] ?Datamining for pernicious anemia?

FOR POSTING TO THE LIST

Thanks Janel,

You bring up the need of creating profiles for immune complex disease patients, with percentages for each symptom - when we get to computer use to better support diagnoses, we may be better to arrive at the answer more quickly.

How would we go about collecting such data. And who would do it?

And also percentages for the diagnose itself compared to all other diagnoses.

For any diagnostic computer program to be reasonably accurate it would seem that % frequencies for symptoms and diagnoses would first be necessary.

It seems that we need better data on nearly everything, if significant progress is to be made in diagnostic accuracy.

Rob Bell




On Dec 8, 2013, at 2:02 PM, Janel Hopper <janelhopper at COMCAST.NET<mailto:janelhopper at COMCAST.NET>> wrote:

Hello All,

Thank you Dr. Kohn for bringing up my story of pernicious anemia again and giving me an opportunity to clarify a few things.  I just looked into Dr. Sanders book that you mention.  In the article excerpted from the same, she says:

"This patient’s hemogram would have shown anemia and almost certainly an elevated mean cell volume."

I had intermittent iron anemia for many years that was treated by my internist. I sometimes had a subnormal hematocrit that was attributed by a very well-known teaching hospital to be due to celiac and not investigated further at that time. I was taking high-dose vitamins, which was communicated to every physician.

A close examination of the medical literature will show that pernicious anemia can be doing neurologic, gastric and other tissue damage including bone, heart, and CNS without any overt macrocytosis necessarily appearing. A query into problems with B-12 blood testing, especially in cases of pernicious anemia, will retrieve a number of results in PubMed. If anyone is interested in links or further investigation into this, I would like to share more details. Any particulars of my experience needs to maintain the privacy of all the physicians and institutions along the way that certainly all did their best to help me.

When a patient keeps insisting that there is something physicially wrong but "the screen" (electronic medical record) already marginalizes a patient as having "imaginary pain," many subsequent physicians won't think anything needs re-thinking. I am lucky in that I could go for consultations across the country. I am also lucky that the national data record was not yet aggregated at that time.  Eventually (I had enough data points of my related autoimmune conditions), that I stumbled upon pernicious anemia as a related disease and possibility. I was lucky enough to be one with IF blocking antibodies, later also confirmed with parietal cell antibodies, elevated MMA and homocysteine. This simple, specific test (IF Blocking Antibody) that only cost me $35 has been available since 1983. In spite of numerous teaching hospital workups and awareness of my other autoimmune conditions, this test was never used until I requested it. After 40 years of work-ups, this was indeed a shocking finding.

Since decision support software is not widely adopted, a retrospective analysis of patients who have say known thyroid disease along with other, unclear diagnoses such as "fibromyalgia, chronic fatigue, irritable bowel syndrome, mitral valve prolapse, intention tremor, and somatic pain disorder" (all diagnoses I previously had), this population subset could be data-mined to sample test those who have a risk for undetected pernicious anemia.

I hope that some of you may be interested in learning from my experience. I am just a lucky layperson, so please correct me if I make errors.

Janel

On Dec 7, 2013, at 10:23 PM, Kohn, Michael wrote:

In sick (i.e. symptomatic) patients, there are two kinds of diagnostic error: we can fail to identify and treat the disease that is causing the illness or we can misdiagnose and inappropriately treat a disease that isn't causing the illness.

One of the participants on this thread mentioned pernicious anemia.  The Lisa Sanders book (Chapter 9) has an example of a delayed diagnosis of that disease.  I could see missing this as an explanation for chest pain in a busy ED and recommending outpatient evaluation, but it's hard to believe that it wouldn't be picked up eventually.    We also have to worry about overdiagnosis/misdiagnosis.  See Chapter 8 on chronic Lyme Disease.

Reducing both kinds of diagnostic error is a challenge.  Computerized decision support tools integrated into the EHR may be part of the answer, but we're not there yet.

MAK


Michael A. Kohn, MD, MPP

Associate Professor

Epidemiology and Biostatistics

Emergency Physician

Mills-Peninsula Medical Center

________________________________
From: Alan Morris [Alan.Morris at IMAIL.ORG<mailto:Alan.Morris at IMAIL.ORG>]
Sent: Saturday, December 07, 2013 4:55 PM
To: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>
Subject: Re: [IMPROVEDX] Missed and Erroneous Diagnoses Common in Primary Care Visits

I propose that an error is any deviation from the intended, evidence-based, target decision.  Some errors are trivial and of little consequence.  Some carry a major risk to the patient. In an unusually tightly controlled ICU at the Hebrew University Hospital in Jerusalem, Gopher et al. reported about 174 clinician interactions/patient/day.  99% of these clinician interactions were performed correctly by RNs and MDs. (It is the rare, if any, clinical unit that can aspire to a 99% correct performance rate.)  However, the 1% error rate led to a major threat to life or limb per patient every other day, on average!  Most people think that technical rock climbing is a dangerous activity.  However, few climbers would ever tie onto a rope and climb, if they thought they would be subjected to a major threat to life or limb every other day because of error.

Surprising low clinician error rates associated with almost unachievable correct performance, can lead to unacceptable risks to patients.

Extrapolating these ICU data to the outpatient setting is difficult.  Nevertheless, I believe we should pay attention to all errors, and categorize them with respect to patient risk.

Have  a nice day.

Alan H. Morris, M.D.
Professor of Medicine
Adjunct Prof. of Medical Informatics
University of Utah

Director of Research
Director Urban Central Region Blood Gas and Pulmonary Laboratories
Pulmonary/Critical Care Division
Sorenson Heart & Lung Center - 6th Floor
Intermountain Medical Center
5121 South Cottonwood Street
Murray, Utah  84157-7000, USA

Office Phone: 801-507-4603
Mobile Phone: 801-718-1283
Fax: 801-507-4699
e-mail: alan.morris at imail.org<mailto:alan.morris at imail.org>
e-mail: alanhmorris at gmail.com<mailto:alanhmorris at gmail.com>




________________________________

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/



________________________________

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/



________________________________

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/

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.

________________________________

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/20131209/853b77c9/attachment.html>


More information about the Test mailing list