Error discovery

Bruno, Michael mbruno at HMC.PSU.EDU
Mon Jan 19 15:51:07 UTC 2015

Dear SIDM List Serve Colleagues,

I’m just catching up after a week away – there have been some wonderful, thought-provoking posts on this ListServe over the past 10 days or so, and I have appreciated the discussion very much.  This particular thread, started by Bob Bell back in November, is definitely worth revisiting.  Karen (below) raises an important point about how we even detect errors, which raises the question of how many we are not detecting.  If we are only detecting the ones that cause harm, for example, we will miss most of them, and it is unlikely in that scenario that we will ever learn enough about diagnostic error to make much of a difference in error reduction or prevention.  We clearly need to find a more sensitive, reliable way to discover at least MOST of our errors, so that we can begin to think about them in a rational and quantitative way.   I am currently reading a fantastic book, “How Not to be Wrong,” written by mathematician Jordan Ellenberg (see  He gives many examples of how even rigorous analysis of partial or incomplete data – such as if we detect only a certain type or severity of errors – can lead to false conclusions and wasted effort/lost opportunity.

I think what Karen’s message highlights is the need for us to develop what have become known as “trigger tools,” i.e., ways of detecting errors or (hopefully) even latent errors.  Getting a more accurate, comprehensive understanding of the epidemiology for error as it exists in medicine will help us in so many ways.  This may be the most significant challenge for research in this area in the years ahead.

All the best,

[Description: Description: Description: \\\files\Staff\M\mbruno\Signature2.gif.gif]
Michael A. Bruno, M.D., F.A.C.R.
Professor of Radiology & Medicine
Director of Quality Services & Patient Safety
The Milton S. Hershey Medical Center
Penn State College of Medicine
500 University Drive, Mail Code H-066
Hershey, PA  17033

Phone: (717) 531-8703
Fax:      (717) 531-5596

[cid:image002.png at 01CF4E6D.0D6442B0]

From: Karen Cosby [mailto:kcosby40 at GMAIL.COM]
Sent: Thursday, January 15, 2015 11:21 AM
Subject: Re: [IMPROVEDX] Fwd: [IMPROVEDX] Commonest error in medicine? kindly post to list

This is not that straightforward a question.  It comes down to how you screen for errors.  I reviewed all our M&M cases over the last 20 years, and I have access to other quality and oversight cases.  But those cases are subject to a selection bias…..mostly cases that provoked a response because they were "interesting" or had poor outcomes. There are certainly many more errors that we don't know about or that were less interesting.  I think we do better to recognize that error is everywhere.  You can choose to look at instances where delays in diagnosis or misdiagnoses most likely have greatest impact on outcome, or just recognize that even simple things that seem mundane impact care, and those deserve attention too.  If you need a list, most people will come up with diagnoses that are unforgiving of delay and need rapid recognition to optimize outcome.  The usual list…….AAA, aortic dissection, MI, PE, ectopics, mesenteric ischemia, CVA, sepsis, aortic occlusions/ischemic legs, tamponade…..Add to this items from malpractice that includes missed fractures, wound complications.

On Mon, Dec 1, 2014 at 7:12 PM, Robert Bell <0000000296e45ec4-dmarc-request at<mailto:0000000296e45ec4-dmarc-request at>> wrote:

Thanks John - good thoughts thoughts.

How do we get a top 10 for the ER. ? Return to ER data if available.


Sent from my iPad

Begin forwarded message:
From: "Ely, John" <john-ely at<mailto:john-ely at>>
Date: December 1, 2014 8:41:48 AM MST
To: Robert Bell <rmsbell200 at<mailto:rmsbell200 at>>
Subject: RE: [IMPROVEDX] Commonest error in medicine?
Fortunately there is a MeSH term (“Diagnostic Errors”) that should help with a Pubmed search.  Gordy Schiff’s study of 583 errors and Mark Graber’s study of 100 errors should provide an entry into the literature.  Also lots of autopsy studies summarized by Shojania (Shojania KG, Burton EC, McDonald KM, Goldman L. Changes in rates of autopsy-detected diagnostic errors over time: a systematic review. JAMA. 2003 Jun 4;289(21):2849-56.).  Maybe we could start with a top-10 list.  Criteria might include
“Don’t miss” diagnoses
Diagnoses amenable to quick easy rule-out tests, or historical points that rule them out or physical exam points that rule them out
Most common presenting symptoms for each of the 10 diagnoses.

From: Robert Bell [mailto:rmsbell200 at<mailto:rmsbell200 at>]
Sent: Sunday, November 30, 2014 4:43 PM
To: Ely, John; Society to Improve Diagnosis in Medicine
Subject: Re: [IMPROVEDX] Commonest error in medicine?

Great John,
 A literature search would be good. I do not have access. Would we be interested in, "commonly missed dignoses with serious consequences seen In the ER?" Readmission data should provide some information. Are there studies showing return to ER data?
You bring up good points. Perhaps focussing on the the Emergency Room would provide us the best benefit. This would also bridge in-hospital and private practice patients. We could expand to those in time.
The other thing is that the missed diagnoses, will take us back to symptoms. So for pulmonary embolism we would be starting with shortness of breath?
 Rob Bell

Sent from my iPad

On Nov 30, 2014, at 1:49 PM, "Ely, John" <john-ely at<mailto:john-ely at>> wrote:
I would be happy to contribute to this project.  I suppose the first step would be a literature review to see what’s been done in the past.  By “hospital diagnostic errors” do you mean errors made on inpatients or errors made with ER patients who are sent home but admitted later or something else?

From: Robert Bell [mailto:rmsbell200 at]
Sent: Saturday, November 29, 2014 8:13 PM
To: Society to Improve Diagnosis in Medicine; Ely, John
Subject: Re: [IMPROVEDX] Commonest error in medicine?

John and everyone - sorry I have not replied sooner. Have just been through a busy period. My general thoughts at this time are to tackle common errors in diagnosis in standard format, similar say a drug monograph in content. Then to fill in the "chapters" to the best of of our abilityq with those on line. Then to ask for comment on the draft from one or two experts in the specialty being discussed. Then slowly deal with other diagnostic errors until we have a book for publication. ? Title: Common hospital diagnostic errors - supported and published by the Society. Something like that.
To start, what is the purpose of the book? What would/should the headings in each monograph be? Could we use as our test error, incorrectly and over diagnosed pulmonary embolism?
Is this all feasible, so we all contribute to something positive, even if never published, it will get our thoughts organised. And maybe promote other ideas and lines of research within our small community.
Further thoughts welcome.

Rob Bell

Sent from my iPad

On Nov 25, 2014, at 6:23 AM, "Ely, John" <john-ely at UIOWA.EDU<mailto:john-ely at UIOWA.EDU>> wrote:
I think this has potential.  If we wait for hard data, we’ll wait a long time.  We would want to do it systematically with something like a Delphi method.  The most memorable errors might not be the most common or even the most important to try to reduce.  We should consider who to include (primary care, pediatrics, OB-GYN, surgeons, radiologists, pathologists).  I’m guessing misdiagnosis of the common cold is far more common than PE (i.e., calling the common cold bacterial sinusitis).  Maybe we should also collect diagnoses that are never misdiagnosed (sometimes you can learn from the opposite situation).  We should anticipate criticisms – why we should not do this (e.g., it would be based on opinion, not evidence; lots of potential for selection bias; what would an angry grant reviewer say).  We would want to have inclusion and exclusion criteria.  In our checklist study we found lots of potential for misclassification of diagnostic errors, which we defined as an important difference between the initial diagnosis and the one-month follow-up diagnosis.  (Important defined as patient management would have been different).  For example,
1. Evolution of disease:   The diagnosis of viral upper respiratory infection (common cold) may have been correct at the time of the initial visit, but the patient developed bacterial sinusitis after this visit.
2. Nomenclature discrepancy:  The initial diagnosis was “viral illness.”  The final diagnosis was “bronchitis” without designating the etiology (which was likely viral).
3. Diagnostic evaluation in process:  The initial diagnosis was nausea of unknown etiology.  Subsequent testing, which was ordered at the initial visit, revealed diabetic gastroparesis.
4. Vague initial impression but specific initial plan:  An initial impression of “abdominal pain” accompanied by a prescription for a proton-pump inhibitor, followed by a final diagnosis of “peptic gastritis.”
5. One patient with two diseases:  The chest pain in the emergency room really was caused by reflux esophagitis even though the patient died of a myocardial infarction at home 6 hours later.
6. Discrepancy in disease severity:  A patient with “dehydration” initially treated and released from ER subsequently required admission for the same diagnosis.
In all these cases, the initial diagnosis differed from the one-month follow-up visit, but they were not really “diagnostic errors” the way we usually think of them.
John Ely

From: robert bell [mailto:rmsbell at ESEDONA.NET]
Sent: Monday, November 24, 2014 6:41 PM
Subject: [IMPROVEDX] Commonest error in medicine?

A thought.
Do we know what is the most common serious error in diagnosis in medicine in the USA?
If not can we guess at what is the commonest error in diagnosis? And then perhaps the top five.
Then could we discuss intensely on line and come to conclusions as to how we could lower that rate?
When we have come up with a list of suggestions for that one error we move on to Number 2  on what we think is the list of commonest serious errors in Medicine and how to reduce their incidence.
 This way we would be doing something positive rather than just talking. Not that talking is important.
 My contribution for serious No. 1 is Pulmonary embolism.
 This way we could just on line make a seripous contribution to medicine particularly if someone would write up our discussions.
 A thought

Robert M. Bell, M.D., Ph.C.



To unsubscribe from IMPROVEDX: click the following link:

Visit the searchable archives or adjust your subscription at:

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

To learn more about SIDM visit:

Notice: This UI Health Care e-mail (including attachments) is covered by the Electronic Communications Privacy Act, 18 U.S.C. 2510-2521, is confidential and may be legally privileged.  If you are not the intended recipient, you are hereby notified that any retention, dissemination, distribution, or copying of this communication is strictly prohibited.  Please reply to the sender that you have received the message in error, then delete it.  Thank you.

To unsubscribe from the IMPROVEDX:
or click the following link: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG


For additional information and subscription commands, visit:

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

Visit the searchable archives or adjust your subscription at:

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

To unsubscribe from the IMPROVEDX list, click the following link:<br>
<a href="" target="_blank"></a>

HTML Version:
URL: <../attachments/20150119/db089b73/attachment.html> ATTACHMENT:
Name: image001.png Type: image/png Size: 2137 bytes Desc: image001.png URL: <../attachments/20150119/db089b73/attachment.png> ATTACHMENT:
Name: image002.png Type: image/png Size: 13674 bytes Desc: image002.png URL: <../attachments/20150119/db089b73/attachment-0001.png>

More information about the Test mailing list