Introducing Dx Biases and too much information ... studies needed?

Stefanie Lee stefanieylee at GMAIL.COM
Wed Aug 14 02:18:08 UTC 2013

I agree that more information is better, and that sometimes we cannot
leave reviewing clinical history until the end, as it is required to
protocol the study correctly in the first place.

As a radiologist, I try to reduce the risk of bias by focusing on the
building blocks of that information (e.g. RLQ pain, 24 hours, WBC 15
-- searching the electronic medical record if necessary), rather than
information that has already been interpreted (r/o appendicitis). I
have heard of many others who also try to avoid reviewing past reports
until they have looked at the images first.

In the setting where only limited clinical information has been
provided (r/o dissection), I will try to work backwards and consider
what the clinical presentation may have been (chest/back pain) in
order to expand my search pattern to look for other differential
diagnoses (thoracic compression fracture).

At the end, I have found 'closing the loop' (discussing the case
directly with the referring physician) extremely valuable, especially
for more complex or less clear-cut cases. Together we can discuss how
well the imaging findings support various diagnoses, as well as the
pretest likelihood of each possibility, and I can ask for additional
clinical information to confirm or refute any other hypotheses I may
have generated based on the findings.

Even if the referrer already has a favoured diagnosis, I have found
people who request imaging studies are quite open to discussion of
various possibilities based on the results, and hopefully this allows
us to provide more useful interpretations to guide management (rather
than the dreaded hedge, or laundry list - as many findings are not
pathognomic and could conceivably fit more than one scenario, it takes
clinical information to narrow down the possibilities to what is
likely going on in this patient).

Stefanie Lee
Abdominal Imaging Fellow
University of Toronto

On 13 August 2013 19:42, Perez-Ordonez, Bayardo
<Bayardo.Perez-Ordonez at> wrote:
> I am surgical pathologist with administrative responsabilities in a large medical centres in Canada. As general principle, the more information we have, the more accurate or helpful our reports will be. The information needed by a liver pathologist dealing with a core biopsy of a patient with acute liver failure are quite different from those needed by a head and neck pathologist dealing with biopsy of  a tongue mass.  Some information is useful, some might represent "red herrings" but knowing what is useful requires a combination of knowledge and experience which is hard to compress in a forum like this. Clinical information influences even how specimens are processed. We sample differently a resected femoral head fracture in a patient with no history of malignacy from a femoral head in a patient with radiologic evidence of a pathologic fracture. Far too often we even lack basic clinical information and diagnostic pathology is a poor and expensive screening procedure. More information is better!
> Bayardo Perez-Ordonez, MD
> ----- Original Message -----
> From: Lorri Zipperer [mailto:Lorri at ZPM1.COM]
> Sent: Tuesday, August 13, 2013 07:16 PM
> Subject: [IMPROVEDX] Introducing Dx Biases and too much information ... studies needed?
> Forwarded by moderator
> From: Robert Bell [mailto:rmsbell at]
> Sent: Tuesday, August 13, 2013 4:42 PM
> Could it be that histologists and radiologists do not want too much
> information for fear of a diagnostic bias being introduced?
> After all one presumes accuracy is pretty good now. Would more information
> improve or compromise accuracy? Studies needed?
> Rob Bell
> Sent from my iPad
> On Aug 13, 2013, at 2:45 PM, James Oldham
> <James.Oldham at SESIAHS.HEALTH.NSW.GOV.AU> wrote:
>> Dear Mark
> Great point. So we may need to define the diagnostic process between path
> referral and path interpretation, which may be different for many common
> path requests. (Is is just abnormal histology and imaging that presents this
> too many alternatives problem?)
> Clinical situation (differential would help) -> path request (how can it
> help?)
>> Great discussion
>> Best wishes
>> James
>> James Oldham
>> Chief Psychiatrist Mental Health Services ISLHN & Clinical Associate
>> Professor University of Wollongong
>> -----Original Message-----
>> From: Graber, Mark [mailto:Mark.Graber at VA.GOV]
>> Sent: Wednesday, 14 August 2013 6:51 AM
>> Subject: Re: [IMPROVEDX] Errors in Oncology Pathologies
>> Thanks Jena for corroborating the value of Peggy's advice. I've heard very
> similar comments from my Radiology colleagues, that supplying clinical
> information improves their ability to make the right call.
>> I'm worried though that this may sometimes bias the review and 'frame' the
> case prematurely.  These kind of framing effects are SO common elsewhere, as
> illustrated by the Rory Staunton case (His pediatrician communicated to the
> ER that she was sending over a boy with gastroenteritis, and no surprise
> that was the ER diagnosis too, missing his sepsis).  So if I send you a
> lymph node saying 'Pt with weight loss, night sweats, lymphadenopathy,
> suspected lymphoma', aren't you going to be biased towards a diagnosis of
> lymphoma by that?
>> Maybe the better way to frame this is to ask your advice:  How can
> clinicians best provide clinical information WITHOUT inducing undue bias in
> the subsequent pathologist's (or radiologist's) diagnosis?
>> Mark
>> Mark L Graber, MD FACP
>> Senior Fellow, RTI International
>> Professor Emeritus, SUNY Stony Brook School of Medicine Founder and
>> President, Society to Improve Diagnosis in Medicine
>> Phone:   919 990-8497

More information about the Test mailing list