Patient Safety Tip of the Week
January 14, 2014
Diagnostic Error:
Salient Distracting Features
Salient distracting
features (SDF’s) are one mechanism by which cognitive biases may enter our diagnostic
reasoning and potentially lead to diagnostic errors. These are features in a
patient’s history that may be strongly associated with a disease or condition
and thus catch the attention of the clinician. Particularly when reasoning in a
pattern recognition mode, the clinician may be so distracted by the feature
that a cognitive bias is introduced. When that SDF is not actually relevant to
the patient’s case, a diagnostic error may ensue.
In our May 15, 2012
Patient Safety Tip of the Week “Diagnostic
Error Chapter 3” we discussed a study from the Netherlands (Mamede 2012) looked at diagnostic reasoning in medical
students and internal medicine residents. They had previously noted that
salient distracting features are a major contributor to diagnostic errors,
particularly when in the non-analytic reasoning mode. They showed that reflective
reasoning led to significantly more correct diagnoses. Interestingly, students
did not benefit from reflective reasoning. The implication is that certain
salient features may attract a physician’s attention and misdirect the
diagnostic reasoning process. Reflective reasoning may help overcome the
influence of these distracting features.
Now the same group (Mamede
2013) has further studied the role of salient distracting features in internal medicine residents analyzing
simulated cases based upon actual patients. They asked the residents to
diagnose 12 clinical cases (6 simple, 6 complex) in three different formats.
Some had an SDF inserted near the beginning of the case, some near the end of
the case, and some with no SDF. When an SDF was presented early in a complex
case there were 58% fewer accurate diagnoses than when no SDF was present. And
the accuracy of diagnoses was about 50% fewer when an SDF was presented at the
beginning rather than end of a case. The presence of an SDF also increased the
amount of times spent making a diagnosis.
An example of a
salient distracting feature in the above studies was noting early in the case
that the patient had a positive PPD and family history of remote treatment for
TB.
But I can provide a
great personal example where salient distracting features led me astray in
diagnosing myself! Several years ago I began to limp, primarily due to weakness
in my left leg. It was pretty much painless except for some low-level aching in
the mid-thigh region. The thigh was visibly smaller on the left than the right
(measuring 2 cm. smaller in circumference). My father had died from ALS (amyotrophic
lateral sclerosis). Being a neurologist, my first thoughts were, of course,
that I had ALS (as many as 10% of ALS cases may have a familial component). So
I went to one of my trusted colleagues for an EMG. I even informed him which
muscle groups he’d likely see fasciculations, since by now I was looking for
any other signs that would support a diagnosis of ALS (and almost everyone gets
fasciculations from time to time that are benign in nature). How relieved I was
when the EMG was normal. But that still left the limp unexplained. So we
scheduled some hip X-rays. Because what little pain I had was in mid-thigh and
because something in mid-thigh seemed to suddenly “catch” sometimes while
walking, I insisted we also get X-rays of the thigh as well. The X-rays did
show fairly significant hip arthritis on both sides. The thigh X-ray showed
about a 6-inch ossification of a muscle and tendon in mid-thigh. I suspected
that was likely due to a very old lacrosse injury. But it was also right where
my main symptoms were located. I did at that time get my left hip injected
(with both a local anesthetic and a steroid) as a diagnostic test. It produced
no significant improvement. Six weeks of physical therapy led to only slight
improvement. I was convinced my problem was fully related to the myositis
ossificans. My limp got progressively worse. I used to walk/jog almost 8 miles
a day. Now I could only walk about 2 miles a day with difficulty and could no
longer jog at all. If I stood with all my weight on just the left leg, I would
hear a “sproing” sound, further increasing my concern that the symptoms were
related to the myositis ossificans. The literature paints a very pessimistic
picture for treatment of myositis ossificans. Surgery is not likely to be helpful.
There are a few case reports of symptom improvement after extracorporal shock
wave therapy. Local injections with iontophoresis have also been reported to
improve symptoms in some cases. I finally tracked down an orthopedic surgeon
who had some experience with extracorporeal shock wave therapy. But when I saw
him he took one look and said “this is all coming from that hip”. Even though I
obviously understand the concept of referred pain, I still remained skeptical
since I lacked the more typical hip or groin pain seen with hip arthritis. I
could still kayak and I slogged through a trip-of-a-lifetime to Alaska where I
was able to go on most hikes because I could use hiking poles. But I finally
acquiesced to have a left total hip replacement.
My surgeon was
great. The surgery went smoothly and I was out of the hospital in less than 24
hours! My thigh pain disappeared almost immediately. My rehab consisted of
daily exercises at home. (I recall the “old days” when a total hip replacement
patient stayed in the hospital for a couple weeks!). My limp is gone. By six
weeks post-op I was cross-country skiing and by eight weeks I was downhill
skiing!
My self-misdiagnosis
obviously shows multiple types of cognitive bias that we commonly see in
diagnostic error. Anchoring, framing, early closure and confirmation
bias were all clearly operative. But very significant were the two salient
distracting features (the family history of ALS and the muscle ossification
on X-ray). Fortunately, my diagnostic error was not on one of my patients. And,
fortunately, it didn’t lead to any significant harm. But it did have a
significant impact on my lifestyle that could have been avoided if I had gone
ahead with the hip replacement a couple years earlier!
In the diagnostic
process we often need to step back. Most often when we do that we ask ourselves
“what might I be missing?”. But just as important is
to ask “what might I be giving too much weight to?” or “would I be thinking
differently if {blank} was not present?”.
Some of our prior columns on diagnostic error:
- September 28, 2010 “Diagnostic Error”
- November 29, 2011 “More
on Diagnostic Error”
- May 15, 2012 “Diagnostic
Error Chapter 3”
- May 29, 2008 “If You Do RCA’s or Design Healthcare Processes…Read Gary Klein’s Work”
- August 12, 2008 “Jerome Groopman’s “How Doctors Think”
- August 10, 2010 “It’s Not Always About The Evidence”
- January 24, 2012 “Patient
Safety in Ambulatory Care”
- October 9, 2012 “Call
for Focus on Diagnostic Errors”
- March 2013 “Diagnostic
Error in Primary Care”
- May 2013 “Scope
and Consequences of Diagnostic Errors”
- August 2013 “Clinical
Intuition”
- January 2014 “Trigger
Tools to Prevent Diagnostic Delays”
- And our review of Malcolm Gladwell’s “Blink” in our Patient Safety Library
References:
Mamede S, Splinter TAW, van Gog T, et al. Exploring the role of salient distracting clinical features in the emergence of diagnostic errors and the mechanisms through which reflection counteracts mistakes. BMJ Qual Saf 2012; 21:295-300 doi:10.1136/bmjqs-2011-000518
http://qualitysafety.bmj.com/content/21/4/295.abstract
Mamede S, van Gog T, van den Berge K, et al. Why Do Doctors Make Mistakes? A Study of the Role of Salient Distracting Clinical Features. Academic Medicine., POST AUTHOR CORRECTIONS, 25 November 2013
Print “PDF
version”
http://www.patientsafetysolutions.com/