Discussion Board

Expand all | Collapse all

Should stroke be ruled out before the diagnosis conversion is made?

  • 1.  Should stroke be ruled out before the diagnosis conversion is made?

    Posted 07-12-2020 18:54

    I am a retired (abdominal) radiologist, still working as a patient safety officer.

    At the moment we are performing a RCA of a patient, 30 years-old woman,  who was misdiagnosed as having  a conversion disorder but in the end appeared to have an ischemic stroke.
    She presented with acute-onset complaints, which were considered atypical.  
    The diagnosis was made by the paramedic on the ambulance and established by the emergency physicians and more than 6 hours later (by phone) by the neurologist).

    As a radiologist I asked the EP and the neurologist: why didn't you request a CT?

    They replied that the distinction between a true stroke and conversion is not usually clear-cut but an accurate clinical history and competent examination are key to identifying which patients are likely to have had a stroke. In patients with a clinical diagnosis of stroke, CT or MRI is required to exclude some stroke mimics and differentiate between an ischemic or hemorrhagic accident. Because ischemic accidents are often not obvious, lowering the threshold for imaging will lead to an overuse of intravenous thrombolytic treatment of patients with conditions mimicking stroke (overtreatment) and to the overuse of imaging facilities.

    What I found in literature was that stroke mimics were infrequent among intravenous thrombolytic–treated stroke patients, and their treatment did not lead to harmful complications.

    But I didn't find guidelines that a CVA should be ruled out by imaging before the diagnosis conversion can be made.

    My question is: Is there any evidence?

    Thanks in advance,

    The best,


  • 2.  RE: Should stroke be ruled out before the diagnosis conversion is made?

    Posted 07-12-2020 19:17
    Dear Dr. Jager, 

    I found your case report very interesting, and I hope someone will be able to answer your question about evidence. 

    What occurred to me upon reading it was the question: Would conversion disorder have been the guiding diagnosis to start with if this had been a 30-year-old man? 

    Thank you, 

    Terry Graedon (anthropologist)

  • 3.  RE: Should stroke be ruled out before the diagnosis conversion is made?

    Posted 07-13-2020 01:23
    Dear Terry Greadon,

    Thanks for your reply.

    I think ("gut feeling")  there is a gender diparity in the number of misddiagnosis, but I do not have figures to state this. In this systematic review I could not found figures about bias du to gender. ( Systematic review of misdiagnosis of conversion symptoms and "hysteria".  https://www.bmj.com/content/331/7523/989 )

    " There was a significant (P < 0.02) decline in the mean rate of misdiagnosis from the 1950s to the present day; 29% in the 1950s; 17% in the 1960s; 4% in the 1970s; 4%  in the 1980s; and 4% in the 1990s. This decline was independent of age, sex, and duration of symptom in people included in the studies."

    The questionis if  a 4% misdiagnosis rate could still be seen as too high, because misdiagnosis can also happen in the opposite direction e.g. In a study, 8% of patients with a diagnosis of multiple sclerosis were later found to have conversion disorder.

    Best Gerrit

  • 4.  RE: Should stroke be ruled out before the diagnosis conversion is made?

    Posted 07-13-2020 14:46
      |   view attached

    Thanks for this important question, Gerritt.  I'm a bioethicist and philosopher of medicine whose research is focused in this area.  The article you're quoting is a case of foxes guarding the henhouse.  Psychosomatic medicine is a very small field and outsiders rarely step in to examine the science.  When we do that we quickly find problems.

    • The claim that error rates have declined to 4% is unsupported. The 17% rate in the 60s arose from studies where independent clinicians actively searched for disease in patients dx with CD.  Studies that claim we're now at 4% retrospectively examine records to determine the rate at which CD dx is overturned by clinicians themselves.  So 4% is not the error rate.  It's the rate at which clinicians themselves look for, and catch, this error.  (Imagine what the field of dx error would look like if we were generally satisfied with that kind of study!)

    • There's plenty of research to show that error rates are much higher than 4%.  See esp the first paper, which carefully addresses the situation you describe:

    • Psychosomatic medicine has long advocated unusual confidence in dx of this kind.  This recent commentary suggests clinicians actually have an ethical duty to adopt extraordinary confidence about CD dx (that's now "functional neurological disorder").  Again, imagine how the field of dx error would respond to this recommendation in any other area of medicine.

    • It's clear that error of this kind predominantly affects women, because dx of this kind is predominantly made in women.  Current recommendations generally suggest a 2-1 female-to-male ratio for conversion dx, and a 10-1 ratio for SSD (the new name for somatization).  These are straightforward directives, not personal biases.

    I continue to be mystified by lack of attention to error of this kind, given the incredible rate at which clinicians rely on psychological explanations.  My sense is that resistance to the issue has nothing to do with science.  It's widespread personal discomfort with everyday conflict about psychosomatic dx in the exam room.

    I am at your service if you have an interest in pursuing this issue -

    Diane O'Leary

    (I've attached an article of mine.  You can follow the sections "The problem of error" and "Is it true that psychogenic diagnosis rarely errs?" without reading the other articles.)

    Diane O'Leary, PhD
    Visiting Fellow, Rotman Institute of Philosophy
    London, ON Canada
    Adjunct Full Professor and Course Chair in Philosophy
    University of Maryland University College
    Adelphi MD

  • 5.  RE: Should stroke be ruled out before the diagnosis conversion is made?

    Posted 07-13-2020 14:52

    Thanks, Diane!  I think that this is a very important topic for SIDM.

    All the best,


    Michael A. Bruno, M.D., M.S., F.A.C.R.  
    Professor of Radiology & Medicine

    Vice Chair for Quality & Patient Safety

    Chief, Division of Emergency Radiology

    Penn State Milton S. Hershey Medical Center
    ( (717) 531-8703  |  6 (717) 531-5737

    * mbruno@pennstatehealth.psu.edu  


  • 6.  RE: Should stroke be ruled out before the diagnosis conversion is made?

    Posted 07-14-2020 02:38
    Thanks Diane,

    I agrree with Mike that this is an important topic for SIDM.
    I realy recommend reading your well-balanced article concerning "Ethical Management of Diagnostic Uncertainty".
    It is a challenging subject to acheive epistemic humility.

    I do have interest in pursuing this issue at a later moment.

    All the best



  • 7.  RE: Should stroke be ruled out before the diagnosis conversion is made?

    Posted 07-14-2020 10:43
    Thanks for the positive words, Gerrit.  Humility always seems like such a soft issue, but it's actually a scientific necessity.  If physicists were not very precise about what they do and do not know, physics would be impossible.  

    Best -

  • 8.  RE: Should stroke be ruled out before the diagnosis conversion is made?

    Posted 07-14-2020 21:47
    Hi Diane,

    Gridlock seems to be the basic situation in the USA in so many walks of life.

    Politically one could ask are we in trouble?

    I have written recently to the list suggesting that there may be some things in medicine that prevent us from better caring for our patients. 

    So a few questions!

    Do physicians with patient responsibilities ask enough questions of themselves and others?

    Is one of the main things that dominates and hinders clinical medicine at so many levels what I refer as the loyalty bias?

    • This is not as strong in all branches of medicine including academic, and clinical research.
    • In clinical medicine speaks to the situation that many find they are in at this time, namely little say/control in the future of their patient's health.

    One might ask are there ways to lessen the impact of the loyalty bias? I have thought of ways to do this but not sure if they could/would work!

    I have also wondered how the Scandinavian countries handle loyalty biases, appointments, communications and compromise to reach what is best for patients. What are the differences to the US systems? Is there anything to learn their?

    Rob Bell, M.D.

  • 9.  RE: Should stroke be ruled out before the diagnosis conversion is made?

    Posted 07-15-2020 08:38
    As a PT with extensive experience in the area of neurological examination who has encountered many patients for which conversion disorder is in the differential, I’d like to contribute to the discussion. I believe that stroke and other neurological diagnoses should be ruled out before a conversion diagnosis is made. I have seen cases of cognitive bias in diagnosing patients who have histories of psychiatric illness or personality disorders, or who may present embellishing symptoms on top of real signs and symptoms (this is what I call “a kernel of organicity with a cherry on top”). It is easy to dismiss these patients’ complaints quickly if care is not taken to objectively examine the patient and perform the appropriate workup. There is a tendency to make a conversion disorder diagnosis when the workup and problem-solving required might be challenging, require extra time, or an alternative diagnosis is not readily apparent, especially in this era of “conveyor belt” movement of patients through our medical system.

    Dana B. Thomas
    WakeMed Health and Hospitals
    Raleigh, NC

    Sent from my iPhone

  • 10.  RE: Should stroke be ruled out before the diagnosis conversion is made?

    Posted 07-14-2020 02:32
    Edited by Charles Pilcher 07-14-2020 02:55
    Not only should stroke be ruled out, so should any other possible stroke mimic. Any physician, especially an Emergency Physician, who diagnoses a "conversion reaction" does so at his/her peril. Conversion reaction is a diagnosis of exclusion. I am aware of a similar case involving  a middle-aged female complaining of new onset intermittent blindness over a period of hours. Not only was there no imaging, there was also no documentation of visual acuity. Turned out to be an occipital stroke.
    In my  several decade career as an EP and ED Medical Director, I can't recall a single patient among those seen by myself or colleagues, who was diagnosed as a "conversion reaction" without multiple non-psychogenic etiologies being ruled out first. If you're doing an RCA, the "root cause" is most likely to be anchoring bias.

    Charles Pilcher MD FACEP
    Chair, Board Quality & Safety Committee
    Kirkland, WA
    Editor, Medical Malpractice Insights - Learning from Lawsuits

  • 11.  RE: Should stroke be ruled out before the diagnosis conversion is made?

    Posted 07-14-2020 03:12
    Thanks Charles,

    You are right, one of the most prominent bias was indeed anchoring bias. Furtehermore sartisficing, diagnostic momentum or premature closure, gender bias, overconfidence bias and alliterating error.

    However, in the Netherlands the Inspectorate does not accept "Human Error" as a "root cause". In their excellent book "Behind Human Error", David Woods and Sidney Dekker et all state: '"Human Error" is an attribution after the fact'  ...  'erroneous actions assessments are a symptom not a cause'  and 'should be taken as the starting point for an investigation, not an ending'

    Sincerely Gerrit,



  • 12.  RE: Should stroke be ruled out before the diagnosis conversion is made?

    Posted 07-15-2020 01:40
    Dear Gerrit,

    I appreciate your posing the question to the group to take advantage of crowd-sourcing and prompting this discussion. The first thought that came to my mind, though, was that hindsight bias is embedded in the entire exercise. Could it be that your presentation to us was influenced by that and other unconscious biases such that you would get responses that confirmed what you were thinking?

    I agree that “conversion reaction” is a diagnosis of exclusion & should not be made in the ED; however, your brief case presentation left me with so many questions that I feared I would not be doing justice to your effort if I offered a critique of the care without posing those questions so I could better understand the patient’s presentation and situational factors.

    The work of Diane O’Leary, Woods et al & a host of others, many of whom participate on this listserv, have a lot to teach us about humility, and the lessons go beyond our work as diagnosticians. We need it as post hoc case reviewers, too. The Netherlands agency’s caveats seem appropriate to foster such humility in the RCA process, thereby forcing reviewers to avoid the too-easy attribution to human error which would lead to a "blame & train” corrective approach rather than getting at the root cause(s), i.e., trying to perfect the humans rather than the systems. The temptation to do the former is still quite strong, attested to by employee surveys though denied by most institutions.

    I would be interested to hear the final result of your RCA and the system factors that played a role in the patient’s care.

    Sent from my iPhone

    David L Meyers, MD FACEP

  • 13.  RE: Should stroke be ruled out before the diagnosis conversion is made?

    Posted 12-18-2020 08:20

    As a recently retired neurologist with expertise in multiple sclerosis and stroke, my 30 year experience indicates that mri imagining should be performed before diagnosing a conversion disorder. Many patients with MS have what recently have been termed prodromal symptoms which may seem psychiatric in nature before a "hard" neurological symptom such as weakness or numbness occurs. It is thought that the earlier you treat MS patients with disease modifying therapy, the more likely you are to decrease long term disability so that it is important, as it is in stroke, not to miss the diagnosis by labeling the patient as a conversion disorder. 

    I suspect that I am "preaching to the choir" but once a patient has been labelled with a psychiatric diagnosis, in my experience he is less likely to be worked up for future complaints. While there is no doubt some psychiatrists can accurately diagnose a conversion disorder based on symptoms and examination, the quality of psychiatrists (as in all medical specialties) varies but the misdiagnosis of a conversion disorder by a poor psychiatrist may stick with a patient as long as an accurate diagnosis by a true expert.

    Similarly to MS, the earlier a true conversion disorder can be diagnosed, the more likely that it can be adequately treated. Thus the cost  of an MRI to rule out a conversion disorder is well worth it.

    With regards to conversion disorders and stroke, stroke is much more likely a diagnosis and there is, as you mention, evidence that the benefit of treating most stroke mimics acutely with rTPA outweighs the risk.

    Jonathan Harris

  • 14.  RE: Should stroke be ruled out before the diagnosis conversion is made?

    Posted 12-18-2020 09:19
    Given the estimated diagnostic error rate of 40-60% in psychiatry, the rarity of true conversion disorder, and the overconfidence problem generally, it would be extremely important to rule out an underlying medical condition.
    Pat Croskerry

  • 15.  RE: Should stroke be ruled out before the diagnosis conversion is made?

    Posted 12-18-2020 09:58
    I completely agree with both Dr. Harris and Dr. Croskerry. And I would go a step farther. "Conversion reaction/disorder" should never be diagnosed by anyone other than a psychiatrist or neurologist. An emergency physician recently did exactly that in a middle-aged female with both motor and visual sx and only a CT scan to r/o anything more serious. She returned with full-blown stroke symptoms hours later. Litigation is ongoing. 
    UpToDate Online lists MS as the first of a dozen or so "rule out's" that should be in the differential.

    Charles A. Pilcher MD FACEP
    Editor, Medical Malpractice Insights - Learning from Lawsuits
    Kirkland, Washington

  • 16.  RE: Should stroke be ruled out before the diagnosis conversion is made?

    Posted 12-28-2020 13:18

    I confess to having just discovered this site, and NOT having digested this entire thread.  But I wholeheartedly agree with my colleagues Chuck Pilcher and Pat Croskerry that NO EMERGENCY PHYSICIAN should EVER presume to make a diagnosis of conversion disorder in an ED.  It is indeed cringeworthy hubris, reflecting ignorance and arrogance, and extremely perilous behaviour from a humanitarian as well as a risk management perspective.  

    If you examine migraine alone (a condition that notoriously affects females and a great many healthcare providers) in all of its myriad manifestations and its close association (especially those variants with aura) with cardiovascular disasters, one would be an idiot to presume to know to any degree of medical certainty that an odd syndrome is a conversion reaction.  

    I am glad to have discovered this forum.  Although retired, I still have much to learn from my esteemed colleagues.  

    Louise Andrew
    Rejuvacare PC