By Charles A. Pilcher MD FACEP
April, 2015
Over 100 years ago Dr. Richard Cabot reported that autopsies showed a remarkable lack of correlation between the pre-mortem diagnosis and the actual cause of death. The findings were shocking and embarrassing, but the house of medicine anticipated that they would lead to improvements in the diagnostic acumen of physicians.
But have they?
This newsletter has frequently posted articles and “Perspectives” 1, 2, 3, 4 on what leads to physician error. Readers will find a JAMA editorial particularly interesting for the light it sheds on the progress we are making (or not making) in improving patient safety. [There’s a powerful punch line here, so read on./CP]
The editorial’s author reviews the work of Cabot and other researchers in the field and distills medical error into the following categories [edited for brevity/CP]:
- ignorance, i.e., simply missing the obvious, such as cerumen in the ear as a cause of deafness, or a full bladder as the cause of an abdominal mass.
- laziness, i.e., simply believing that one knows enough to skip any further investigation. An example of this is failure to read the nurses notes or discuss an equivocal x-ray with a radiologist.
- obsession, i.e., when prior training and focus inclines one to assume that every problem is within his/her area of expertise. An example might be the endocrinologist who interprets a patient’s fever to thyroid storm when that patient is actually septic.
- failure to think critically, i.e., to assume that one’s initial impression must be correct and close off the pursuit of further options, even if to reject them. An example of this is the diagnosis of a patient with back pain as a drug seeker before addressing obvious risk factors for epidural abscess.
- incomplete examination, i.e., to prematurely reach a conclusion before ruling out other possibilities, which the author of this editorial claims is by far the most common error.
Perhaps the last point was true – when this editorial was published: March 27, 1915.
Little seems to have changed in 100 years! The only difference is that we now call these errors “cognitive biases,” and we have developed more esoteric names to describe the ones listed above from that 1915 editorial:
- “ignorance” can now be called, among other descriptors, “selective perception bias.”
- “laziness” can now be called “overconfidence effect.”
- “obsession” can now be called “knowledge bias.”
- “failure to think critically” can be called “anchoring bias.”
- “incomplete examination” can be called “confirmation bias.”
For a complete list of cognitive biases, click here.
The 1915 editorial points out that physicians must always be aware of their own biases and weaknesses and seek consultation or refer a patient when the scenario becomes cloudy. One of the references [Abrahams] advises “It is better to lose a patient [as a result of referring that patient to a consultant/CP] than to lose a reputation.” Or phrased in contemporary medical-legal parlance, “It is better to refer the patient for a consult than it is to lose a lawsuit.”
The 1915 editorial concludes with: “Mistakes due to gross ignorance and faulty judgment are being overcome by… increased improvements in medical education and by an endeavor on the part of most physicians to keep abreast with the advance in medical knowledge” [and now improved technology/CP.]
Really? Have we learned nothing in 100 years?