A member of our editorial team kindly brought me some blog fodder last week – a recent article from the New Yorker. It was an inflammatory piece, describing four types of errors that doctors make in diagnosing patients:
- Representativeness error – when a physician fails to consider diagnoses that contradict their mental templates of a disease. E.g. thin, fit, young male with chest pain – unlikely to have heart attack, but did have one.
- Availability error – the tendency to judge the likelihood of an event by the ease with which relevant examples come to mind. E.g. a patient coming to the ER in the middle of a flu epidemic, with flu-like symptoms was diagnosed with flu but actually had aspirin poisoning.
- Confirmation bias – confirming what one expects to find by selectively accepting or ignoring information. E.g. “sub-clinical pneumonia” diagnosis given even though lungs are clear.
- Affective error – the tendency to make decisions based on what physicians wish were true. E.g. nice young patient has mild fever – physician presumes it’s a typical post-op fever rather than early sepsis.
“The implicit assumption in medicine is that we know how to think. But we don’t.”
I have mixed feelings about this – for as many examples they can think of that demonstrate how physicians got the wrong diagnosis, I can also think of examples of physicians getting the right diagnosis against all odds.
Consider the middle aged woman who came to the ER with a headache – one sharp physician had a “gut feeling” that this headache was not typical, and resisted the protocol to do a head CT to rule out a sub-arachnoid hemorrhage and send her home. Instead he got blood tests that revealed the underlying diagnosis: advanced leukemia. Her blood was so thick with dividing leukemia cells that it was causing her to have a headache. She underwent immediate dialysis and survived what could have killed her.
Or what about the man who complained of chronic sinusitis? Instead of giving him an antibiotic with outpatient follow up, one physician took a detailed history and realized that this man had been having sinus pain since a recent fall from a ladder (while using a nail gun) at a construction site. The doc got a head X-ray and found a nail lodged in his sinus! During the fall the nail gun had shot a nail into the corner of his eye, leaving no entrance wound. Because of the jarring nature of the fall, the man didn’t even realize he had been shot. The man had an ENT surgeon remove the nail, and she also cleaned out what could have become a life threatening abscess.
The truth is that doctors (like anyone else) are vulnerable to making false assumptions about people – and that we would all benefit from using a software program that would automatically generate a large differential diagnosis to consider each time we see a patient (just to keep other possibilities in the forefront of our minds). However, if you ask patients if they’d rather be treated by a machine or a human being – I’m sure the majority would choose the latter. I think we can all agree that instinct and judgment still have value in this information age. The trick is to marry accurate information with good instincts without ordering every single test in the book to rule out rare diagnoses on everyone! That’s a tough balance to achieve.
Do you know of any examples of a physician making an unexpected diagnosis based on gut instincts? I’d love to hear about it.
This post originally appeared on Dr. Val’s blog at RevolutionHealth.com.