Confirmation bias in both physicians and patients

We are all guilty of confirmation bias from time to time.

Confirmation is something that resident physicians in particular are guilty of more so than experienced, qualified physicians.   Resident physcians and attending physicians alike may quickly form a diagnosis in their mind during a brief discussion with a patient.  Now they will try to convince themselves (sub-consciously) that the other complaints and physical exam fit that diagnosis.   They force the rest of the interaction in a certain direction.  Surprisingly, when a cross sectional study such as an MRI does not reveal the diagnosis they were suspecting, the usual response is that the MRI “missed it.”  And the confirmation bias persists.

I have seen many second opinions when the patient either had surgery or plans surgery for a presumed diagnosis, not yet confirmed with imaging studies.  After taking 5 to 10 minutes to listen to the patient, the real diagnosis usually reveals itself.   Now, I’m not claiming to be Dr. House (who is guilty of confirmation bias too), but I certainly give the patient enough time to talk until a solid list of differential diagnoses have formed in my mind.   And we don’t stop talking until that occurs.  If the light bulb doesn’t illuminate, I simply say “I don’t know” and “let’s go to the computer and look this up.”

This is a very important concept for residents to grasp early on in their training.  Not all snapping in the knee is a meniscus tear, not all shoulder pain is a rotator cuff tear and not all back pain is a pinched nerve.  Listen to your patient, ask the right questions and the correct diagnosis will usually reveal itself.

Patients are also guilty of confirmation bias.  When many patients are online looking for information, they are usually (certainly not always) starting with a presumed diagnosis given by a physician, friend, or a Googled page.  Now they search deeper and deeper into that diagnosis, even if the hints are there that this may not be the correct diagnosis.  It is not very hard to force a soft square object into a hexagon shaped cup.

One of the roles of experienced physicians and diagnosticians is to educate residents, colleagues at the water cooler and, of course, patients.  Patients are online and searching.  If they start their search with a reasonable foundation and instructions on how to minimize confirmation bias, a more educational and more productive search will ensue.

Howard Luks is an orthopedic surgeon who blogs at his self-titled site, Howard J. Luks, MD.

Submit a guest post and be heard on social media’s leading physician voice.

View 3 Comments >