Overconfident Doctors Make More Mistakes

A study on false-positive mammography findings, published in the Journal of the American Medical Association in 2003, noted a huge difference between US and UK radiologists. US radiologists recommended follow-up (imaging, biopsy, or clinical follow-up) twice as often as their UK colleagues. Not coincidentally, US patients had twice the rate of normal biopsies as patients in the UK. However, the actual rate at which cancer was detected was similar in the two countries. Meaning? US radiologists tend to read normal mammograms as potentially abnormal more often. This can result in unnecessary invasive diagnostic procedures (e.g., biopsy) and additional testing which contributes to increasing healthcare costs as well as anxiety and greater risk of adverse events for those misdiagnosed. Interestingly, radiologists in the US are required to read a minimum of 480 mammograms per year to maintain their certification while those in the UK must read 5000 (that’s right - a 5 followed by 3 zeroes!) to meet the standards.

I found this interesting tidbit in an article titled Overconfidence as a Cause of Diagnostic Error in Medicine, published in the American Journal of Medicine in 2008. As is true in general (and there’s research to back this up), confidence tends to be inversely proportional to knowledge. That is, the less knowledge of a particular subject one possesses, the more confident that person is likely to feel or, conversely, “He who knows best knows how little he knows” (attributed to Thomas Jefferson). According to Charles Darwin, “Ignorance more frequently begets confidence than does knowledge.” A large majority of medical mistakes, something like 80%, is due to cognitive errors, i.e., mistakes in thinking. For example, if a doctor has characterized a particular patient as a whiner, a hypochondriac, that doctor may dismiss a seemingly trivial complaint without thinking of other, more serious possibilities. Jerome Groopman, M.D., in How Doctors Think, described exactly this situation. An elderly woman presented with heartburn, and he prescribed an antacid. Had he given it more thought, and thus performed a proper physical examination, he would have discovered the aortic aneurysm that was the cause of the symptom. (I highly recommend this book; it was written for a lay audience and is available in paperback in lots of bookstores. It will help you to help your doctor think better!)

Diagnostic error occurs between 5 and 15% of the time, depending on the medical specialty. (Radiologists and pathologists have the lowest rates.) Clinical guidelines and computerized decision-support systems exist to help doctors make more accurate diagnoses and prescribe better treatments, but it seems that most physicians disdain from using these tools. A study published in 2000 showed that lipid treatment guidelines (i.e., treatment for high cholesterol and hyperlipidemia) were followed only 18% of the time, even though 95% of physicians in the study were aware of the guidelines. Decision-support tools are often disregarded, even when easily available and integrated into electronic medical record systems. Overconfidence, it appears, lies at the root of many medical mistakes. (As a chiropractor, I would not exclude members of my own profession, perhaps myself at times, from this characterization.) Patients want their health care providers to be confident of their diagnostic and treatment decisions, but the doctor who appears less certain and thus digs deeper to find the answers may be more competent. Embrace uncertainty!