Social psychologist David Dunning, PhD, recently said he had fallen victim — for years — to the cognitive bias he had helped name. That famous 1999 article in the Journal of Personality and Social Psychology revealed that the Dunning-Kruger effect occurs when individuals assume that their competency in a given area is significantly higher than it is.
“I have asthma, and just the other day I ran across a test on how to appropriately use an inhaler,” said Dr. Dunning, a professor of psychology at Cornell University. “I took the test just for fun because obviously I knew all the right answers. I've been using an inhaler for 15 years!”
But it turned out that he had been using the inhaler wrong for all that time. “I was breathing in heavily really quickly when you're supposed to take breaths slowly. It was a shock to me!” he said. “I had been depriving myself of oxygen that was there for the taking. I've been feeling much better since I began doing it correctly.”
The original Dunning-Kruger effect article opened with the anecdote of a man who robbed two Pittsburgh banks in 1995, undisguised, in broad daylight, under the mistaken belief that rubbing lemon juice all over his face — because it was the key ingredient in invisible ink — would render him invisible to video cameras.
But the effect does not just apply to someone as hilariously clueless as the incompetent thief. We all fall victim to the Dunning-Kruger effect: engineers, accountants — even emergency physicians. And it's not a failing: We just don't know what we don't know. A corollary to the Dunning-Kruger effect is often that people are not completely uninformed, but rather are misinformed, as Dr. Dunning himself was about the asthma inhaler.
“We all have our own pockets of incompetence, and some of them we can identify, but in others, we don't know how far the rabbit hole goes,” he said. “And the worse we're doing, the more pronounced is the difference between our perception of our own competence and how competent — or incompetent — we actually are.”
The Dunning-Kruger effect is a universal human phenomenon, and studies have shown that it plays a role in a number of medical professions, ranging from clinical lab technicians to obstetrics and gynecology. Emergency medicine, however, may be particularly vulnerable for a few reasons, said Robert Wears, MD, PhD, a professor of emergency medicine at the University of Florida and an expert in diagnostic errors.
First is the issue of pace and stakes. “In a specialty like family medicine, for example, the pace at which things evolve is generally slower, and you have more ability to compensate for ill effects before they become permanent and damaging,” Dr. Wears explained. “In emergency medicine, often you have only one shot to get it right, and if you mess that up, the consequences could be severe and are probably out of your control. It's a lot less forgiving environment.”
The Dunning-Kruger effect can be a particular problem for residents and young attendings, said Lauren Westafer, DO, MPH, an emergency physician and research fellow at Baystate Medical Center in Springfield, MA, who wrote about the effect on her blog, The Short Coat (http://bit.ly/2cjqDxt). “In my experience during residency, people would say the most dangerous residents are in their mid-second year. They pick up speed, they see more patients, and feel more comfortable. That's when they're most dangerous,” she said. “Residents are probably safest at the beginning when they're scared and know that they don't know what's going on. When you get part way through your second year, you feel more comfortable and you may not be aware of all the things you still don't know.”
This is backed up by research that compared the actual accuracy of physician, resident, and medical student diagnoses with their perceptions of that accuracy. (J Gen Intern Med 2005;20:334.) Attendings were the most accurate and highly confident in their accuracy. Medical students were less accurate, but they also lacked confidence. Residents were more confident about the correctness of their diagnoses, but were not as accurate as the attendings.
Residents may also overestimate their psychomotor skills in doing tasks like intubating or inserting a central line. “After they've done a few, they tend to have an inflated idea of their abilities because their experiential base is not big enough for them to have encountered one of the rare, horrific problems that can come up,” Dr. Wears said. “Someone with more experience, on the other hand, has run into those things, and they're more aware of how close to the edge they might be even without having a clear warning that they're in danger.”
Emergency medicine residents may also have more autonomy during these high-risk years of overestimated competence than some other specialties, Dr. Westafer said. “As an attending, I now have to trust my residents in the first 30 minutes or an hour when they see a patient, and missteps during that time could mean life or death. A surgical resident, on the other hand, theoretically at least has the attending surgeon there watching at all times.”
Emergency physicians also often get less feedback about the outcomes of their care decisions than other physicians who practice in more longitudinal specialties. And feedback is one potential way to compensate for the Dunning-Kruger effect, said Eta Berner, EdD, the director of the Center for Health Informatics for Patient Safety/Quality at the University of Alabama-Birmingham, who has written on overconfidence as a cause of diagnostic error in medicine.
“For the most part, a physician's errors are not frequent. So a given physician is not going to have a lot of experience making a lot of mistakes. Most of the time they're right, so they're reinforced most of the time,” she said. “But when they are aware of errors, they are often devastated and take it very seriously. In emergency medicine, unless something goes very wrong, you probably don't ever hear about your patient's outcome after he is transferred to an inpatient unit or goes home.”
That's something Dr. Westafer is trying to address with her own residents. “Emergency medicine is a snapshot, and you don't always get feedback unless you go looking for it,” she said. “So I sit down with my residents and have them each make a master list in the EMR. Whenever they sign up for a patient, they add him to that list. When they log in, with one click they can look at the most recent of their patients. I ask them to look up a few patients a couple of times a week and find out what happened. Who's still in the hospital? Who's in the CICU? What happened to them and why?”
Physicians can also try to address their own Dunning-Kruger effect by pursuing additional training. “The effect is a property of lack of competence, and the solution to lack of competence is to get more competence,” Dr. Wears said. “Improving yourself as a doctor is the single best thing you can do. You can't diagnose Brugada syndrome if you've never heard of it, no matter how good you otherwise are at reading ECGs.”
Young doctors also should be aware of the Dunning-Kruger effect to remind themselves to cultivate a sense of humility, he added. “Novices, when they reach a tentative working diagnosis, often hold onto it like the tree of life because it's so much work to go back and revise. Experience teaches you to hold other possibilities idling in the background. And always be ready to bring fresh eyes to a situation to get out of your own blinders. Someone else can come in, and say, ‘That's not a stroke; that's a dissection,’ and you'll say, ‘Why didn't I see that?’”