Thinking about thinking is the science of cognitive psychology and addresses the cognitive aspects of clinical reasoning underlying diagnostic decision-making. It is an area of study in which few medical professionals are versed. “Except for a few of these guys who trained in psych or were voices in the wilderness that have been largely ignored,” most physicians are unaware of the cognitive psychology literature, Dr. Groopman says.
Common biases and errors in clinical reasoning are presented in Table 1 (right).4,5 These are largely individual mistakes for which physicians traditionally have been accountable.
Patterns and Heuristics
The following factors contribute to how shortcuts are used: the pressures of working in medicine, the degrees of uncertainty a physician may feel, and the fact that hospitalists rarely have all the information they need about a patient.
“That’s just the nature of medicine,” says Dr. Groopman. “These shortcuts are natural ways of thinking under those conditions. They succeed about 85% of the time; they fail up to 10-20% of the time. The first thing we need to educate ourselves about is that this is how our minds work as doctors.”
Dr. Groopman and those he interviewed for his book have a razor-sharp overview of clinical practice within hospitals throughout the U.S. and Canada, including academic centers, community centers, affluent areas, suburbs, inner cities, and Native American reservations. But except for Pat Croskerry, MD, PhD, in the department of emergency medicine at Dalhousie University’s Queen Elizabeth II Health Sciences Center in Halifax, Nova Scotia, none of the experts he interviewed had rigorous training in cognitive science.
Although how to think is a priority in physicians’ training, how to think about one’s thinking is not.
“We are not given a vocabulary during medical training, or later through CME courses, in this emerging science—and yet this science involves how our mind works successfully and when we make mistakes,” Dr. Groopman says.
The data back this assertion. In a study of 100 cases of diagnostic error, 90 involved injury, including 33 deaths; 74% were attributed to errors in cognitive reasoning (see Figure 1, right).1 Failure to consider reasonable alternatives after an initial diagnosis was the most common cause. Other common causes included faulty context generation, misjudging the salience of findings, faulty perception, and errors connected with the use of heuristics. In this study, faulty or inadequate knowledge was uncommon.
Underlying contributions to error fell into three categories: “no fault,” system-related, and cognitive. Only seven cases reflected no-fault errors alone. In the remaining 93 cases, 548 errors were identified as system-related or cognitive factors (5.9 per case). System-related factors contributed to the diagnostic error in 65% of the cases and cognitive factors in 74%. The most common system-related factors involved problems with policies and procedures, inefficient processes, teamwork, and communication. The most common cognitive problems involved faulty synthesis.
Dr. Groopman believes it is important for physicians to be more introspective about the thinking patterns they employ and learn the traps to which they are susceptible. He also feels it is imperative to develop curricula at different stages of medical training so this new knowledge can be used to reduce error rates. Because the names for these traps can vary, the development of a universal and comprehensive taxonomy for classifying diagnostic errors is also needed.
“It’s impossible to be perfect; we’re never going to be 100%,” Dr. Groopman says. “But I deeply believe that it is quite feasible to think about your thinking and to assess how your mind came to a conclusion about a diagnosis and treatment plan.”