A few years ago, when Andrew Olson, MD, was chief resident at the University of Minnesota Medical Center, he noticed it was relatively easy to identify and intervene on a variety of medical errors. But, at the same time, he also noticed that it was extremely difficult to identify a diagnostic error.
In fact, Olson said it seemed no one even used the term “diagnostic error,” which is defined as a missed, delayed, or incorrect diagnosis. Olson always loved problem-solving, so after finding a dearth of existing educational tools on diagnostic errors, he set out to create his own.
After conducting a needs assessment among fellow residents, Olson and colleagues found that only 27 percent of respondents were aware of strategies to reduce cognitive errors that contribute to diagnostic errors. In addition, only one-third said their supervisors encouraged them to discuss cognitive errors. With feedback in hand, Olson and collaborator Emily Ruedinger, MD, created a six-part curriculum on diagnostic reasoning and errors for pediatric residents that included personal stories from physicians, malpractice attorneys, and from a family whose child died after a diagnostic error. Olson and Ruedinger’s work was funded in part with an AAMC Clinical Care Innovation Award. The interactive curriculum aims to help residents recognize the strengths and weaknesses of different thought processes, as well as define common cognitive biases that can lead to diagnostic error.
“Diagnosis is a very special challenge because it’s just so complicated—there are over 10,000 diseases and over 5,000 lab tests and putting it all together is difficult.”
Mark Graber, MD
Society to Improve Diagnosis in Medicine
In evaluating the curriculum, which has been taught for the past three years, Olson and colleagues found residents reported significant improvements in their ability to define cognitive errors. “This is the next frontier in patient safety,” said Olson, now an assistant professor of medicine and pediatrics at the University of Minnesota Medical School.
New focus on diagnostic error
Most medical professionals point to the 1999 Institute of Medicine report To Err Is Human as the beginning of the modern patient safety movement. However, that report focused mostly on errors such as health care–associated infections, falls, and wrong-site surgery, with little focus devoted to diagnostic error, said Mark Graber, MD, founder and president of the Society to Improve Diagnosis in Medicine.
“Diagnosis is a very special challenge because it’s just so complicated—there are over 10,000 diseases and over 5,000 lab tests and putting it all together is difficult,” said Graber, also a senior fellow at RTI International and professor emeritus of medicine at Stony Brook School of Medicine.
Outside of research studies, Graber noted that while little is known about the true effect of diagnostic errors, a 2013 article in BMJ Quality and Safety reported that diagnostic discrepancies were identified in 10 to 20 percent of autopsies of patients who died in inpatient and emergency room settings. Another BMJ Quality and Safety article published in 2014 estimated that at least one out of 20 U.S. adults will be affected by a diagnostic error. The need to better track and address diagnostic error received greater attention in 2015 with the release of the National Academies of Sciences, Engineering, and Medicine publication Improving Diagnosis in Health Care, which concluded that “urgent change is warranted to address this challenge.” Along with other health care organizations, the AAMC is part of the Coalition to Improve Diagnosis, which is bringing awareness, attention, and action to the problem of diagnostic error.
Graber, a member of the report’s authoring committee, said diagnostic error typically occurs for one of two reasons: a breakdown in clinical reasoning, such as jumping to a conclusion, or an issue in the health care system, for example, a lack of communication between members of a care team. The National Academies report on diagnostic error notes that the “diagnosis process often involves intra- and inter- professional teamwork.” There are strategies to prevent diagnostic errors through improved cognitive reasoning, but few medical schools have incorporated such education. To fill that gap, Graber is in the process of creating a diagnostic error curriculum for medical students.
“This is a problem that’s leading to a huge amount of harm,” he said. “We need doctors who … know when they need to slow down and get a second opinion.”
Jennifer Myers, MD, an associate professor of clinical medicine at Perelman School of Medicine at the University of Pennsylvania, said a first step to teaching about diagnostic error is developing a shared vocabulary around the topic. Myers and colleagues began developing a diagnostic error curriculum for medical trainees about five years ago.
“In terms of clinical reasoning, different schools teach it in different ways,” said Myers, who also directs quality and safety education in the Department of Medicine and is director of Perelman’s Center for Health Care Improvement and Patient Safety. “But when you put a diagnostic error lens on it, it really allows you to make it a burning platform and say, ‘We’re a profession that learns from our mistakes, and if one of those mistakes is faulty reasoning then let’s talk about how we reason.’”
Today, Perelman School of Medicine students learn how errors can happen and how to recognize suspect reasoning during the second-year Differential Diagnosis course. Diagnostic error education has also been integrated into clerkships and the internal residency program.
“For us, this is a starting point,” Myers said. “But it needs to be in every residency and fellowship program and in every medical school.”
New educational strategies emerge
Pat Croskerry, MD, PhD, director of the Critical Thinking Program in the Division of Medical Education at Dalhousie University in Nova Scotia, Canada, said research estimates that while both systems problems and physician thinking contribute to diagnostic errors, cognitive issues are responsible for up to two-thirds of diagnostic errors. And when that diagnostic error is attributed to an individual physician, it’s not typically a knowledge failure but a thinking failure, Croskerry said.
“Medical colleges do an exceptional job imparting knowledge, but we don’t historically put much emphasis on how to train doctors to think,” said Croskerry, who was also a member of the committee that authored the 2015 National Academies report on diagnostic error.
Dalhousie implemented a critical thinking program for medical students about four years ago, with lectures that run throughout the undergraduate years. For example, students receive a lecture on how doctors think that includes four emergency room scenarios in which the patient for each scenario has the same right-lower abdomen pain but is given a different diagnosis. Students learn that discussing symptoms with patients is only the beginning of the diagnostic process, Croskerry said. They also learn that uncertainty is okay.
Back in Minnesota, Olson is the primary investigator on a project to develop six online cases that medical schools can use to teach diagnostic error. The cases are now being piloted in seven medical schools and are expected to be available in 2017.
“We all have a stake in the diagnostic process and we all benefit from it getting better,” he said.