During a recent evening in the critical care wing of the emergency department where I teach, I worked with a new second-year emergency medicine resident. He was enthusiastic and well organized, and we functioned well as a team. With an accurate and timely reading of the EKG, he rapidly diagnosed that an elderly woman with chest pain was having an acute myocardial infarction, started the initial medications, and activated the cardiac catheterization team. He also provided the procedural sedation to reduce a dislocated hip for a trauma patient and intubated a patient who had overdosed on antidepressants before sending her to the intensive care unit. While the nurses bustled around frantically attempting to keep up with the inflow of new patients, the resident’s calm approach helped keep the situation from seeming chaotic or out of control. Of the 16 patients we treated that night there were probably only 3 where my presence may have made a difference in the outcome—one where I helped with a difficult procedure, another in which I elicited certain information from the patient that changed our diagnosis, and a third in which I detected an abnormality in a patient’s laboratory test that led to a change in the resident’s treatment plan.
The similarities between my approach and that of the resident prompted me to consider the notion of expertise and its importance in our health care system. How was it possible for a resident with only one year of experience to perform at almost the same level that I did? How important were the differences?
Much of our work in medical education is focused on the goal of guiding our students and residents in their efforts to achieve competence if not genuine expertise. Fortunately, in this month’s Academic Medicine we have a cluster of articles1–9 that can increase our understanding of the most effective ways for us to accomplish that goal. These articles, coordinated by William McGaghie, PhD, are about mastery learning and describe the mastery learning methodology, its value in medical education, its use in education research, and its potential to improve the quality of medical care through implementation in clinical settings.
McGaghie introduces the concept of mastery learning and describes each of the articles in a Commentary.3 In this editorial I will focus primarily on one of those articles, written by Ericsson.1 He offers some interesting observations about expertise and expert performance in numerous fields outside of medicine that may help our thinking about how to achieve expert performance in medicine. His insights helped me understand why the differences in performance between experts and nonexperts may not be as great as we might imagine. In describing how to distinguish experts in chess through the study of the moves of chess pieces during a game, Ericsson notes:
All chess moves are not equally important for outcomes of chess games; thus, some moves will not differentiate chess players with different skill levels. In fact, evidence shows that only a small number of critical moves per game will clearly distinguish superior players from those with lower chess ratings.
If these same principles apply to critical care, might they explain the small differences in performance between me and my resident? If so, are they enough to justify our enormous investment in medical education to develop expertise? What else might we learn from Ericsson’s study of expertise in other fields and McGaghie’s work on mastery learning that might improve our approach in medical education? Later in this editorial I will offer two recommendations that relate to these questions.
To think about the application of Ericsson’s theory of expert performance to medical education, I divided the content of medical education into four areas:
The knowledge needed to understand human illness—what does a student need to know to practice medicine? This includes all the vocabulary of medicine, the names and functions of the various organs, and how they become diseased. We teach this in various ways—lectures, problem-based learning, team-based learning, reading, flipped classrooms. Traditionally, the first two years of medical school have been dedicated to this area of medical education. Around the world, students learn this same core knowledge in diverse medical schools, and it has often amazed me how even those students from the least technologically sophisticated training programs can still learn this knowledge well enough to achieve good scores on the tests that we use to assess knowledge at national levels and readiness for residency education in the United States.
Ericsson refers to this area of education as the acquisition of knowledge and rules, and he notes the extensive emphasis of this aspect of learning in medicine, which can involve thousands of hours of education, compared with a small amount of time dedicated to fostering specific areas of expert performance that will be needed later in practice. He further notes that the ratio of time dedicated to general knowledge acquisition compared with time dedicated to direct practice to develop expertise is the reverse in medicine from that in music, sports, or chess. He states that mastery learning may be a valuable methodology to assist learning when sequential knowledge (i.e., knowledge that must be learned in steps of increasing difficulty) is needed to develop a coherent understanding of complex and challenging content.
The acquisition of procedural skills—what must a student be able to do to diagnose and treat patients? Such skills can include everything from how to examine the heart to how to deliver a baby, how to read an EKG, and how to place a needle in the back to obtain cerebral spinal fluid. Teaching this part of medical practice was once focused on gaining clinical experience with actual patients under supervision, but has more recently included gaining such experience using simulation. In this area of medical education, the risks of inadequate performance can be clearly identified should the procedure be performed incorrectly or a complication develop.
Studies of expertise in medicine have often focused on the acquisition of procedural skills in medical education because that process resembles the acquisition of procedural skills in other areas of human activity, such as playing a musical instrument. Mastery learning, with its standardized, criterion-based advancement strategy, may be helpful in the initial learning of procedures. McGaghie3 and McGaghie et al4 describe the principles of mastery learning, which include an initial assessment; clear learning objectives sequenced with increasing difficulty; educational activities such as deliberate practice or reading; designation of a criterion for advancement; testing; advancement when the criterion for mastery is met; and continued practice and review. Eppich et al7 describe the important role of feedback and debriefing as part of the mastery learning process. The importance of individualized feedback to improve performance may be an area of overlap between the approaches of mastery learning and Ericsson’s model of expert performance with deliberate practice. Ericsson describes his model as depending on an understanding of the goals to be achieved, the use of specific feedback, and concentrated efforts at improvement through practice; this model is similar to that used in developing expertise in music, sports, and chess.
The integration of information to make decisions—essentially, how does a student think about a problem and arrive at a diagnosis and treatment plan? This area of medical education is about filtering, selection, and integration of information into cohesive patterns that can be recognized and acted on. Although medical knowledge and procedural expertise can assist in the collection of the information, the recognition of complex patterns typically requires experience and problem-solving abilities. The learning involved in this aspect of medical education has been characterized by years of reflective experience, exposure to a variety of problems, and close supervision by expert teachers. Ericsson suggests that theories of expertise posit the gradual replacement of acquired knowledge with patterns, based on experience, that become retrievable through memory. Croskerry et al10 have described a similar idea in their discussion of two types of diagnostic decision making, “the reflexive, autonomous processes that characterize intuitive decision making and the deliberate reasoning of an analytical approach.” Experts not only have a more extensive set of recognizable patterns stored in their memories but also can identify situations where stored patterns do not fit a current problem and in which the use of the second, less intuitive type of thinking must be used. Both mastery learning and also training based on achieving expert performance with deliberate practice can be used to develop problem-solving skills through the presentation of illness scripts or simulated problems that allow a student to practice on problems that might occur only rarely in actual practice, thus expanding the student’s practice skills.
The attitudes and behaviors that are related to physician identity—how does the student become transformed into a good physician? These attributes develop as the student proceeds through residency and then practice, all the while learning how to communicate, develop trust, empathy, compassion, confidentiality, and other characteristics of professionalism considered the hallmarks of a good physician. The teaching, learning, and assessment of these skills continue to be elusive, and discussions of these challenges continue to fill many pages of this journal. Ericsson believes that communication skills may be taught like other skills, with specific goals for performance and improvement.
In presenting these four areas I intentionally left out other important topics of medical education such as leadership, teamwork, and development of health care systems as well as the important role of the learning environment that nurtures learners and teachers. Although all of these topics contribute to the development of expertise in medical care delivery, they do not fit easily into Ericsson’s theory of expert performance with deliberate practice.
Based on Ericsson’s work on expert performance with deliberate practice presented in this issue of Academic Medicine and the cluster of articles developed by McGaghie on mastery learning, I make the following two recommendations, which are relevant to my earlier questions about the teaching and application of expertise in medical education.
- Findings from the study of expertise in music, sports, and chess can provide valuable insights into the development of expertise in medicine. The current predominant emphasis on acquisition of general knowledge is one strategy in which medical education varies from education in other fields. Education focused on the principles of expert performance with deliberate practice should be explored as a possibly more rapid and effective route to the development of expertise in some areas of medicine.
- Mastery learning provides an organized approach to the development of competence and should be explored as a methodology that is compatible with outcomes-based medical education. There are now many examples of how mastery learning methodology has improved learning and the quality of care to patients that could inform its adoption into many areas of medical education.
While I could easily end the discussion of mastery learning and expert performance with deliberate practice with these recommendations, I must also recognize the limitations of these approaches as solutions to the health care problems in our country. Expert performance by a physician, although important, does not always ensure excellent care for the patient. Studies of medical errors have shown that systems failures rather than individuals’ lack of expertise are often responsible for medical errors and bad patient outcomes.11
In addition, we have substantial variability in the quality of care provided in our country based on geography, race, class, and economic resources. Although the aspirational ethic in the United States is access to the highest quality of care for all regardless of class, race, gender, or economic resources, the reality of expert medical care is that it depends not only on the expert skills of individual physicians but also on organized systems of referral and regionalization of expertise. The development of trauma systems, which depend on field triage by paramedics and preferential transport of injured patients to trauma centers, provides a good example of how organized systems of care can improve the outcomes of patients.12
As I reflect on the questions I raised earlier about the performance of the second-year resident, I realize that I may have been asking the wrong questions. Although the work of Ericsson helps to explain the rapid learning curve of the resident and the partial overlap in the skills of experts and trainees, it does not provide guidance about which skills our future physicians will need to function as parts of teams to provide excellent care for patients. Nor does it describe what features of our future health care delivery system will be most important in helping us to improve the health and health care of our population at a cost we can afford. For example, how critical will the production of various physician experts be compared with the training of other health care team members or the design of regional networks of care? This is not to say that we must choose between highly expert physicians and excellent systems of care. We must have both.
Decisions about resource allocations will require active discussion between medical educators, policy makers, and the public as we envision our future health care system. We can reaffirm our commitment to excellence in health care through the training of our medical students and residents to the highest levels possible, and mastery learning and education for expert performance with deliberate practice can make valuable contributions to achieving this goal. However, we must be careful not to equate physicians’ expertise with excellence in health care. Unlike music, where the expert solo violinist may captivate an audience, excellent medical care requires the coordination of an entire orchestra as well. Our medical education programs must educate and support the entire orchestra as well as the soloists.
1. Ericsson KA. Acquisition and maintenance of medical expertise: A perspective from the expert-performance approach with deliberate practice. Acad Med. 2015;90:1471–1486
2. Lineberry M, Park YS, Cook DA, Yudkowsky R. Making the case for mastery learning assessments: Key issues in validation and justification. Acad Med. 2015;90:1445–1450
3. McGaghie WC. Mastery learning: It is time for medical education to join the 21st century. Acad Med. 2015;90:1438–1441
4. McGaghie WC, Barsuk JH, Cohen ER, Kristopaitis T, Wayne DB. Dissemination of an innovative mastery learning curriculum grounded in implementation science principles: A case study. Acad Med. 2015;90:1487–1494
5. Yudkowsky R, Park YS, Lineberry M, Knox A, Ritter EM. Setting mastery learning standards. Acad Med. 2015;90:1495–1500
6. Cohen ER, McGaghie WC, Wayne DB, et al. Recommendations for Reporting Mastery Education Research in Medicine (ReMERM). Acad Med. 2015;90:1509–1514
7. Eppich WJ, Hunt EA, Duval-Arnould JM, Siddall VJ, Cheng A. Structuring feedback and debriefing to achieve mastery learning goals. Acad Med. 2015;90:1501–1508
8. Griswold-Theodorson S, Ponnuru S, Dong C, Szyld D, Reed T, McGaghie WC. Beyond the simulation laboratory: A realist synthesis review of clinical outcomes of simulation-based mastery learning. Acad Med. 2015;90:1553–1560
9. Inui TS. The charismatic journey of mastery learning. Acad Med. 2015;90:1442–1444
10. Croskerry P, Petrie DA, Reilly JB, Tait G. Deciding about fast and slow decisions. Acad Med. 2014;89:197–200
11. Kohn LT, Corrigan JM, Donaldson MS To Err Is Human: Building a Safer Health System. 1999 Washington, DC Institute of Medicine, National Academy Press
12. Celso B, Tepas J, Langland-Orban B, et al. A systematic review and meta-analysis comparing outcome of severely injured patients treated in trauma centers following the establishment of trauma systems. J Trauma. 2006;60:371–378