More Than a 100 Years After Flexner: Are We Achieving Expertise Through Medical Education?

I present a brief overview of how medical education has changed for the better in the United States since the early 1900s. Despite significant advancements, there is room for improvement. I propose a new goal of "expert performance" that is not in keeping with the status quo of attaining basic competence. I believe that improvement in individual performance requires a change in medical education, and in practice. My opinions have developed from a longstanding interest in the development of expertise, combined with many years as a practicing anesthesiologist. In addition, I have served as a member of the Accreditation Council for Graduate Medical Education (ACGME) Anesthesiology Residency Review Committee, an Anesthesiology Residency Program Director, an American Board of Anesthesiology (ABA) Examiner, and a departmental Vice Chair for Education. Although applicable to an international audience, my experiences are limited to US organizations. The fact that I see room for improvement even in the privileged environment of a high-performing academic medical center suggests that there is room for improvement in all locales. Medical education has advanced substantially since the late 1800s when freshly graduated medical students could immediately engage in independent medical practice. Medical schools were often "cash for diploma" arrangements with little direct patient care in the curriculum. Key stakeholders, particularly the Carnegie Foundation, commissioned Abraham Flexner, a nonphysician familiar with the European medical training model, to report on the state of US medical education. After visiting every one of the 155 medical schools in the United States, he published his highly influential Flexner Report in 1910. Flexner found that many medical schools were making a profit by offering students a medical degree despite low or nonexistent standards. His report resulted in the closing of many medical schools and led to the entrance requirement of at least 2 years of university education before embarking on 4 years of medical school. After the Flexner Report things changed for the better. Medical graduates no longer go directly into practice.1 Instead, they now embark on a course of graduate medical education (GME) which offers practical clinical training during an internship, residency, and possibly a fellowship. Flexner would likely be amazed by our medical knowledge, diagnostic capabilities, and therapeutic options. We have metaphorically reached the moon—with safe open-heart surgery, aortic aneurysm repairs, joint replacements, and transplants. But, after accommodating to these miracles, he might then wonder why we sometimes fail to render the level of care that seems so very attainable. Flexner was prone to asking tough questions. He may wonder why, for example, patients who have seen their health care provider only receive 55% of recommended care for even basic medical conditions.2 Residents who finish their Internal Medicine or Family Medicine training appear little better at diagnosis than fourth-year medical students.3 Approximately 70% of compliant hypertensive patients remain inadequately treated.4 Large variations in physician performance still exist.5 Experienced anesthesiologists often manage critical events suboptimally6 or fail to administer the very drugs that research shows will prevent postoperative nausea and vomiting in high risk patients.7 The answers to these questions are multifaceted but we could begin by looking at the overarching goals of GME. The ACGME is charged with overseeing the quality and standards of GME programs (ie, internships, residencies, and fellowships) in the United States. The ACGME is the group that promulgated duty hours, created the Outcomes Project, developed the 6 core competencies, and ushered in milestones. This same group requires program directors to place a letter in each graduate's file deeming them "competent" to engage in independent practice in the field in which they have received training. A formal definition of competent means "sufficient" or "adequate but not exceptional." It is not a high bar. After graduation, there is no requirement to improve performance beyond this minimum level of competence. The ACGME Milestones 2.0 for anesthesiology have 23 milestones each being scored on a 5-level progression from novice (1) to expert (5).8 Importantly, the ACGME notes that Level 4 is designed as a graduation target but does not represent a graduation requirement.9 Thus, expertise remains the domain of the exceptional graduate instead of the typical graduate. A minimum standard is again used during the board certification process—national oversight bodies responsible for both residency programs (ACGME) and individual practitioners (medical specialty boards) do not ask for expertise. Perhaps this is not a goal for GME programs because it takes so very long to reach expert levels of performance. In fact, the "10,000 hour" or "10 year" rules are derived from the work of Ericsson and colleagues studying the practice habits of elite performers in a variety of domains.10 Deliberate practice11 is specifically designed to improve a specific area of performance by working at the edge of competency. Deliberate practice is effortful, usually difficult, requires feedback, takes time, and focuses on areas of weakness or on areas that can be improved. Ericsson and colleagues found that even elite performers could only sustain deliberate practice for approximately 3 to 4 hours per day. Even if a resident wanted to adopt this model a 3-year residency only accommodates 2520 hours of deliberate practice. Unfortunately, typical physicians do not usually deliberately engage in learning during their workday (ie, deliberate practice).12 Today, medical education and the practice of medicine require competence but are not designed to push physicians to attain expert levels of performance. I believe that we can raise the bar by changing our approach to medical education, starting in residency, so that habits are forged that then persist over a career.13 A focus on attaining a higher bar of expert performance will establish a new norm. Habits make actions easier because they run automatically. It was Aristotle who said "We are what we repeatedly do. Excellence is, then, not an act, but a habit."14 Desirable work habits are more common among better-performing anesthesia residents.13,15 Some desirable habits are aspects of personality and are stable over time (eg, conscientiousness) whereas other habits can be specifically developed (eg, reading at least 2 articles per week from major anesthesia journals). Broadly, attaining expert performance has a least 5 key components: Professional development should occur in a culture that pursues a "learning orientation" (also known as "mastery orientation").16 Simultaneously, the prevailing culture must reduce its "performance orientation" since this only encourages the goal of looking good in the eyes of others. Performance-oriented individuals usually will not risk looking bad. Put another way, a learning orientation prioritizes improving, even at the cost of occasional failure (which in due time pays off), whereas a performance orientation prioritizes looking good. Individuals can become more learning-oriented over both the short-17 and long-term.18 Incremental self-improvement focuses on the elements of a challenging task, looking for ways to improve even when things are going well, using effort and strategy as normal components of improving and appreciating that errors and failure, although not the goal, are very potent learning tools. Strategies leading to higher performance should be used.19 The strategy of deliberate practice leads the way.6 Additional strategies include: (a) reflective reasoning20 (involving structured reanalysis of case findings such as listing symptoms of a diagnosis that are expected but are not present or listing symptoms that are present but not accounted for by the diagnosis); (b) mastery learning21 (ensuring correct task performance where any aspect that is missing, or done sub optimally, is practiced repeatedly until success is achieved); (c) spacing for better retention and performance22 (inserting time between related learning sessions where the total learning time is kept constant - for example, a single 60-minute learning session could be changed to 3 20-minute sessions each separated by a day or more); (d) retrieval23 (the active recall of information at some time after it was initially learned - the longer the interval between successful retrievals the more durable is the learning); (e) feedback to enhance learning24 (feedback is only effective at improving performance when it directs the learner to alter an aspect of what they are doing such that it becomes more effective - praise, although pleasant to give and receive, does not typically result in improved performance); (f) comparing and contrasting to enhance transfer25 (finding all the commonalities and differences between topics; for example, listing the similarities and differences between septic and hemorrhagic shock would help identify the fundamental similarities and differences between them which would lead to a more effective diagnosis with a new and somewhat different manifestation of either); (g) teaching to enhance metacognition (metacognition is the process of realizing when you do, or do not, know or understand something). Obstacles that interfere with high performance should be avoided. Working memory capacity (WMC) is used by the brain when we actively think, pay attention, process information, and learn. This processing capacity is extremely limited in humans (the typical person can only hold 7 numbers in their mind at a time) and is sufficiently vulnerable that simply walking along a prescribed path while learning words will result in less learning compared to learning while sitting.26 This limitation can be optimized by training a task to a level of automaticity which then reduces the involvement of WMC to complete the task. In turn, this leaves more WMC available to attend to other cognitive processes.27 Additional performance limitations include (a) cognitive biases and heuristics28 (biases are cognitive errors made when processing information; heuristics are cognitive shortcuts used to make quick and easy decisions); (b) dysfunctional thinking dispositions (a variety of thinking dispositions can reduce optimal decision making and performance. For example, the "need for cognition" is the tendency to engage in, and enjoy, effortful cognitive activity.29 Those with a high need for cognition perform better on intellectual tasks.29 "Actively Open-Minded Thinking" is another disposition that is favorably, or unfavorably, linked to normatively rational thinking when is high or low, respectively). Coaching or a structured curriculum is needed to ensure that components 1 to 3 above are followed and to provide the right environment. Mentors in the medical profession seldom dedicate themselves to coaching a junior colleague the way a coach of an elite athlete would dedicate themselves to improving the performance of an athlete. Prior studies have shown that physician self-assessment is poor and sometimes it is even inversely related to performance.30 In essence, physicians need to receive frequent, timely and actionable feedback from knowledgeable others that they will listen to. The learner must be persistently motivated to achieve increasing levels of performance. Significant motivation and self-discipline are needed to stay on the path of incessant improvement. Motivation is fostered through autonomy, relatedness to important others, and success in attaining goals.31 Implementation intentions have been shown to help individuals reach goals that require self-discipline.32 If items 1 to 5 above were applied to all 6 of our ACGME core competencies the result would be a truly new pathway leading towards expert performance across the landscape of GME. AN EXAMPLE FROM OUR SPECIALTY Cormack and Lehane published one of the earliest examples of deliberate practice in anesthesia. Their 1984 paper defined the 4 grades of glottic exposure during laryngoscopy; however, the paper was more focused on learning how to intubate the difficult but rare Grade 3 glottic view. They realized their need for practice with these difficult airways under controlled conditions, recommending that practitioners actively convert Grade 1 views into Grade 3 views by reducing the force on the laryngoscope handle. However, it is a rarely used method for teaching. Although modern videolaryngoscopes and fiberoptic scopes may have reduced the need for this type of practice, it remains an important example of designing methods for improving performance. SO, WHERE ARE WE GOING? For both the good of our patients and the profession of anesthesiology, we must select the expert pathway. We will need to adopt a culture founded on a learning orientation—one where the singular issue is increasing competency. Learning is a process fraught with failure and mistakes. Accepting this as an inevitable component of one's professional development is foundational. The daily use of strategies such as deliberate practice,6 the sine qua non of exceptional performance, is essential. The expert pathway is designed to lead to ever-increasing levels of performance, a goal not currently pursued or attained by the typical medical practitioner. At this point, the reader may see why a modern Flexner Report would likely make a case for improving the state of GME. In the last 110 years, humans began to reliably break the 4-minute mile, children began playing violin pieces that heretofore were considered unplayable by even the best violinists of the day and normal humans were reliably able to memorize numbers that were previously relegated to savants33—all this with a modicum of talent, a learning orientation, deliberate practice, feedback, serious motivation and the magic of long-term effort. In terms of GME, we must approach each day and each case with a plan to deliberately practice some element from 1 or more of the 6 core competencies (Table). We must use all the strategies that are becoming known which lead to higher levels of performance, and we must work to avoid those aspects of human cognition that lead to suboptimal performance. I expect we all know of individuals who independently pursue expert-level performance in some area of medicine. Thus, we currently leave it to the motivated individual to follow this pathway. Unfortunately, the number of people who spontaneously do this is only a very small percentage of all physicians. If we changed our GME programs to make expertise the norm instead of the exception, then we could potentially increase the number of physicians who pursue expertise as a life-long goal. Structural changes in GME programs aimed at teaching how to attain expert levels of performance would act to indoctrinate trainees into the process of developing expertise as a habitual part of their practice. In time, if we follow the expert pathway for the next 100 years, then the next Flexner Report might be much shorter indeed. E Table. - Examples for Deliberate Practice During a Week in the Operating Room Monday Airway skills: Combine a fiberscope and videolaryngoscope for faster intubation. This takes practice to become familiar with the logistics of the process and to yield the published time advantage. The time needed to obtain the extra equipment makes this a good choice for the first case of the day, each day of the week. For the second and third cases of the day, you can switch to practicing the left molar approach to intubation since no extra equipment is needed. This method leads to useful glottic exposure in approximately 50% of difficult intubations without the need for extra equipment. Learning the logistics of the process when the stakes are low allows for ready use of the technique when the stakes are high. Tuesday What-if scenarios: For example: what, exactly, would you do if your patient developed acute ST segment elevations in inferior leads. The proposed sequence of actions must be explicit and achievable. The process must run through all possible outcomes up to, and including, the need for an acute coronary intervention. What-if scenarios can be created to span most acute medical crises and can also be created to cover most of the core competencies. Wednesday Teamwork: Ask the nurses in your room what aspects of their job they wish you understood better so that their work would be made easier or more effective. Also, ask the circulating nurse for their input on how you positioned the patient. Adopt any constructive suggestions. Design behavioral changes to support at least one of these suggestions and enact that change within the next day. Thursday Communication: Ask patients, individual nurses (preoperative, intraoperative, and postoperative) and surgeons for a single suggestion that would have improved your communication in some manner, even if it were a minor change. Design behavioral changes to support at least one of these suggestions and enact that change within the next day. Friday Medical knowledge: For each abnormality that arises during the day (hypotension, oliguria, etc) create a complete differential diagnosis for that problem, a "most likely" top 3 list, and an explicit sequence of actions that would differentiate the various causes. Check with other physicians to ensure that your lists are complete. Complete a literature review and select articles to fill in any gaps that were identified on your initial lists. All cases have no contraindications to the proposed practice. ACKNOWLEDGMENTS The author wishes to thank the residents and faculty who have participated in "The Expert Pathway" seminar series. DISCLOSURES Name: Keith Baker, MD, PhD. Contribution: This author helped wrote the article and approved the final version of the article. This manuscript was handled by: Edward C. Nemergut, MD.