In Albuquerque, when we are lucky, a storm may blow through town with clouds and rain just before sunset, and a little later the sun will shine from the west. At that moment, a rainbow will sometimes appear over the mountains to the east. I’ve seen several such rainbows, but each time one appears it feels like magic. I become pensive and full of awe, clearing my mind so that I can appreciate the beauty in front of me for the brief time it will last.
As I reflect on this feeling, I am reminded of a quotation from Pablo Neruda1 in The Book of Questions: “Where does the rainbow end, in your soul or on the horizon?” Will the rainbow inspire our imagination, or will it become an object of study and measurement? Or can it be both?
Today as I write about the continuum of medical education, I think a rainbow, with its gradual change of colors that blend into each other, is an apt metaphor for the transformation of our students that occurs incrementally from day to day, the way red in a rainbow gradually turns into orange and orange into yellow and yellow into green until finally it comes to violet. Somehow, over several years as students progress through the phases of education, the transformation happens, although I’m not sure exactly how or when.
Challenges to a Continuum
While we can take joy in the eventual competence of our trainees, what happens during their transformation is not without challenges. The medical education process is expensive and stressful and may not produce the right mix of physicians who will go to the places where they will be needed for the future. We also lose potential physicians all along the continuum. For example, those who are eliminated during large science lecture courses in premedical programs because they lack adequate preparation. Or the students who survive the admissions process but leave medical school because of the mental and physical distress they experience that leads to burnout. Or those students who survive medical school but fail to match in their desired specialties and lose their enthusiasm for a medical career. In spite of the sense of awe that I feel at the many successes of our medical education system, I am saddened by our failures all along the continuum of medical education and wonder whether we could reduce them.
A Confederacy Rather Than a Union
Iobst and Holmboe2 note that
for over 100 years medical education has been delivered along a continuum that can be characterized as a series of linked but independent silos. While the design of these silos may vary worldwide, they typically include premedical education, undergraduate medical education, graduate medical education, and continuing medical education. Historically, the alignment of learning outcomes across these silos has reflected a confederacy rather than a union of stakeholders.
This lack of a union has resulted in fierce competition at each stage of education. Students are confronted with a series of barriers akin to the moats that surrounded medieval castles. Their portfolios and standardized tests serve as armor to protect them on their long and uncertain journey. The curriculum that guides their education has become a fragmented mix of commercial testing aids and courses that increasingly focus on preparing students for successful competition rather than giving them the knowledge and skills they will need at the end of training to provide high-quality health care.3 An example of the consequences of this fragmented continuum is the Step 1 climate at many medical schools described by Chen et al,4 in which preparation for the Step 1 examination of the National Board of Medical Examiners overwhelms other areas of study because students recognize the influence of that test score on their future career options. Chen et al report that
we have witnessed classmates balk at the incorporation of clinical ethics, social determinants of health, health systems, and social justice in preclinical curricula—such topics are deemed “low yield” for Step 1.
Finding Connection Through Competencies
Aschenbrener et al5 have described how graduate medical education (GME) is helping to overcome the separations of the phases of medical education with its creation of an outcomes-based model with designated competencies that are now being embraced by organizations responsible for education and accreditation in the other phases of medical education. They note that “the ultimate outcome for medical education is physicians who consistently provide safe, effective, patient-centered health care.” They describe opportunities to improve premedical education so that the competencies that will be desired later in the education process (such as service orientation, ability to work as a team, and ethical behavior) can be taught and assessed. They also mention changes in continuing medical education (CME) that align with population health, team-based care, patient safety, and quality improvement. Aschenbrener et al provide a vision that could counteract the current fragmented continuum if fully embraced.
Alignment of Education and Clinical Care
CME is the part of the educational continuum closest to patients and their experience. When CME is integrated into clinical care improvement, it may provide important insights for the rest of the educational continuum. McMahon6 suggests,
An empowered CME program with its multiprofessional scope and educational expertise can contribute to initiatives in both clinical and nonclinical areas, such as quality and safety, professionalism, team communication, and process improvements.
He encourages health care organizations to use CME to convene faculty, fellows, residents, and students across the continuum to address institutional priorities through education, which would improve quality of care at the institution and improve learning at all levels. I believe this suggestion could help provide a unified purpose for medical education that could guide a competency-based continuum.
Similarly, Gupta et al7 in this issue describe aligning education and delivery system missions centered on GME:
Traditionally, medical education and the health delivery system have existed in different silos and addressed different [academic medical center] missions, which can affect the training of the future physician workforce.
They encourage the development of new leadership roles that bridge the clinical and GME missions in quality improvement and patient safety, and highlight 4 additional steps: aligning strategic priorities between delivery systems and educators, supporting alignment with data, leveraging existing staff and infrastructure, and having incentives to encourage alignment.
While integration of education and clinical care in GME with its heavy patient care component could be possible, what about in the undergraduate and premedical education phases?
Moriates et al8 in this issue describe the development of online modules for teaching about high-value care that can be used by medical students, residents, and other health professionals. The use of the Internet can be a useful way to spread new content across the continuum. The authors describe integrating the content into the second year of medical school; the modules used for second-year students could also be adapted for premedical studies, creating an alignment of content and values at the earliest stages of the continuum. Gonzalo et al9 have described proposals for implementation of health systems sciences in undergraduate curricula at 30 medical schools that indicate additional ways of integrating education and clinical care to provide high-value care.
Using Entrustable Professional Activities to Create the Continuum
With the common purpose for medical education and clinical care being to achieve high-value health care, the stage is set for an implementation strategy across the continuum, and, in this issue, ten Cate and Carraccio10 describe a competency-based strategy. They explain how entrustable professional activities (EPAs) could knit together the continuum:
The current structure of medical education—with sequential and separated rather than integrated programs; one-size-fits-all, fixed lengths of training; additional training sequences of prescribed duration for subspecialties; and ever-increasing costs to the individual, the profession, and the public—appears unsustainable. However, educators now can define the EPAs needed for physicians within highly diverse practice types, map them to the competencies that are critical to making an entrustment decision, and build curricula and assessments that foster learning experiences with progressive autonomy.
In their model, trainees would progress to competence through core EPAs in medical school that would provide the foundation for their training in all residencies. In the residency phase, some EPAs from medical school would continue to develop, new EPAs specific to the specialty would be added, and some EPAs might become unnecessary and atrophy. In clinical practice, there would be further differentiation and growth. With the rapid evolution of new technologies and research findings, practicing physicians would need to continue to learn and refresh prior skills to maintain competence and build new areas of expertise. The authors maintain that
in essence, EPAs act as building blocks. For example, the foundations of many houses are similar—they all may be brick or stone. Once the foundations are laid, the houses will be built with a limited number of materials, but they ultimately will become different sizes and styles, like the EPAs in GME and practice for individual physicians.
ten Cate and Carraccio also go on to describe how their model will be applied for practicing physicians:
Mastered EPAs constitute the building blocks of professional work. Building a dynamic portfolio of these EPAs requires: (1) continued practice of those EPAs for which summative entrustment decisions have been made; (2) adaptation to enhanced standards of practice, reflecting up-to-date abilities if techniques and insights have evolved; (3) acquisition of new knowledge and skills leading to entrustment for additional EPAs in one’s field of practice; and (4) deliberately leaving behind those EPAs, beyond the requirements of the appropriate certifying bodies, that one is not practicing. Practice, and thus the EPAs that define a physician’s practice, will change.
The authors provide a mechanism for the progression of physicians through the educational continuum in a way that prepares them to understand and deliver high-value health care to our population.
Impediments to a Continuum
Unfortunately, there are several impediments to a continuum based on competencies and EPAs. The first is the distracting effect of the standardized tests for selection to medical school and residency, which emphasize content expertise and skills that are of little relevance to our current health system goals. These tests can divert a student’s attention from developing competencies that create a continuum that aligns with care delivery goals, and can also distract selection committees into giving undue influence to test scores. The dominance of standardized testing in student assessment has created a self-imposed tyranny that can dwarf other assessments of competence or expertise.
This problem is particularly relevant to the assessment of practicing physicians and decisions about granting maintenance of certification. In other fields such as music, sports, or aviation, observation and analysis of data provide the most important information about performance and expertise. Statistics about batting average, strikeouts, errors, and team performance in baseball have been collected for many years and have been used to make decisions about the performance of a player and that individual’s value to the team. The idea of testing players’ knowledge about baseball rules or the physics of a bat hitting a ball would seem silly. Yet we do not hesitate to ask basic knowledge questions to physicians to assess their expertise. While there is a role for standardized tests of content and problems relevant to the care delivery goals to assess competence, their use to compare trainees in selection processes or certification processes should be balanced with other relevant information, such as our substantial data on students, residents, and practicing physicians and efforts to link those data to the outcomes of patients.
Another impediment to the adoption of EPAs throughout the continuum is the time and training required for our faculty to provide workplace-based training and assessments of our students and residents. Previous models of medical education assumed that a basic science foundation followed by a period of clinical knowledge and experience would provide sufficient preparation for clinical practice. Faculty time could be limited to scheduled lectures and informal clinical assessments.
Competency-based education requires more time for formal assessment. Procedural expertise requires coaching, deliberate practice, and assessment of competence. While some of the learning can occur in simulation laboratories, some will need to occur in the clinical environment with a faculty supervisor. Other clinical competencies will require faculty who are trained to observe, assess, and intervene if necessary to create a continuum of education that is aligned with clinical goals. Trained faculty with time to implement competency-based education would be the foundation of an outcomes model. How will clinical faculty be able to negotiate the requirements for clinical productivity and the time needed to perform increasingly complex assessments of students to determine entrustment? The resolution of these competing interests will determine whether implementation of competency-based education and assessment with EPAs are possible. If we can demonstrate that investment in our educational programs will add value to our health care programs, we may have a better chance to resolve this conflict in favor of increased time for education.
A final impediment is the philosophical foundation of competency-based education, the idea that there is a set of domains that describe the roles or activities of physicians, and that we can identify, name, and measure them. When the Accreditation Council for Graduate Medical Education (ACGME) and the American Board of Medical Specialties (ABMS) proposed the 6 original competencies, some rejected the concept as unnecessary and too complex, and others accepted the concept but believed that one or another topic had been left out. For example, the Royal College of Physicians and Surgeons of Canada11 identified the roles of collaborator, medical expert, health advocate, and scholar, which were not specifically identified by the ACGME and the ABMS. Over time, other topics have been suggested, such as quality and safety, community orientation, teamwork, cross-cultural care, and humanism. As competency-based education evolves, there will be a continuous tension between the need to continually improve education with additions and revisions to the original core competency domains and the need to maintain a stable and manageable competency curricular framework.
I believe the key to overcoming the challenges I have identified is to demonstrate that improvements in education across the continuum using the competency framework will lead to improvements in health and health care value for the population. As we use education to improve quality of care and reduce cost, we will begin to bridge the divide between the clinical and educational silos. In addition, we should attempt to address the problems of inadequate time for the clinical assessment of competencies and EPAs. While allocation of increased educational resources using savings from clinical care would be helpful, we could also reduce the time needed for assessments by making the process less complex and time-consuming. Imagine if answering one question about a student or resident could provide the same quality of assessment information as does answering the 10 or more questions that currently often appear on student evaluation forms.
Recently, Ware et al12 have shown how assessment of a patient’s functional status can be reduced to answering one question. Why not use a similar approach to identify a more simple and effective assessment of a student? While advocating more time and payment for clinical education and assessment is reasonable, these goals are unlikely to be met immediately because of the current financial pressures on hospitals and physicians. But simplifying the assessment process and maintaining its accuracy could boost the acceptance of competency-based education. I suggest that we resist the temptations to add more competencies or EPAs and instead work toward reductions and simplification that may encourage faculty participation and open our eyes to a larger assessment horizon.
Such an approach returns us to the question Neruda posed in his rainbow poem. Will we struggle with the impossible task of finding where the rainbow touches the horizon? Or will we choose other ways to appreciate phenomena that are beautiful, ephemeral, and transitory—like the rainbow with its gradual changes of colors and like the remarkable transformation of our students to doctors?
On July 31, Deb Weinstein ended her term as deputy editor of Academic Medicine. Deb started as deputy editor in 2013 when I took over as editor-in-chief, and I have been grateful for her hard work these past 7 years. Deb’s commitment to excellence has contributed greatly to the journal’s success. Working with Deb has been a pleasure for the editorial team, journal staff, and me, and we all wish her the best as she takes on new challenges.
David P. Sklar, MD