What are the attributes of an excellent medical school? How important are medical students’ scores on national exams, match rates for residency programs, or students’ decisions to practice in underserved areas? What about research productivity of faculty or the quality of care provided in the institution’s clinical facilities? All of these factors and more are important measures of excellence, and different medical schools will set different priorities for these and other important activities based on the schools’ histories, values, resources, and the current needs within their communities.
U.S. News and World Report has for years used a few measures, such as reputation, research activity, student selectivity, and faculty resources, in its annual rating system of medical schools.1 McGaghie and Thompson2 strongly criticized that rating system on methodological, conceptual, and ethical grounds and maintained that
the assessment of quality should be based on criteria … that go beyond wealth and reputation. They should be measures that are proven to be directly related to graduating better doctors.
They suggested six approaches to evaluating medical schools’ quality: accreditation, impact on students (e.g., their orientation to service), prosperity (the school’s financial health), public service, reputation, and research. And they ask, “Given accreditation … what is a medical school’s value added for its graduates?”
As examples of value added, they suggested increased inclusion of underrepresented minority students in a medical school’s classes, improved assessment of clinical skills, and service to the underserved as categories of importance. Mullan et al3 suggested that medical schools could be ranked based on their accomplishments in meeting the social mission of training physicians to care for the population as a whole, taking into account issues such as primary care, underserved areas, and workforce diversity. How important is the production of a workforce that will meet the needs of the population compared with other important goals? How are we to choose between so many different ideas of what constitutes excellence in medical education?
One approach might be to look at the most serious problems in our health system and how educational programs contribute to their causes and solutions. Rising health care spending with poor-quality outcomes is one such overarching problem.4 Health care spending in the United States reached 17.8% of the gross domestic product in 2016, far higher than that in 10 other high-income countries,5 yet the life expectancy in the United States was the lowest of the 11 countries at 78.8 years, and infant mortality was the highest (5.8 deaths per 1,000 live births). Papanicolas et al5 concluded that the high health care spending in the United States is driven by high prices for labor, pharmaceuticals, and administrative costs. Health care spending in the United States is projected to rise at an average rate of 5.5% per year from 2017–2026 and grow one percentage point faster than the gross domestic product.6
These statistics should concern health professions educators because they train the health professionals whose decisions about tests, procedures, and pharmaceuticals are part of the difference between health care costs in the United States and other advanced countries.7 Efforts to train students about using best evidence to select cost-effective treatments have been disappointing according to Ioannidis,8 who noted that evidence-based medicine (integrating individual clinical expertise with the best external evidence) has not led to more effective, efficient care. He asked, “How likely is it that physicians will design studies whose results may threaten their jobs by suggesting that less procedures, testing, interventions are needed?” Ioannidis points out the intrinsic conflicts of interest of physicians and hospitals, who benefit from the dollars spent in health care even as those same dollars become unavailable to communities and individuals for other important needs such as housing, transportation, or food. What, then, might health professions educators and academic health centers do to better meet the needs of the public for a high-quality and less costly health system?
I can suggest a possible direction by sharing a story. Some time ago I was working with a patient—whom I will call Tom Dolan—who had end-stage liver disease and recurrent fluid accumulations leading to tense ascites. He made frequent visits to the emergency department, often resulting in admissions to the hospital. Although he had appointments in the outpatient liver disease clinic, he had trouble getting there on time because of transportation problems and would come to the emergency department when his discomfort became intolerable. In one year, Mr. Dolan accumulated over $100,000 in medical bills from various admissions and procedures; most of the care was unreimbursed.
One day I was talking to our hospital administrator about this patient and wondering whether we could provide more social support and perhaps transportation to the liver clinic for him. I also mentioned that we had several other patients who received frequent care in the emergency department for chronic problems that could be better managed in an outpatient clinic if we could only provide transportation and social services and get the cooperation of the clinic staff and doctors. I also mentioned the effect Mr. Dolan had on the emergency department staff, who felt helpless to solve his underlying problems.
Our conversation ultimately led to an analysis of data and an initiative focusing on the care of high-cost patients like Mr. Dolan.9 This program involved the analysis of hospital billing data to identify high-cost patients with chronic illnesses, an initial physician and nursing assessment to develop a care plan, priority access to specialty clinics, social services and nurse case management, psychological support, and connection to primary care teams. The program resulted in reductions in costs and emergency department visits. However, in spite of these successes over 10 years, the lessons from this initiative have not become integrated into the medical school’s educational programs. The social challenges and chronic conditions of our patients were not as dramatic as the stimulating cases that illustrated pathophysiology in the problem-based learning classes or as the clinical problems of patients seen on clerkships.
Blumenthal et al10 have described efforts similar to ours at other academic medical centers to identify and manage high-cost patients, have encouraged the adoption of such programs more broadly, and have urged integrating the study of these initiatives into educational programs. It is interesting that as important as high-cost patients are to the financial success of health care providers, health plans, and government programs, there is little known about the attributes of these patients or how our educational systems might serve them. I believe that while this approach to cost saving has so far been a lost opportunity, it illustrates some of the potential that a careful analysis of the distribution of spending in health care can provide to our redesign of health professions educational programs.
Mitchell11 has shown that in 2014, 1% of the noninstitutionalized U.S. patients accounted for 22.8% of expenditures, with a mean of $107,208. At the same time, the bottom 50% of the U.S. population accounted for only 2.8%, or $264 dollars on average, of total health care spending. These are huge differences with implications for population management: That is, by understanding the characteristics of high-cost patients, how to identify them before they become high-cost ones, and how to prevent them from experiencing the events that lead to high cost, we might be able to eliminate or mitigate serious clinical—and associated—cost problems. Even if we were not able to prevent the initial high-cost events, we might be able to prevent future high costs for the same patients.
At the same time, by understanding that 50% of the patients spend little on health care, we could design programs to help keep them healthy with preventive services while making sure that they did not get offered unnecessary care that could raise overall population health costs without providing benefits. The patients who were in the middle of the cost pyramid could also be analyzed to understand what factors might move them up or down in their spending patterns. Health professions education focuses on the diagnosis and treatment of high-cost patients but not on early recognition and prevention of conditions that can lead to high costs. What if excellence in health professions education recognized the importance of preparing a workforce that could recognize and change the trajectory of a serious illness rather than demonstrating procedural competence for treatments that might not have been needed?
Johnson et al12 identified a group of patients in Denver who were super users of health care, with an average annual spending of $113,522, and found them to cluster in diagnostic and social groupings such as patients with multiple chronic diseases, terminal cancer patients, recipients of emergency inpatient dialysis, trauma patients, individuals with serious mental health diagnoses, and orthopedic surgery patients. Recipients of emergency dialysis might not even need the dialysis on an emergency basis if financial and political barriers could be overcome; similar convenience and economy could certainly be achieved for at least some of the other groups. The approaches to reducing the spending for these patients would differ depending on the diagnostic/social group, and health professions education that fostered skills of data analysis and health policy advocacy might improve the quality of care and reduce the costs for segments of the population.
Dow et al13 have described a segmentation of their institution’s patient population based on cost and types of chronic conditions and the implications for the health professions workforce. Education of interprofessional teams to manage groups of patients could be a sign of care delivery and educational excellence. Lipstein et al14 have further developed recommendations for the health workforce, based on the perceived future needs of the population, that would involve the development of teams that could address both social and health-care-related problems. Meltzer and Ruhnke15 have described ambulatory physicians who manage high-risk, high-cost patients; these physicians could lead teams of caregivers. The skills needed by such teams would likely overlap with those needed in primary care and would provide a unique skill set to coordinate specialists, social care systems, nursing, and pharmacy services. Current payment models do not adequately reimburse physicians who spend time helping to avoid a hospitalization by arranging alternative approaches to care for complex patients like Mr. Dolan, but if payments shifted away from a fee-for-service model to a population management approach, there could be more financial support for the kind of physicians described by Meltzer and Ruhnke.
How might current health professions educational programs enlarge their perspective to address the needs of populations? While medical schools have different missions, with some focused on training clinician scientists and others oriented toward primary care for underserved populations, the acknowledgment of social accountability of medical education to address workforce needs would be an important step. Canada has adopted a social accountability goal for medical education. The Association of Faculties of Medicine16 noted the need for a national approach to physician resource planning and that
medical schools collaborate with the federal, provincial, and territorial governments and other national medical educational and licensing bodies to ensure the right number and mix of physicians entering practice.
By matching the population health needs with production of the right mix of health professionals who have the right set of skills, health care spending and quality of care could be vastly improved.
While the United States has struggled to find a consensus on health insurance and health care systems, the growing support for competency-based medical education might provide the opportunity to focus on population health management for medical students and residents. Gourevitch17 has described the way in which population health education and research could be integrated into an academic medical center to create a competency in population health management. However, adding more competencies to the growing list of competencies has drawbacks, as noted by Norman et al18 in their critique of competency-based medical education. The addition of any new competencies could create an unmanageable list that could dilute the importance of all of them and be unacceptable.
Another option would be to identify those organizations and projects that are already leading in developing population management programs and work with them to learn what knowledge and skills their teams require. This would encourage the development of educational programs, and the education would be specific to the clinical needs of specific populations. Bodenheimer and Berry-Millett19 provide examples of organizations and projects that are models of how to reorient the care delivery system to manage high-cost patients in an efficient, high-quality way. There are others, as noted by Blumenthal et al.10 Financial incentives that would encourage population segmentation and chronic care management, including the funding of interprofessional teams, could nurture the spread of innovative care models and the educational programs needed to support them. Education might begin with continuing professional development of practicing physicians and diffuse into the graduate and undergraduate medical systems as necessary prerequisites for employment.
We could also follow the lead of Gonzalo et al20 to establish health systems sciences as a third and coequal branch of medical education, joining basic sciences and clinical sciences as a foundational building block of medicine. In this way, the preparation of a health workforce interested in population health and trained to manage health spending would be a part of the education of every medical student and other health professions students. The success of the medical school and its rating for excellence would partly depend on the effectiveness of its education and care in health systems sciences, which would include population management. If we did this, the models of care that demonstrated success in meeting population health goals would be integrated into the curriculum. By better aligning the educational and care delivery goals, decisions about curricular change could be made in a logical and consistent manner.
While I am skeptical of medical school ratings, if such ratings are to have any value at all, they must include recognition of medical schools that train students and residents with population health skills and that motivate trainees to reduce health care spending, narrow disparities in care, improve health care quality, and provide access to care for our underserved populations. And while we are considering recognition of medical schools, we should include those that reduce or do not require tuition, such as New York University School of Medicine, which recently eliminated tuition. Lifting this great financial burden could influence medical students to more favorably consider practicing in underserved locations and choosing specialties most needed for improving the health of the public.