Secondary Logo

Journal Logo


A Case for Change

Disruption in Academic Medicine

Kahn, Marc J., MD, MBA; Maurer, Ralph, PhD; Wartman, Steven A., MD, PhD; Sachs, Benjamin P., MBBS, DPH

Author Information
doi: 10.1097/ACM.0000000000000418



In the article by Kahn and colleagues in the September 2014 issue, item 20 in the reference list was incorrect. The correct reference citation is presented below, with the corrections in italics.

For more information about the report cited in reference 20, please see “Benchmarks and Metrics” at the Association of Academic Health Centers’ Web site ().

Academic Medicine. 89(12):1663, December 2014.

Disruptive innovation is “a process by which a product or service takes root initially in simple applications at the bottom of a market and then relentlessly moves up market, eventually displacing established competitors.”1 In his seminal book The Innovator’s Dilemma, C.M. Christensen1 explains why incumbent firms usually miss the next wave of change: “They are both focused on organizational needs under current paradigms and have the overriding goal of driving efficiencies into their current operational processes.” This seemingly rational behavior actually slows adaptation and hinders competitive solutions. Eventually, it is too late to institute necessary changes, and these established but less agile businesses fail.

Innovation is also disrupting every sector of academic medicine, largely because of unsustainable business models, changing workforce needs, and the need to improve the quality of care. This includes academic health centers, with their three missions of leading biomedical and clinical research, educating the next generation of health professionals, and offering comprehensive and cutting-edge care.2 If academic health centers make the same mistakes Christensen warns against, they risk being overtaken by more adaptable enterprises and fading in their overall preeminence and impact.

Disruptions in Medical Education

Often overlooked, however, is the need for disruptive innovation in medical education, which has far-reaching implications—not just for the job prospects of future doctors but also for the long-term quality of health care services in the United States and globally.

An important driver of disruption in academic medicine is the doubling, roughly every five years, of medical information3 and the even faster evolution of medical technologies. How can physicians absorb and retain this vast amount of knowledge given time constraints and cognitive limits? How can they keep up with new tools and the new skills they require?

Are we in medicine teaching the next generation of physicians skills, or are we teaching them adaptive expertise?4 In this notion, skill is the ability to perform a delimited set of tasks, such as diagnosis and treatment. By expertise we mean the adaptive ability to both efficiently use past knowledge and experiences and innovatively create new knowledge and ideas in response to novel problems.5,6 If medical school consists mainly of acquiring skills, then physicians risk obsolescence as new technologies come into play. However, if physicians are taught expertise, they will be able to adapt in highly complex, dynamic, and uncertain environments. The key is that how we teach students is probably as, if not more, important than what we teach.

What will happen when computers can diagnose, recommend treatment, and—in the process—make fewer medical errors? Will computers that are also up-to-date and available 24/7 replace some providers? Today, IBM’s Watson has partnered with WellPoint Inc., which operates in 14 states, to eventually guide treatment decisions for its 34.2 million members.7 Watson is both an answer to the information dilemma and an example of a technology that will disrupt traditional medical diagnostics and, by extension, education.

If machines like Watson can handle basic diagnosis and treatment, then health care teams essentially become the human interface between patient and machine. They will be the front line for translating machine-generated recommendations into personalized care attuned to the specific context of a patient. The new modality will become both high-tech and high touch, requiring physicians to nimbly meld human intuition and emotional intelligence with the outputs of expert computer systems.

It is sobering to reflect that the class of 2016 will still be practicing medicine in 2050. Most physicians today are spend ing the majority of their practice treat ing chronic diseases such as diabetes, heart disease, cancer, and HIV.8 Treatment of chronic diseases makes interdisciplinary teamwork ex tremely important. It requires the co ordination of care between physicians, nurses, pharmacists, social workers, and other health care professionals.

Unfortunately, it is not clear where in modern medical education interdisciplinary teamwork is taught. Traditional lecture-based education is largely passive and an individual, rather than a team, effort. The Khan Academy has championed the use of the inverted classroom where students learn at home on the computer and come to class to ask questions and solidify what they have learned through interaction with both their teacher and peers.9 Another example is Just in Time Teaching, where students answer multiple-choice questions before class. After reviewing the students’ answers, the professor can efficiently focus classroom activities on areas in which the students are having the most difficulty.10 Team-based learning is an efficient educational method where students form teams to answer questions based on readings completed before class. These techniques, in which students apply, rather than just reiterate, knowledge can them help achieve adaptive expertise.

However, such an approach begs the question: How many, and what kind of faculty members do we really need to teach in this curricular model? In an inverted classroom model, the delivery of course material will raise the standard of education and produce economies of scale. Online lectures, texts, and multimedia presentations can be accessed simultaneously by thousands of students. Multidisciplinary teachers will be needed to facilitate students’ synthesis of material in a collaborative discussion-oriented environment. As a result, medical schools will require fewer teaching faculty and will be able to free up resources to invest in areas where hands-on teaching adds value and cost savings.

Disruptions in Clinical Care

Driven by the need to lower costs, and aided by new technologies, patient care is moving from the academic health center to the outpatient setting, community hospitals, and, ultimately, to wherever the patient happens to be. Clearly, these and other changes in the site and type of care will have a dramatic impact on academic health centers’ finances and on medical education.

Of note, megastores like Walmart are beginning to offer a host of services including primary care clinics staffed by physicians, nurse practitioners, dentists, and pharmacists.11 These innovations are already having a dramatic impact on the sites of primary care delivery. For the academic health center, these “docs in the box” are a direct threat to clinical revenue because of their lower cost of entry into the marketplace and lower fixed costs of practice.

The Affordable Care Act will eventually extend health care coverage to approximately 30 million more Americans. With limitations on physicians’ time and an unequal distribution of physicians’ services, especially in rural settings, telehealth is a technology that will disrupt the way medicine is practiced by simplifying and expanding access to specialty care while decreasing the use of hospital outpatient services. This is not surprising, because although physicians provide patients with information, expertise, and support, there is considerable pressure to see more patients in less time. An important question is how often direct physical doctor–patient interaction is necessary. As with any new modality, there are potential barriers to adoption of telehealth technology. These include state restrictions on reimbursement, issues with hospital privileges and interstate credentialing, malpractice fears, physician culture, internet bandwidth, and, most important, how to educate the workforce to effectively use telehealth.

What is the evidence that telehealth is worthwhile? It has been shown to reduce the cost of Medicaid in California and improve the quality of care for the state.12 However, most of the studies evaluating telehealth have been descriptive. In one major exception, the UK National Health Service used a multisite, cluster-randomized trial to study the effect of telehealth on use of secondary care and mortality.13 The setting was 179 general practices in three rural areas in England and included 3,230 people with diabetes, chronic obstructive pulmonary disease, or heart failure recruited between May 2008 and November 2009. Telehealth was associated with significantly lower mortality (4.6% versus 8.3%), lower rates of emergency admission, and shorter length of stay. However, in spite of these benefits, the study was unable to show overall cost savings.

Although today’s students and residents are very comfortable with technology, telemedicine provides unique educational challenges. Telemedicine does not provide complete access to the patient, particularly with respect to the tactile aspects of the physical exam, so trainees must be more comfortable with incomplete data and tolerant of ambiguity. As they consider their futures, they must also understand that fewer specialists may be needed. Trainees also need to be aware of the changing job description of the future physician practicing telehealth. More than a bedside clinician, a telephysician needs a different skill set than is presently taught in most medical schools. Finally, we need to teach our students to embrace change as this technology evolves. For example, haptic technology will make remote physical exams possible, and improvements in robotics may allow for surgeons to operate at great distance from their patients.

Transitioning care from the inpatient to the outpatient setting will disrupt both graduate and undergraduate medical education. Funding models will need to change, as graduate medical education is largely supported by inpatient care (direct and indirect medical expenses). As care moves from the inpatient setting, not only will hospital revenues be affected but so will graduate medical education. Medical students will need to be educated in ambulatory settings that are patient-centered medical homes and that fully integrate primary and specialty care. Many believe that we will face a shortage of primary care physicians because of the newly insured under the Affordable Care Act and the aging population. However, some disagree and argue that physicians can and should be replaced by other providers, nurses, physician assistants, and pharmacists.14,15 Nevertheless, from a medical educational perspective, perhaps the best way to increase primary care’s attractiveness for our students is to model the part of primary care that is most fun—namely, allowing students to follow patients over time.16 Whereas Flexner advocated for centralizing medical education in the hospital setting, the new paradigm is the integrated longitudinal clerkship. Only with integration across disciplines can team-based care, patient centeredness, and stewardship of limited resources be truly modeled and taught.17

Disruptions in Research

A final driver of disruption is the chang ing nature of research. In 2012, researchers at the University of California, Los Angeles, used an online game accessible by cell phones and personal computers to teach laypersons, including children, to diagnose malaria in infected red blood cells.18 This project allowed large numbers of public nonexperts to view digitally captured images of red cells and identify those infected with malaria. Because gamers are by nature competitive, the project eventually achieved a diagnostic accuracy rate of 98.75% when compared with expert pathologists who served as the reference group. Using similar principles, one can imagine using the public to quickly and efficiently analyze large data sets from research studies. Crowd sourcing allows many different people to collaborate without the need for direct physical interaction. Considering the high costs of clinical trials, including patient enrollment and data analysis, crowd- and cloud-sourced options could provide a more efficient and less expensive way to enroll larger numbers of patients and analyze larger amounts of data in less time.

The ability to collaborate without physical interaction and to do so with many people in real time will disrupt the definition and funding models of research. The traditional single principal investigator (PI) in a lab doing research may be a vanishing breed. External funding agencies such as the National Institutes of Health (NIH) are beginning to allow multiple PIs on projects, and grants are recognizing that the team-based research is important and perhaps even superior to the single-PI model.19 Who, though, when there are hundreds or thousands of contributors in crowd-sourced research, gets the credit? Research performed in such a fashion raises the question of study owner ship, integrity of the data, peer review, distribution of funding, and acknowledgment. Are medical schools ready for such changes?

It has been estimated that for every research grant dollar received by a medical school, the institution must spend an additional 26 to 40 cents to support that research.20 Given that most academic health centers must substantially support their research and teaching through revenues derived from ever-decreasing margins in clinical care, it is not hard to imagine that many will have to restructure their research enterprises. What is likely to happen is an increased differentiation amongst academic health centers, with the top 10% or 20% garnering a larger piece of the research pie, the lower 10% or 20% abandoning research, and the group in the middle forced to make real changes, including partnering with other centers.


Every mission of the academic health center is under threat of disruptive tech nologies and changing economics. Examples from business have taught us that companies that survive disruption do so by being agile, experimental, problem driven, and solution agnostic.21 As noted, academic health centers are likely to be more differentiated than they are today as they will increasingly emphasize one or two missions over others. A possible scenario is that many institutions will jump off the “NIH treadmill” and give up the increasingly futile attempt to catch those in the top 20%. These institutions’ research programs will be smaller and more highly focused on areas where they feel that they can make a real difference. Another potential scenario is that institutions will dramatically reduce the hours faculty spend on teaching as the education of health professionals becomes more “national,” taking advantage of increasingly sophisticated digitized platforms. And a possible scenario in the clinical arena suggests that academic health centers will form highly networked clinical partnerships with a wide variety of care providers, ranging from health systems and hospitals to long-term care providers. As a result, there will be a mix of cultures and priorities that will challenge the traditional ethos of academe. The academic health center of tomorrow may look nothing like today’s. This is probably a good thing if academic health centers are to survive the era of disruption.

To survive and prosper in the dynamic environment of the future, academic health centers will have to radically change. Only through careful strategic planning, balancing resources with mission, and overcoming resistance to change will academic health centers prosper in the era of disruptive innovation.


1. Christensen CM The Innovator’s Dilemma: When New Technologies Cause Great Firms to Fail. 1997 Boston, Mass Harvard Business School Press
2. Wartman SA. Toward a virtuous cycle: The changing face of academic health centers. Acad Med. 2008;83:797–799
3. Alper BS. How much does practice-guiding medical knowledge change in one year? Medicine 2.0. Accessed May 30, 2014
4. Cummings M. Director, Human and Automation Lab, Massachusetts Institute of Technology. Personal communication with Missy Cummings. September 2013
5. Mylopoulos M, Regehr G. How student models of expertise and innovation impact the development of adaptive expertise in medicine. Med Educ. 2009;43:127–132
6. Hatano G, Inagaki KStevenson H, Azuma H, Hakuta K. Two courses of expertise. In: Child Development and Education in Japan. 1986 New York, NY WH Freeman:27–36
7. IBM. What is Watson?. Accessed May 30, 2014
8. Tracy CS, Bell SH, Nickell LA, Charles J, Upshur RE. The IMPACT clinic: Innovative model of interprofessional primary care for elderly patients with complex health care needs. Can Fam Physician. 2013;59:e148–e155
9. Parslow GR. Commentary: The Khan academy and the day–night flipped classroom. Biochem Mol Biol Educ. 2012;40:337–338
10. Handelsman J, Ebert-May D, Beichner R, et al. Scientific teaching. Science. 2004;304:521–522
11. Pollert P, Dobberstein D, Wiisanen R. Jumping into the healthcare retail market: Our experience. Front Health Serv Manage. 2008;24:13–21
12. Nesbitt TS, Dharmar M, Katz-Bell J, Hartvigsen G, Marcin JP. Telehealth at UC Davis—a 20-year experience. Telemed J E Health. 2013;19:357–362
13. Steventon A, Bardsley M, Billings J, et al. Effect of telehealth on use of secondary care and mortality: Findings from the Whole System Demonstrator cluster randomized trial. BMJ. 2012;344:e3874
14. Chen PG, Mehrotra A, Auerbach DI. Response: Effectiveness in primary care is paramount, but need not come at the expense of efficiency. Med Care. 2014;52:99–100
15. Gottlieb S, Emanuel EJ. No, there won’t be a doctor shortage. New York Times. December 4, 2013 Accessed May 6, 2014
16. Bates J, Konkin J, Suddards C, Dobson S, Pratt D. Student perceptions of assessment and feedback in longitudinal integrated clerkships. Med Educ. 2013;47:362–374
17. Berwick DM, Finkelstein JA. Preparing medical students for the continual improvement of health and health care: Abraham Flexner and the new “public interest.” Acad Med. 2010;85(9 suppl):S56–S65
18. Luengo-Oroz MA, Arranz A, Frean J. Crowdsourcing malaria parasite quantification: An online game for analyzing images of infected thick blood smears. J Med Internet Res. 2012;14:e167
19. Kaye J, Hawkins N. Data sharing policy design for consortia: Challenges for sustainability. Genome Med. 2014;29:4
20. Association of American Medical Colleges. . Medical School Financial Data: Academic Expenses, Funding Sources, and Faculty Compensation Project. 2013 Washington, DC Association of American Medical Colleges
21. Wessel M, Christensen CM. Surviving disruption. Harv Bus Rev. 2012:56–64 December
© 2014 by the Association of American Medical Colleges