Secondary Logo

Journal Logo

Should Medical Students Track Former Patients in the Electronic Health Record? An Emerging Ethical Conflict

Brisson, Gregory E. MD; Neely, Kathy Johnson MD; Tyler, Patrick D. MD; Barnard, Cynthia MBA, MSJS

doi: 10.1097/ACM.0000000000000633
Perspectives
Free

Medical students are increasingly using electronic health records (EHRs) in clerkships, and medical educators should seek opportunities to use this new technology to improve training. One such opportunity is the ability to “track” former patients in the EHR, defined as following up on patients in the EHR for educational purposes for a defined period of time after they have left one’s direct care. This activity offers great promise in clinical training by enabling students to audit their diagnostic impressions and follow the clinical history of illness in a manner not possible in the era of paper charting. However, tracking raises important questions about the ethical use of protected health information, including concerns about compromising patient autonomy, resulting in a conflict between medical education and patient privacy. The authors offer critical analysis of arguments on both sides and discuss strategies to balance the ethical conflict by optimizing outcomes and mitigating harms. They observe that tracking improves training, thus offering long-lasting benefits to society, and is supported by the principle of distributive justice. They conclude that students should be permitted to track for educational purposes, but only with defined limits to safeguard patient autonomy, including obtaining permission from patients, having legitimate educational intent, and self-restricting review of records to those essential for training. Lastly, the authors observe that this conflict will become increasingly important with completion of the planned Nationwide Health Information Network and emphasize the need for national guidelines on tracking patients in an ethically appropriate manner.

G.E. Brisson is assistant professor in clinical medicine, Feinberg School of Medicine, Northwestern University, Chicago, Illinois.

K.J. Neely is assistant professor in medicine, Feinberg School of Medicine, Northwestern University, and chair, Medical Ethics Committee, Northwestern Memorial Hospital, Chicago, Illinois.

P.D. Tyler was a final-year medical student, Feinberg School of Medicine, Northwestern University, Chicago, Illinois, at the time this article was written. He is now a first-year resident, Beth Israel Deaconess Medical Center, Boston, Massachusetts.

C. Barnard is director of quality strategies, Northwestern Memorial Hospital, and research associate professor, Feinberg School of Medicine, Northwestern University, Chicago, Illinois.

Editor’s Note: A Commentary by K. McLaughlin and S. Coderre appears on pages 1007–1009.

Funding/Support: None reported.

Other disclosures: None reported.

Ethical approval: Reported as not applicable.

Correspondence should be addressed to Gregory E. Brisson, 1333 W. Belmont Ave., Suite 200, Chicago, IL 60657; telephone: (312) 926-3627; e-mail: gbrisson@nm.org.

In 2009, the U.S. Congress allocated funds to incentivize the development and implementation of electronic health records (EHRs).1 As a result, EHRs are rapidly being adopted in academic health centers as well as in private practice settings, and medical students are increasingly using them during clinical clerkships.2–4 This technology has the potential to transform the way students learn, though few studies have examined its impact on medical education, and published opinions have been mixed.5–9 The primary function of EHRs is to improve patient care, and research on EHRs has focused on this area.10–12 However, the Institute of Medicine has recognized several secondary functions of the EHR, including research, public health, and education.10 Now that the EHR has been widely adopted, it is appropriate to explore these secondary uses. And, though the effect of EHRs on education is presently a matter of debate, there is general agreement among medical educators that we should seek opportunities to use this technology to improve clinical training.2,5,6

Back to Top | Article Outline

Tracking Former Patients in the EHR

One such opportunity involves using the EHR to track former patients in the medical system. Students have always had the ability to follow up on patients after discharge or after completion of a clerkship, but doing so was cumbersome in the era of paper charts. EHRs greatly facilitate this process, providing students ongoing access to physician notes, lab and imaging results, and often outpatient clinic notes, long after they have concluded their direct relationship with a patient. Longitudinal access enables students to observe the history of disease in a manner previously not possible and offers an immediate means of professional self-education. As Adibe and Jain5 observe, the EHR empowers students to “audit their diagnostic impressions and management”; they recommend that educators embrace this technology and suggest that students could track a panel of complex patients “throughout and beyond a given rotation.”5

Of note, there is no standard definition of “tracking” patients. In this discussion, we define it as the action of following up on a former patient in the EHR for educational purposes for a defined period of time after the patient has left one’s direct care, where that patient is someone with whom one has had a meaningful clinical relationship.

Back to Top | Article Outline

Little is known about tracking by medical students

The educational potential of tracking is apparent, though little has been written about its application in training. Some EHRs allow users to flag patient records to their own personal “teaching file,” and as Morrow and Dobbie suggested in an article by Peled et al,8 this feature could be used in the future to query the records of all patients one has cared for in the past. Dhaliwal and Detsky13 speculate that the master diagnostician of the future will use the EHR to measure his or her diagnostic error rate, track outcomes, and examine mistakes “to understand if they were ‘no fault’ or borne of an oversight, knowledge gap, or cognitive misstep.”

Although these authors offer a promising vision of clinical medicine, one need not wait until the future to start tracking patients. We know that some students at our institution already perform this activity, and survey data suggest that tracking may be widespread (G.B., P.D.T., unpublished data, 2014). There is informal evidence that this practice also occurs at other institutions. For example, during an electronic forum in which emergency medicine physicians and residents responded to a student’s question about following up on “interesting” patients, multiple participants reassured the student that the EHR enabled follow-up on such patients.14 One observed, “My curiosity is well sated by following their electronic medical records after admission.” Another stated, “I typically follow up on all critically ill, interesting, or the rare nice patient that I see. In general it’s QI [quality improvement] for myself to see what I missed, what I could have done differently, and what I did right.”

Back to Top | Article Outline

Concerns related to tracking

Interestingly, participants in this forum expressed uncertainty about the appropriateness of tracking even as they continue to do it. For example, when a participant questioned whether tracking violated the Health Insurance Portability and Accountability Act (HIPAA), one responded, “Don’t know for sure, but I seriously doubt it.” Another added, “It is generally allowed for educational/QA [quality assurance] purposes to continue to follow a case you were part of (but now no longer are) however some (stupid, short-sighted) institutions have stricter limitations than what HIPAA allows.”

A first-year resident, blogging under a pseudonym, observed that some medical students track patients and described it as a “powerful example of self-motivated, case-based learning.”15 In a posting titled “How Hospital Privacy Policies Are Inhibiting Medical Education,” the blogger also commented that tracking likely violates the privacy policies of many institutions and made a plea to allow students to track patients “free from fear of medical school expulsion or other disciplinary measures.”

The blogger makes an eloquent case for the merits of tracking from a student’s perspective, but it is not clear that this individual appreciates the full implications of this activity. Ethical and legal concerns associated with this action must be addressed to assure patients that their health records are being handled appropriately. However, on the basis of our observations, many academic health centers do not have a policy to address this activity (ours does not), and those that do offer widely varying approaches, ranging from permission to prohibition.

Back to Top | Article Outline

Ethical Considerations of Tracking

The ambiguous status of tracking prompts an important question—is it ethically and legally acceptable to track former patients in the EHR? Our review of this topic indicates that, although it is legally permissible to track patients for educational purposes, as both an improvement activity and a competency assurance activity under the concept of “health care operations” described in the HIPAA Privacy Rule,16 the law is insufficient to address relevant ethical considerations. Physicians therefore have a professional responsibility to ensure that tracking is conducted in an ethically appropriate manner. However, there is a paucity of guidance on this topic in the medical literature. Numerous articles address concerns about privacy, confidentiality, and autonomy—standards at the foundation of medical ethics which face new challenges in the era of EHRs—but they do not address the trade-off of breaching patient privacy for medical education, as occurs when students track patients.17–20

Compelling arguments can be made on both sides of this question. We examine these arguments below.

Back to Top | Article Outline

Arguments in support of tracking

In support, one could argue that educators have an obligation to employ tracking in medical education because it improves training and therefore offers long-lasting benefits to society. Redelmeier21 has commented that diligent follow-up offers an “opportunity to learn from mistakes for the benefit of future patients.” This argument is supported by utilitarian theory, but it is also based in beneficence because, as explained by Beauchamp and Childress,22 “the theory sees morality in terms of the legitimate goal of promoting welfare.”

If one accepts as self-evident that improved training benefits society, then the soundness of this argument hinges on the premise that tracking improves training. This premise is supported by analogy to case-based learning, a teaching method which, similar to tracking, links theory to practice through the application of knowledge to authentic cases—a process that students believe improves training.23 It is also supported by precedent: The value of patient-based learning has been recognized since the origins of modern medical education. Sir William Osler is often quoted as having said, “He who studies medicine without books sails an uncharted sea, but he who studies medicine without patients does not go to sea at all.”24 More recently, Prober and Heath24 stated that patients’ stories “make the acquisition of knowledge compelling” and “serve as the scaffolding on which facts and concepts are organized and reinforced.” Moreover, ethicist Kenneth Goodman25 has written that a physician who declined to learn from experience, citing patient privacy as the reason, would be “medically and morally fatuous,” and that society “would regard such a physician as unfit for practice.”

Tracking could also be supported by the ethical principle of distributive justice, which describes a duty to advance the fair, equitable, and appropriate distribution of benefits and burdens within a community—acknowledging that such a commitment may conflict with autonomy.22 Indeed, what is fair for most may encroach on the preferences of an individual, and in medicine we sometimes determine that maximizing benefit to the group should prevail over the wishes of the individual. For example, most hospitals have policies that ban smoking. In applying the principle of distributive justice to tracking, then, one could reasonably argue that it is fair to permit tracking, despite attendant burdens (such as risk to privacy), given the benefits to society of enhancing the training of physicians.

Back to Top | Article Outline

Arguments in opposition to tracking

In opposition, one could argue that tracking compromises patient autonomy, as expressed through the rule of privacy, and students should therefore not be permitted to do it. Ethicist Eike-Henner Kluge20 has written that universally accepted ethical principles, such as respect for autonomy, applied to the context of EHRs, guarantee all patients a “fundamental right to privacy, and … control over the collection, use and disposition of data about themselves.” Although this right is conditioned by the legitimate needs of society and the “equal and competing rights of others,” Kluge suggests that most nonconsensual secondary uses of the EHR do not meet this standard.26 Given this assertion, one might reasonably conclude that students should not be granted longitudinal access to the evolving story of a patient’s health and life, as contained in the EHR. Moreover, ethicist Mark Rothstein27 warns of potential harms resulting from loss of privacy: “If restrictions are not imposed, then too many providers increasingly will have access to too much information,” resulting in “patients’ anxiety … and perhaps a reluctance to seek care.”

Additionally, one could argue that tracking patients is not necessary to train a competent physician—that any educational value gained is incremental compared with training before EHRs. The validity of this premise is debatable, considering that there are no published data on the impact of tracking on education. However, it is supported by literature which is inconclusive in its assessment of the value of case-based learning.23,28 It is also supported by historical precedent: Modern medical education has produced generations of exceptional physicians in the era preceding EHRs—many of them are our respected teachers and mentors. Therefore, in the absence of clear and substantial added benefit to education, one could reasonably conclude that, according to the principle of nonmaleficence, students should not perform this action because it is potentially harmful to patients.22

Back to Top | Article Outline

Other considerations

Whatever our own positions may be, it is essential that we consider our patients’ perspective. There are few data to guide us in this respect, though it is reasonable to expect that patients would be divided on this issue. Multiple surveys demonstrate broad public concern about privacy and the use of EHRs, including improper access to protected health information by medical staff.29–31 These data suggest that some patients would object to tracking, even if they acknowledge its potential benefits to society.

Back to Top | Article Outline

Emergence of an ethical conflict

Thus, there is an emerging ethical conflict between medical education and patient privacy which is an unintended consequence of secondary use of EHRs for learning in addition to their primary use in patient care. Not surprisingly, innovative application of this technology has outpaced ethical considerations of its use. There are several possible reasons why this has occurred. First, it may simply be that educators are unaware that some students perform this activity—there are no published data on tracking. Research is needed to define the prevalence and logistics of this activity. Second, educators may have been so focused on promoting access to the EHR that we have overlooked pitfalls associated with its use. Many academic health centers have been slow to provide EHR access to students, and therefore educators have spent the initial years after EHR implementation advocating for student access.2–4,32 As Spriggs et al17 observe, the debate about ethical issues related to EHRs has, in general, been overshadowed by technical priorities and security issues. A third possibility is that educational benefits resulting from tracking so closely align with our teaching mission that our recognition of ethical pitfalls has been obscured. Lack of attention to this issue may therefore represent a blind spot on the part of educators focused on training.

Understanding why this conflict has escaped scrutiny is less important than recognizing that it exists and needs to be resolved so that we can optimize the use of the EHR in clinical training.

Back to Top | Article Outline

Balancing Competing Obligations and Future Directions

As discussed above, there are compelling arguments on both sides of this ethical conflict, and it is the work of physicians to balance competing obligations to optimize outcomes and mitigate harms. Tracking poses two sets of competing obligations: (1) the duty to put patients’ interests first and, simultaneously, train exceptional physicians; and (2) the duty to care for patient populations as well as individuals. On balance, we conclude that the potential benefits of tracking—to patients and students—outweigh potential harms, and students should therefore be permitted to perform this activity. Furthermore, we believe tracking will promote professional behaviors that can be challenging to teach or measure. For example, longitudinal follow-up reinforces the power of narrative in medicine and encourages curiosity, a professional virtue which can improve clinical reasoning and generate empathy.33,34 It also fosters professionalism by tempering the self-focus of medical students and promoting patient-centeredness. Thus, tracking can augment training of competencies which are highly valued by patients. Conversely, we acknowledge doubts about the measurable educational value added through tracking and stipulate that research is necessary to explore methods to best use this technology in clinical training. However, this action should not be without limits.

Back to Top | Article Outline

Limits to tracking former patients in the EHR

Despite its considerable benefits, tracking must be balanced with respect for patient autonomy. This would be effectively achieved by providing patients the option to permit or refuse access to their records by students, thereby eliminating the need to make an absolute choice between education and privacy.

Furthermore, our obligation does not end with permission—tracking is a privilege that demands proper stewardship. It is widely accepted that infringement on privacy rights of patients should “occur in the least intrusive fashion and with a minimum of interference with the rights of the affected person.”20 Consequently, even with permission, tracking does not confer unlimited access to health records—students must have legitimate educational intent and must self-restrict review of records to those essential for training. Given this substantial responsibility, one could return the argument against tracking with the assertion that students lack the requisite experience to judiciously review records and should therefore not perform this action—or, at least, not without supervision.

Although we acknowledge this concern, there is precedent that supports independent tracking. Fundamentally, all practice of medicine involves access to information that patients value as private, and even the most ancient medical codes include a commitment to use this information only as needed to heal; the rule of confidentiality and, more recently, HIPAA protections further enshrine the doctor–patient relationship.35 These fiduciary responsibilities extend into the educational setting, where patients generously share private information with students whose clinical contribution may be modest or even nonexistent (but who are nonetheless bound by professional duty). In the digital age, these responsibilities also extend into the realm of health records—information in the EHR originates from a confidential encounter and thereby acts as a “patient-analogue,” giving rise to its protected status.36 By this reasoning, accessing the EHR acts as an analogue of a clinical encounter, and information acquired through this activity is similarly protected by professional duty. Applied to the question of tracking, then, there is precedent in education that supports it: We trust properly trained students to examine patients independently; therefore, we can similarly trust them to review health records of those patients. As Kluge37 has reasoned, just as patients may “grant access to their bodies to individuals of their choice” (as during clinical training), patients may “grant others permission to access their record.”

Back to Top | Article Outline

Future directions

Just as students receive training before being granted the privilege of examining patients, they should receive training on tracking before being permitted to do it. Training could be embedded in the curriculum; for example, discussions of tracking could supplement existing ethics and professionalism training by offering concrete examples of concepts that define the doctor–patient relationship, including autonomy, confidentiality, privacy, and fidelity. And, similar to clinical skills training, students could begin tracking under supervision and expand their privilege in a graded fashion based on demonstration of competency.38

We propose that students (indeed, all clinicians) would benefit from guidelines that preserve the unique educational benefits of tracking while safeguarding privacy and patient choice. Development of guidelines is beyond the scope of this paper; however, we emphasize that they should be generated from a patient-centered perspective as it is widely accepted that EHRs must always be managed in such a way as to minimize harm to patients.19,27,37,39,40 Additionally, we recommend that guidelines should achieve national consensus in order to maintain public trust in the medical profession. It is particularly important to achieve consensus in advance of the planned Nationwide Health Information Network, which will greatly expand access to protected health information and increase the potential for harm.41 As Spriggs et al17 have stated, when it relates to matters of privacy and the EHR, “confidence in the ethics of the system is ‘mission critical.’”

Back to Top | Article Outline

Conclusion

Some medical students track patients in the EHR, and, though the prevalence is unknown, this activity will likely occur with increasing frequency given widespread adoption of EHRs. Both patients and students benefit from tracking. However, there is an inherent ethical conflict between medical education and patient privacy, and ambiguity regarding the appropriateness of tracking has likely hindered its evolution as a learning tool. After considering ethical arguments, we conclude that students should be permitted to track patients for education, but with defined limits: Students should obtain permission from patients and should receive training to ensure that they grasp the privilege and responsibility associated with tracking. Students, patients, and educators would benefit from an open discussion on this issue, and guidelines should be developed on how to conduct this activity in a manner that optimizes benefits and safeguards privacy and patient choice. Thus, we can maintain public trust in the profession and demonstrate that a commitment to medical education can thrive in conjunction with dedication to placing patients’ interests first.

Acknowledgments: The authors wish to thank members of the Northwestern Memorial Hospital medical ethics committee, especially Mike Shapiro, MD, Andrew Naidech, MD, Joan Anzia, MD, and Katie Watson, JD, for vigorous discussion and debate which informed the work leading to this paper. They also wish to thank R. Bernard Mims, MSHI, CHC, CHRC, chief privacy officer, Northwestern Memorial HealthCare, for clarifications related to HIPAA privacy regulations. Additionally, the authors would like to thank Rob Hirschtick, MD, Dave Neely, MD, Lyle Berkowitz, MD, and David Liebovitz, MD, for discussion of these topics.

Back to Top | Article Outline

References

1. American Recovery and Reinvestment Act of 2009. . HR 1, 111th Congress, 1st Session, Title XIII, sec. 3001 6(E). 2009 http://www.govtrack.us/congress/bills/111/hr1/text. Accessed November 13, 2014
2. Hammoud MM, Margo K, Christner JG, Fisher J, Fischer SH, Pangaro LN. Opportunities and challenges in integrating electronic health records into undergraduate medical education: A national survey of clerkship directors. Teach Learn Med. 2012;24:219–224
3. Mintz M, Narvarte HJ, O’Brien KE, Papp KK, Thomas M, Durning SJ. Use of electronic medical records by physicians and students in academic internal medicine settings. Acad Med. 2009;84:1698–1704
4. Hammoud MM, Dalymple JL, Christner JG, et al. Medical student documentation in electronic health records: A collaborative statement from the Alliance for Clinical Education. Teach Learn Med. 2012;24:257–266
5. Adibe BA, Jain SH. Electronic health records: Potential to transform medical education. Am J Manag Care. 2010;16(12 suppl HIT):SP62–SP63
6. Elliott K, Judd T, McColl G. A student-centred electronic health record system for clinical education. Stud Health Technol Inform. 2011;168:57–64
7. Tierney MJ, Pageler NM, Kahana M, Pantaleoni JL, Longhurst CA. Medical education in the electronic medical record (EMR) era: Benefits, challenges, and future directions. Acad Med. 2013;88:748–752
8. Peled JU, Sagher O, Morrow JB, Dobbie AE. Do electronic health records help or hinder medical education? PLoS Med. 2009;6:e1000069
9. Schenarts PJ, Schenarts KD. Educational impact of the electronic medical record. J Surg Educ. 2012;69:105–112
10. Institute of Medicine, Committee on Data Standards for Patient Safety, Board on Health Care Services. Key Capabilities of an Electronic Health Record System: Letter Report. 2003 Washington, DC National Academies Press
11. Hillestad R, Bigelow J, Bower A, et al. Can electronic medical record systems transform health care? Potential health benefits, savings, and costs. Health Aff (Millwood). 2005;24:1103–1117
12. Menachemi N, Collum TH. Benefits and drawbacks of electronic health record systems. Risk Manag Healthc Policy. 2011;4:47–55
13. Dhaliwal G, Detsky AS. The evolution of the master diagnostician. JAMA. 2013;310:579–580
14. Student Doctor Network 2011. . Physician/Resident Forums, Emergency Medicine Forum 12/12/11. http://forums.studentdoctor.net/showthread.php?p=11908212. Accessed November 13, 2014
15. EyeSteve. . How hospital privacy policies are inhibiting medical education [blog posting]. 2013 http://www.eyesteve.com/privacy-policies/. Accessed November 13, 2014
16. OCR. . Summary of the HIPAA Privacy Rule. 2013 http://www.hhs.gov/ocr/privacy/hipaa/understanding/summary/privacysummary.pdf. Accessed November 13, 2014
17. Spriggs M, Arnold MV, Pearce CM, Fry C. Ethical questions must be considered for electronic health records. J Med Ethics. 2012;38:535–539
18. Wynia M, Dunn K. Dreams and nightmares: Practical and ethical issues for patients and physicians using personal health records. J Law Med Ethics. 2010;38:64–73
19. Layman EJ. Ethical issues and the electronic health record. Health Care Manag (Frederick). 2008;27:165–176
20. Kluge EH. Professional codes for electronic HC record protection: Ethical, legal, economic and structural issues. Int J Med Inform. 2000;60:85–96
21. Redelmeier DA. Improving patient care. The cognitive psychology of missed diagnoses. Ann Intern Med. 2005;142:115–120
22. Beauchamp TL, Childress JF Principles of Biomedical Ethics. 20097th ed New York, NY Oxford University Press
23. Thistlethwaite JE, Davies D, Ekeocha S, et al. The effectiveness of case-based learning in health professional education. A BEME systematic review: BEME guide no. 23. Med Teach. 2012;34:e421–e444
24. Prober CG, Heath C. Lecture halls without lectures—a proposal for medical education. N Engl J Med. 2012;366:1657–1659
25. Goodman KW. Ethics, information technology, and public health: New challenges for the clinician–patient relationship. J Law Med Ethics. 2010;38:58–63
26. Kluge EH. Informed consent to the secondary use of EHRs: Informatic rights and their limitations. Stud Health Technol Inform. 2004;107(pt 1):635–638
27. Rothstein MA. The Hippocratic bargain and health information technology. J Law Med Ethics. 2010;38:7–13
28. Williams B. Case based learning—a review of the literature: Is there scope for this educational paradigm in prehospital education? Emerg Med J. 2005;22:577–581
29. HarrisInteractive. . Survey on Medical Privacy. 2004 http://patientprivacyrights.org/media-center/polls/#July2004Harris. Accessed November 19, 2014
30. Bishop L, Holmes B, Kelley C. National Consumer Health Privacy Survey 2005. 2005 http://www.chcf.org/~/media/MEDIA%20LIBRARY%20Files/PDF/C/PDF%20ConsumerPrivacy2005ExecSum.pdf. Accessed November 13, 2014
31. Sailpoint. . The SailPoint Market Pulse Survey. 2011 http://assets.fiercemarkets.com/public/newsletter/fierceemr/sailpoint.pdf. Accessed November 13, 2014
32. Gliatto P, Masters P, Karani R. Medical student documentation in the medical record: Is it a liability? Mt Sinai J Med. 2009;76:357–364
33. Dyche L, Epstein RM. Curiosity and medical education. Med Educ. 2011;45:663–668
34. Fitzgerald FT. Curiosity. Ann Intern Med. 1999;130:70–72
35. Lucia SP. The lure of medical history: Invocatio medici: Code of Fushi Ikai No Ryaku, Oath of Hippocrates, and Supplication of Maimonides. Cal West Med. 1929;30:117–120
36. Kluge EH. Security and privacy of EHR systems—ethical, social and legal requirements. Stud Health Technol Inform. 2003;96:121–127
37. Kluge EH The Ethics of Electronic Patient Records. 2001 New York, NY Peter Lang Publishing
38. Halpern SD, Detsky AS. Graded autonomy in medical education—managing things that go bump in the night. N Engl J Med. 2014;370:1086–1089
39. Hoffman S, Podgurski A. Finding a cure: The case for regulation and oversight of electronic health record systems. Harv J Law Technol. 2008;22(1):54–55
40. Mandl KD, Szolovits P, Kohane IS. Public standards and patients’ control: How to keep electronic medical records accessible but private. BMJ. 2001;322:283–287
41. Dobalian A, Claver ML, Pevnick JM, Stutman HR, Tomines A, Fu P Jr.. Organizational challenges in developing one of the Nationwide Health Information Network trial implementation awardees. J Med Syst. 2012;36:933–940
© 2015 by the Association of American Medical Colleges