Share this article on:

“No One Has Yet Properly Articulated What We Are Trying to Achieve”: A Discourse Analysis of Interviews With Revalidation Policy Leaders in the United Kingdom

Archer, Julian MD, PhD; Regan de Bere, Sam PhD; Nunn, Suzanne PhD; Clark, Jonathan MA; Corrigan, Oonagh PhD

doi: 10.1097/ACM.0000000000000464
Research Reports

Purpose To analyze prevailing definitions of revalidation (i.e., a recently instituted system of ongoing review for all physicians in the United Kingdom), the circumstances of their origin, and proposed applications, after a protracted and sometimes difficult decade in development. This was to support a more consensual approach to revalidation policy before its launch in 2012.

Method In 2010 and 2011, the authors carried out a critical discourse analysis of interviews with 31 medical and legal revalidation policy makers. These individuals represented the main stakeholder bodies, including the General Medical Council, Academy of Medical Royal Colleges, British Medical Association, National Health Service Employers, and the departments of health from across the United Kingdom.

Results The authors identified two overarching discourses: regulation and professionalism, held together by patients as “discursive glue.” Regulation frames revalidation as a way to identify “bad apples,” requiring a summative approach and minimum standards. Professionalism looks to revalidation as a process by which all doctors improve, requiring evolving standards and a developmental model.

Conclusions These two discourses were not mutually exclusive; indeed, most interviewees used them interchangeably. However, they are in some regards at odds. Their coexistence has been supported by a shared discursive formation around patients. Yet the authors found little patient-centered policy in revalidation in its current form. The authors concluded that patients need to be recognized, making them present with an active voice. They also stressed the importance of established and ongoing evaluation of medical regulation as a policy and process.

Dr. Archer is National Institute for Health Research Career Development Fellow, senior clinical lecturer, and director, Collaboration for the Advancement of Medical Education Research and Assessment, Plymouth University Peninsula Schools of Medicine and Dentistry, Plymouth, Devon, United Kingdom.

Dr. Regan de Bere is lead for medical humanities and deputy director, Collaboration for the Advancement of Medical Education Research and Assessment, Plymouth University Peninsula Schools of Medicine and Dentistry, Plymouth, Devon, United Kingdom.

Dr. Nunn is research fellow, Collaboration for the Advancement of Medical Education Research and Assessment, Plymouth University Peninsula Schools of Medicine and Dentistry, Plymouth, Devon, United Kingdom.

Mr. Clark is lecturer in sociology, Plymouth University, Plymouth, Devon, United Kingdom.

Dr. Corrigan is honorary fellow, Plymouth University Peninsula Schools of Medicine and Dentistry, Plymouth, Devon, United Kingdom.

Funding/Support: This report is the result of independent research funded by the Health Foundation, an independent charity working to continuously improve the quality of health care in the United Kingdom.

Other disclosures: None reported.

Ethical approval: Ethical approval was awarded by the Peninsula College of Medicine & Dentistry research ethics committee. Participants gave informed consent before taking part.

Disclaimer: The views expressed in this report are those of the participants and the authors and do not reflect those of the Health Foundation.

Previous presentations: An overview of this report was presented by Dr. Archer on October 4, 2012 at the International Association for Medical Regulators and Authorities Conference 2012, Ottawa, Ontario, Canada.

Correspondence should be addressed to Dr. Archer, Plymouth University Peninsula Schools of Medicine and Dentistry, C521 Portland Square, Plymouth University, Drake Circus, Plymouth, United Kingdom PL4 8AA; telephone: +44 (0) 1752-586750; e-mail: julian.archer@plymouth.ac.uk.

Many countries are trying to improve patient care by encouraging individual doctors to maintain their clinical knowledge, skills, and professional behaviors through ongoing assessment. In the United States, maintenance of certification requires most certified physicians to seek recertification periodically (mostly every 10 years).1 Engagement in continuing professional development (CPD) is now compulsory in Australia,2 the majority of Canadian jurisdictions,3 and New Zealand,4 where CPD is explicitly linked to annual recertification. But such programs are not without criticism.5

In the United Kingdom, a system of ongoing review called revalidation was launched on December 3, 2012. Revalidation asks the doctors who are being evaluated to collect and reflect on supporting documentation as part of an annual appraisal. These include the doctors’ self-reported CPD, a practice review such as a clinical audit, significant events, complaints and compliments, and, for each five-year revalidation cycle, feedback from colleagues and patients.6

However, revalidation is unique in a number of important ways. First, revalidation is the only system in the world in which all doctors must successfully take part so they can retain their licenses to practice. This includes all specialists, family physicians, and doctors in training (i.e., residents and fellows). Second, revalidation was informed by the Royal Colleges, the United Kingdom’s professional medical bodies, but has been designed and implemented by the United Kingdom’s national regulator, the General Medical Council (GMC), whose council are half lay members. Third, decision making is initially carried out locally, structured around a preexisting National Health Service (NHS) appraisal process. Documentation feeds into annual appraisals. After five years, the responsible officer, who is the most senior doctor in the region, recommends to the GMC whether a doctor’s license should be revalidated.

As a policy, revalidation has suffered a long and checkered history. It has been under active discussion since 1998,7 but the idea was first raised in 1972.8 Revalidation’s development over time has polarized the medical community, placing strain on the relationship between government and the medical profession.

In the rest of this report, we describe the first of three major studies concerning revalidation, part of a program funded predominately by the National Institute for Health Research. The second and third studies, currently under way, explore the impact of revalidation on doctors in practice and the role and relationships of revalidation to patients and the public.9 In this report, we focus on our independent evaluation of the controversies and events that have beleaguered the progress of revalidation. We also shed light on the role of policy makers’ and senior decision makers’ agendas by using interview data from such individuals to analyze the discourses shaping the conditions within which revalidation has been implemented. By understanding these discourses, we sought to develop new information that can be translated into recommendations for policy, education, and practice.

Back to Top | Article Outline

Method

In 2011, using a critical discourse analysis of interviews with 31 leading medical and legal revalidation policy makers since 1998, we analyzed prevailing definitions of revalidation, the circumstances of their origin, and proposed applications in order to support a more consensual approach to revalidation policy.

Back to Top | Article Outline

Data collection

Using a list generated from the authors of key policy documents, we identified 44 policy makers in key revalidation policy positions, including leaders past and present. Ethical approval was awarded by the research ethics committee of the Peninsula College of Medicine & Dentistry (the former name of our institution). Thirty-one of the 44 policy makers whom we approached by letter consented in writing to be interviewed. We interviewed them between June 2010 and July 2011, allowing us to speak to members of all the main stakeholder bodies such as the GMC, the Academy of Medical Royal Colleges (AoMRC), the British Medical Association (BMA), the NHS Employers (an agency of the NHS), and the departments of health from across the United Kingdom.

Back to Top | Article Outline

Interviews and analysis

Interviews were audiotaped and transcribed and managed by the NVivo 8 qualitative data package (QSR International, Melbourne, Australia). To index the large and complex dataset, two of us (J.C., S.N.) initially coded the data to nine generic themes (e.g., challenges, identity). Once the data were indexed, all of us conducted the discursive analysis using the following three complementary analytical approaches.

First, we applied Foucault’s genealogical analysis10 to understand the continuities and discontinuities that have characterized the history of revalidation. Academics use genealogical methods to challenge the apparent “self-evidence” of prevailing discourses, by revealing how policies and practices are historically contingent and open to change.10 This helped us to understand how policy now has been shaped by events then 11 and to understand what lies beneath complex changes in governmentality.12 We undertook an empirical investigation of governmentality in terms of “what was said and how it was said” across time.13

Second, we drew on Torfing14 for a political analysis of revalidation as a discursive system that is formed by lines of similarity and difference and is played out through identifiable discursive practices. We conducted a systematic analysis of the discourses15 that have shaped revalidation strategies (objects and activities) over time. This approach is political. We explored how power is exercised within the system of revalidation, in this case within the profession of medicine. The analysis focused on the ways in which professional discourses have been used to support or negate particular ideas or practices.

Finally, Freidson16 provided the specificity of the professions’ framework for the overall analysis.

Back to Top | Article Outline

Results

The policy makers we interviewed expressed widespread agreement and support for revalidation in principle. Despite continuing concerns, and despite the reluctance of many doctors, a member of the NHS Professionals (an employment agency within the NHS) concluded, “I think everybody is signed up to say it is a good thing.”

Participants widely believed that, although there were benefits to doctors and medical managers, the ultimate beneficiaries would be patients and the public. A member of the NHS Confederation (another NHS agency) claimed, “If it does not serve the interest of the patient, it has no point.” And a past member of the GMC 1 reminds us, “Licensing is for the public, not for the medical profession.”

Our first-level analysis identified the multiple signifiers*—or subjectively interpreted elements—of revalidation that were provided by interviewees. We explored how these discrete signifiers were articulated within broader discourses, and in what ways they served to reinforce or negate particular agendas. Although interviewees highlighted elements of discourses variously, each interviewee tended to reference elements that were compatible within specific equivalent systems of meaning. On closer inspection, these different systems supported two key discourses: regulation and professionalism (see Figure 1).

Figure 1

Figure 1

Back to Top | Article Outline

Regulation and professionalism: Discourses over time

The origins of revalidation identified by our interviewees were various. For example, they were situated in specific historical events (in particular the Bristol and Shipman Inquiries).17,18

I think there has been a whole range of high-profile cases where people have begun to wonder, “Why were these doctors allowed to get away with it?” and I think this is partly to do with that.

(Strategic health authority director)

The origins were also situated in the broader social and political changes reflected both in the rise of reform movements within the medical profession and also in the development of new standards and regulation by successive governments since the 1980s.

I think if you asked most doctors they would say that revalidation came from Shipman, but it is not true.… I think in my mind the drive for revalidation came from the profession well before Shipman … and was really to quality assure.

(Member of the AoMRC 1)

Therefore, revalidation’s genesis was constructed by our participants either as a reaction to medical malpractice brought to public attention by the media and the subsequent major government inquiries, or as part of the medical profession’s ongoing process of development. The apparent “confusion” of purposes (which we discuss critically later) is illustrated by a member of the Revalidation Support Team 1 (i.e., a body set up to support the implementation of revalidation in England), who asserted that

although it is about regulation of doctors, it is about safety and quality.… Everybody has a vested interest in safety and quality at all the levels.

This participant suggested further that

the purpose of revalidation is by helping doctors to maintain fitness-to-practice, and demonstrate that it will drive up quality of medical care, thereby increasing the effectiveness of the health service.

The quality agenda is arguably contrary to the safety agenda unless the two agendas are conflated. For example a member of the Scottish government claimed,

We have got a big patient safety program in Scotland that is translating into a quality strategy looking at not just patient safety but standards.

But as a member of the BMA pointed out:

We had become engaged in a search for excellence rather than a search for a basic safe standard. The problem, of course, is that when you say there is a basic safe standard that we should look for, it is boring, it is minimum, it is not very exciting, it does not describe the quality of being a doctor, and it feels very much more attractive to design a framework that identifies doctors who exemplify all of the attributes that you would like a doctor to have. The trap, I think, is that in identifying all of those you can become attracted to thinking that a doctor must exemplify all of those in order to get through revalidation.

The problem with this emphasis on quality, according to a member of the BMA 1, is that

it is not realistic, it is based upon an aspiration—“excellence”—that may be admirable but cannot be practically achieved.

As a member of the UK Revalidation Programme Board asserted, “Quality has no end.” The interviewee further explained:

It relates to where you draw the standard.… The temptation for the patient is to say, “I want this chap up here because he’s clearly the best,” but you can’t run a service with everyone getting the best surgeon, so the quest for quality goes on and on and on because we would then have all our patients saying, “Well, actually, we want to be treated by the top three surgeons of your hundred surgeons” and so it goes on and the thing is a never-ending trail.… So all you can have is … the minimum acceptable standard. That’s all you can draw.

The sustained use by our participants of the phrase “up-to-date and fit to practice,” to the point of it forming a simple rhetorical device, indicates that lengthy debates about revalidation have served to fuse the differences between the two key discourses of professionalism and regulation. Through discourse analysis we were able to locate the frustration about revalidation’s lack of clarity and progress, by identifying the conflicting underlying ideologies, such as quality and safety ideologies, that over time have become uncritically fused.

Perhaps the most significant practical consequence of this fusion of ideas is the focus on appraisal (a traditionally developmental instrument) as the focus of revalidation. This marriage of a formative and summative component in appraisal was questioned by some participants:

I think by and large the public see [revalidation] as excluding those who are not fit for purpose [sic], and the profession sees it as shifting the curve to the right.… Essentially the mechanisms you use for one are not necessarily the mechanisms you use for the other. One process has to be formative and the other is summative, and one of the dangers of our current position is that we are trying to use a single system to be both formative and summative and that is not easy to do.

(Member of the NHS Confederation)

The member of the UK Revalidation Programme Board concurred:

I think there has been confusion as to where appraisal between the so-called summative components and the other components lies. I think that’s not yet, even now, been clarified.

We considered this frequently reported “confusion” in terms of the inconsistencies between discourses of regulation and professionalism that might be provoking conflict. The focus here is on understanding that stakeholders (consciously or otherwise) used inconsistencies as argumentative or rhetorical strategies to inform particular definitions of revalidation as being both policy and practice ones. The interview data revealed that these strategies appeared to be related to the nature of the positions of the interviewees espousing them. In other words, doctors and their representative organizations tended to reference professional discourse, whereas governing institutions and policy makers pointed towards regulatory discourse.

This latter point is perhaps unsurprising, but the importance of these findings is in the details. To move toward a more positive approach, we looked for any consistencies that could be used as a gateway to consensus. The strategy here was to identify a third discursive formation that overlapped and thus conjoined the two conflicting discourses. In this final analysis, the patient discourse emerged as a unifying force, prevalent and positive in the discursive systems of all parties: “the discursive glue.”

Back to Top | Article Outline

The patient as the discursive glue

The key point of agreement between our participants was that revalidation should benefit patients.

My aspiration for revalidation is to get to the point where that continuing professional development and sense of professionalism drive quality improvement for patient care. It has to be about the best interests of the patient.… [It is that] desire to improve outcomes for patients that, I think, has to be at the heart of it.

(Member of NHS Quality Improvement Scotland)

However, the patient emerged as a “secondary” discourse—that is to say, patients were spoken ‘“about” and used as justification “for” policy rather than patients having a direct voice in revalidation “as” policy. For example:

We have regular discussions about how the public need to be kept informed. That is why we have a representative of the public and the patients to help us with that. There is going to be a specific communication strategy to inform the public about what this process means and how it will affect them.

(Member of the BMA)

Figure 1 shows graphically how patients, as a point of intersection, have acted as the glue that has held the professionalism and regulatory discourses together and yet has, at the same time, kept the debates described above from being resolved.

Back to Top | Article Outline

Discussion

Revalidation is one of the most significant developments in medical regulation internationally. For the first time, all doctors must undergo a national regulatory system to be able to practice. However, the drivers for these changes are contested even amongst those who have shaped the policy. Although our participants clearly recognized a lack of fit between traditional self-regulation and contemporary views of regulation and society, they demonstrated a continuing level of frustration with a debate that had entered its second decade. In other words, beyond the challenge of reaching agreement on the particulars of revalidation as a process, there is the additional and significant challenge of having to deal with the weight of the policy’s history.

Revalidation will have both direct and indirect consequences for the profession in the domains of patient and intraprofessional relationships. As proposed, it is not an intervention that directly targets the frontline interaction between doctors and patients. Rather, it is a mechanism that operates via the process of appraisal and involves a chain of interactions. However, although it does not act directly upon the doctor–patient relationship, it is anticipated to have an effect that will ultimately influence the dynamic between doctor and patient.

Revalidation aims to improve the quality of health care, reduce its risk, and enhance its relative safety. Our use of critical discourse analysis highlighted how revalidation has been constructed by our participants in very different ways. As a result, revalidation appears to have become a conflation of all these factors rather than a causal relationship between the identification of a specific problem (e.g., doctors over 55 years old struggle to remain up-to-date), a directed solution, and the development of a targeted educational program.

The perpetuation of this diversity of opinion has had significant consequences for the development of policy, since it has caused a divergence of aims and purposes: regulation (catching “bad” doctors) requiring summative assessment, and professionalism (an innate professional drive and duty to maintain high standards) requiring formative assessment (as illustrated in Figure 1).

Functionally, regulation must serve a summative purpose whereby a conclusive and confident point-in-time judgment about a doctor’s fitness to practice is made against a fixed standard. The professionalism discourse, on the contrary, is defined by a formative process whereby conclusive judgment is deferred in favor of ongoing review and support to help foster an ethic of continual improvement that seeks to keep a doctor up-to-date. These two discourses have an impact on the doctor and create tension in terms of what, exactly, the shape of the doctor’s practice should look like.

Untangling the relationships between the competing discourses as they operate within revalidation is vital for three reasons. First, it helps us understand why the policy process has dragged on for so long. Second, it enables us to address the burden of that history in going forward. And third, it allows us to develop a strategy to overcome the conflicts and cross-purposes that continue to threaten this policy’s implementation. One key challenge in drawing up the specifics of revalidation is the divergence of the discourses around the issue of defining standards. A catch-22 situation is created, since without clearly set standards, regulation cannot function. Yet, once standards are fixed, there is a risk that they will undermine the professionalism project to drive up standards.

Such matters are not unique to revalidation; for example, van der Vleuten et al19 offer insights into similar concerns in undergraduate medical assessment that may provide an alternative way of approaching these challenges. They propose that assessment programs should be cyclical, ongoing, and focused on learning. Only after significant periods of repeated assessment do you then bring together the feedback. This, they argue, balances the opportunities to learn from assessment while at the same time periodically allowing for a judgment to be made about performance.

Finally, members of the different stakeholder groups, at different times, all used the patient to promote and/or justify their own agendas, but at no time did they built any real case for revalidation as a truly patient-centered policy. The patient had no active voice within the discourse but was a proxy for other agendas. Patients and the public were terms used interchangeably, but they are not the same. Patients are in small local relationships with physicians or small medical teams, whereas the public encapsulates the collective of patients as a stakeholder group. Patients’ needs are answered by the tenets of professionalism, but the public’s are answered by regulation. Although our participants referred to patients and the public uncritically, we would contend that the role of patients as a unifying force, in both rhetoric and practice, requires further clarification.

All qualitative interview studies can be limited by their sampling. However, we were fortunate enough to gain unprecedented access to many of the main stakeholder groups and individuals involved in the development of revalidation policy. In this way we believe that we have a representative example of opinions. However, as we have discussed, revalidation policy has always been a product in and over time, and it must be acknowledged that our interviews may represent only one moment in this time continuum. Last, discourse analysis, especially Foucauldian,10 seeks to understand power within language. We took a purposeful sample of the most empowered, and so it is important that our work be taken in context and that other research explore the experiences and views of the other actors—namely, doctors and patients.

Our exploration of revalidation, as portrayed by its champions and detractors, was always to better understand its potential benefits and what might facilitate or impede them. By identifying a lack of clarity about what the purpose of revalidation is, we have placed a spotlight on why the pendulum had swung backwards and forwards for 15 years before revalidation was formally established, and why achieving that was so difficult. But our findings are important for the future also, as without a defined purpose for revalidation, it is unlikely that such a complex intervention will deliver the desired impacts.

Back to Top | Article Outline

Summary

Despite agreement around issues of principle, our research identified revalidation as the product of two determining discourses (regulation and professionalism) held together by a discursive formation around patients, which we call the “discursive glue.”

The regulation and professionalism discourses, although not mutually exclusive, are the result of different drivers, with different aims that, importantly, require different processes. The delay and frustration in implementing revalidation has its origins in the inherent contradictions between these discourses. A member of the NCAS summarizes the challenges of trying to advocate a policy that contains contradictions by saying, “I absolutely believe that no one has yet properly articulated what we are trying to achieve.” Within the challenges of complex and divergent policy construction, policy makers have used patients as “glue” to bind these discourses together. However, we conclude that despite patients’ pivotal role in the discourse, policy makers have failed to identify patients as a stakeholder body of equivalence to the other bodies or with an active voice in the debate.

So a clearer composite of the mix between the regulatory and professionalism components of revalidation must be developed—and we would suggest that this mix is determined by the needs of the patient, as voiced by patients and members of the medical profession in partnership.

Back to Top | Article Outline

Conclusions

Revalidation, as implemented, is a political compromise. But to avoid a watering down of the original goals of revalidation, and to avoid tension between stakeholders, there is a need to identify common goals as gateways to making revalidation meaningful to all. It is extremely important that revalidation appear meaningful to the day-to-day work of doctors, and this is the focus of ongoing programmatic research.20

Perhaps most important, to maximize trust and partnership, genuine patient participation should drive a truly patient-centered approach to revalidation, not only in feedback to individual doctors but also in actively shaping the policy and the process. While patients are increasingly viewed as partners in their own care, we conclude that “the patient discourse” needs to be more formally recognized within revalidation, making the patient not the glue but the substance.

Finally, although there is clearly much energy behind medical regulation, there are often too many assumptions about the possible impacts it may have on the patient experience, safety, and quality improvement. Implementing such a complex intervention requires not just a clarity of purpose but also the anticipation (and measurement) of both intended and unintended consequences. As Iglehart and Baron5 conclude,

ABMS [American Board of Medical Specialties] and its boards must actively (and transparently) … strive to navigate a complex system that melds professionalism, government regulation, and market forces.

We would argue that this can be achieved only through established and ongoing evaluation of medical regulation as a policy and process.

Acknowledgments: The authors would like to thank all the interviewees for taking part and speaking so openly and passionately about their involvement in and views on revalidation.

* Objects (such as words, places, ideas, and things) have no intrinsic value. We apply value to them. Therefore, objects are constructed through language and become signified by multiple (sometimes conflicting) signifiers. For example “patient and public involvement,” a regularly used term, especially in medical research, can be taken to mean different things—such as lay representation, public groups, patient feedback, public participation, etc.—which may have very different implications in practice.
Cited Here...

Back to Top | Article Outline

References

1. Brennan TA. Recertification for internists—one “grandfather’s” experience. N Engl J Med. 2005;353:1989–1992
2. Medical Board of Australia. . Registration. 2013 http://www.medicalboard.gov.au/News/Newsletters/November-2013.aspx. Accessed June 26, 2014
3. Ward B. Revalidation in Canada 2010.December 2–3, 2010London, UK Paper presented at: International Revalidation Symposium: Contributing to the Evidence Base
4. General Medical Council. . Continuing professional development: An international perspective. 2011 http://www.gmc-uk.org/CPD___The_International_Perspective_Jul_11.pdf_44810902.pdf. Accessed June 26, 2014
5. Iglehart JK, Baron RB. Ensuring physicians’ competence—is maintenance of certification the answer? N Engl J Med. 2012;367:2543–2549
6. General Medical Council. . Supporting Information for Appraisal and Revalidation. 2012 http://www.gmc-uk.org/static/documents/content/Supporting_information_for_appraisal_and_revalidation.pdf. Accessed June 9, 2014
7. Irvine D. A short history of the General Medical Council. Med Educ. 2006;40:202–211
8. Merrison. Merrison Report: Report of the Committee of Inquiry Into the Regulation of the Medical Profession. 1975 London, UK Her Majesty’s Stationary Office
9. Collaboration for the Advancement of Medical Education Research and Assessment (CAMERA). Revalidation research. 2013. http://www1.plymouth.ac.uk/peninsula/research/camera/revalidation/Pages/default.aspx. Accessed June 10, 2014
10. Petersen A The Body in Question: A Socio-cultural Approach. 2007 London, UK Routledge
11. Foucault M The Archaeology of Knowledge. 1972 London, UK Routledge
12. Foucault MBurchell G, Gordon C, Miller P. On governmentality. In: The Foucault Effect: Studies in Governmentality. 1991 Chicago, Ill University of Chicago Press
13. Rose N Powers of Freedom: Reframing Political Thought. 1999 Cambridge, UK Cambridge University Press
14. Torfing J New Theories of Discourse. 1999 Oxford, UK Blackwell
15. Hodges BD, Kuper A, Reeves S. Discourse analysis. BMJ. 2008;337:a879
16. Freidson E Professional Dominance: The Social Structure of Medical Care. 1970 New York, NY Atherton Press
17. The Bristol Royal Infirmary Inquiry: Summary. 2001 London, UK Department of Health
18. The Committee of Inquiry to Investigate How the NHS Handled Allegations About the Conduct of Clifford Ayling. 2004 London, UK Department of Health
19. van der Vleuten CPM, Schuwirth LWT, Driessen EW, et al. A model for programmatic assessment fit for purpose. Med Teach. 2012;34:205–214
20. Archer J, Regan de Bere S, Brennan N, Lynn N. Evaluating the strategic impact of medical revalidation: Building an evaluation framework. December 2013 http://www.gmc-uk.org/about/research/23660.asp. Accessed August 2, 2014
© 2015 by the Association of American Medical Colleges