Wong, Brian M. MD; Kuper, Ayelet MD, DPhil; Hollenberg, Elisa MSW; Etchells, Edward E. MD, MSc; Levinson, Wendy MD; Shojania, Kaveh G. MD
Despite a strong educational mandate to teach quality improvement (QI) and patient safety (PS) concepts to trainees endorsed by the Association of American Medical Colleges1 and the Accreditation Council for Graduate Medical Education,2 the majority of medical schools and residency programs in the United States and Canada continue to experience difficulty in their efforts to implement and sustain QI/PS curricula. Even with a burgeoning literature base that supports efforts to design and deliver formal training in QI/PS,3–7 recent estimates suggest that less than one-quarter of medical schools in these two countries formally teach QI/PS in their medical curricula.8,9
The authors have informed the journal that the research team for this report agrees that both Dr. Wong and Dr. Kuper completed the intellectual and other work typical of the first author.
The reasons for this gap are unclear and have yet to be fully explored. We previously tried to formally address this by supplementing a traditional systematic review of QI/PS curricula with a thematic analysis of the articles through which we identified factors that their authors regarded as promoting or limiting curricular implementation.4 This review helped generate a set of hypothetical learner, teacher, curricular, and learning environment factors, including, respectively, competing educational demands, degree of faculty expertise, capacity to combine didactic and experiential training, and the strength of the institutional QI culture, that may influence the successful implementation of a QI/PS curriculum. These factors align closely with the traditional considerations that medical educators have used to guide the design of their educational programs so that the learning experiences of their trainees are robust and of high quality.10
However, a critical question remains: Beyond simply considering these factors, how can educators ensure that QI/PS are among the core topics taught to future physicians, both at their particular institutions and more broadly? Our systematic review could not adequately address this concern because most of the relevant articles did not explicitly discuss it. Yet, failing to consider this larger question leads educators to develop and implement well-designed, educationally sound programs that may effectively achieve their desired learning outcomes but lack sustainability.
The use of social theories may help to address this larger sustainability question. Social theories have the advantage of providing unique lenses through which researchers can expose and examine different aspects of medical education, including those related to larger educational structural and policy questions.11,12 Further, applying a social theory is consistent with the increasing call for theory-driven research in medical education.11,13–15
Medical educators are beginning to use the theoretical framework of French sociologist Pierre Bourdieu16–18 to illuminate the forces that influence and change what individuals consider to be legitimate topics to teach and learn.13,19 Recently, investigators have applied his concepts of field, habitus, and capital to various aspects of medical education, ranging from the differences in the kinds of knowledge that different medical schools in the United Kingdom emphasize19 to the forms of legitimate inquiry in the evolving field of medical education research.13
We therefore sought to use insights from Bourdieu’s work to examine current efforts to teach QI/PS to residents and to better characterize the challenges that educators face when trying to introduce QI/PS into graduate-level medical curricula. We intended to identify potential strategies to overcome these barriers and to provide empirically informed suggestions for future steps that may lead to a greater uptake of QI/PS training in graduate medical education.
Bourdieu’s theoretical framework includes two key interrelated concepts: “field” and “habitus” (Table 1). The concept of “field” describes an arena in which players produce, circulate, and acquire resources that relate to a specific area (e.g., medicine, ballet). These resources, otherwise known as “capital,” have different forms, including money (“economic capital”), culture (“cultural capital”), connections (“social capital”)—or any combination of things that give a member of a field prestige or a good reputation (“symbolic capital”).20 Each field is characterized by a “game,” in which, as described by Albert and colleagues,13
the competition for predominance of one definition over competing definitions as the recognized model of excellence in the field results in a struggle between players as each tries to promote a definition that places value on their own products and their own ways of doing things.
The imbalanced structure of power relationships that this struggle inevitably creates represents an additional defining feature of the field.17
Capital is field specific: What is legitimate in one social space or arena may not be in another, and because these definitions and structures are fundamentally arbitrary, they can change over time. Those players who have a lot of symbolic capital (i.e., prestige) have the ability to change their field’s definition of legitimate forms of capital, which they often do to favor their own way of doing things and to further strengthen their position within the field.
The concept of “habitus” encapsulates why individuals from a specific group (e.g., a nationality, a socioeconomic class, a health profession) in a given field tend to have predictable patterns of behavior attributable neither to explicit rules nor to conscious choices.18,21 Past experiences shape these patterns of behavior and tend to feed forward to influence an individual’s notion of him- or herself, such that the future often reproduces the past. The imbalanced structure of power relations that result from the unequal distribution of legitimate capital can strongly influence habitus. Individuals belonging to groups who possess comparatively less legitimate capital have more difficulty acquiring legitimate capital, thereby perpetuating the existing imbalances in power relations that dominate a given field.18,21–23
Field and habitus are, thus, interconnected concepts, each with the ability to influence and modify the other. As described by Brosnan,19
it is the struggle within the field that determines which players’ habitus can access the most capital, while the relations between players are what shape the field. Thus, in order to study a field, one must examine the relations between individual players and the elements that are valued within that field.
Within this constructivist theoretical framework,24 we conducted an exploratory case study25,26 of the incorporation of QI/PS into the medical curriculum. Constructivism, for our purposes, is “a belief […] that the reality we perceive is constructed by our social, historical, and individual contexts, and so there can be no absolute shared truth.”27 In the social sciences, case study methodology is a way of conducting an analysis of an interesting or informative phenomenon in order todevelop a greater understanding of that phenomenon by grounding empirical observations in social theory. The phenomenon of implementing QI/PS curricula constitutes an informative case because (1) different training programs are at different stages in their implementation, and they have used different strategies to achieve this implementation,4 thereby providing a varied study sample, (2) many educators and program directors view implementing QI/PS education as important, yet many lack QI/PS training themselves and have insufficient expertise to teach QI/PS to others, and (3) teaching QI/PS is such a recent phenomenon within medical education that the earliest innovators are still available forinterviews.
Because different drivers, accreditation standards, and learning contexts influence curriculum implementation in undergraduate (medical school) and graduate (residency) medical education, we chose to limit our case study to the residency clinical training environment. We, therefore, recruited participants who were actively involved in implementing QI/PS curricula for residents. We identified the initial pool of potential participants from the list of principal authors whose published QI/PS curricula we had examined in our recent systematic review, published initially in 20104 and then updated in 2012.28 We purposively sampled among them, balancing the training program characteristics (e.g., country, discipline) and educational features (e.g., design, teaching methods) of their curricula. Because sampling from the systematic review would likely identify only participants who had been successful in their curricular implementation, we also conducted confirming/disconfirming snowball sampling29 by asking our study participants to identify individuals whom they knew had attempted to implement a QI/PS curriculum but had not published about it, particularly those who had failed in their implementation. We continued to sample to theoretical saturation.29
Ethical considerations, data collection, and analysis
We gathered data by conducting semistructured, one-on-one interviews between October 2010 and May 2011. The Sunnybrook Health Sciences Centre research ethics board (Toronto, Ontario, Canada) approved this study. After receiving information about the nature and methods of our study, all participants provided informed consent. We did not offer incentives for participation. We anonymized all study data and stored them securely. One of two study investigators (B.W., E.H.) interviewed each study participant for 45 to 60 minutes, either by telephone or, whenever possible, in person. We audiotaped and transcribed interviews verbatim. We based the initial interview guides (Supplemental Digital List 1, http://links.lww.com/ACADMED/A135) for these semistructured interviews on our synthesis of the literature related to facilitators of and barriers to teaching QI/PS to residents and to implementing curricular changes in faculties or schools of medicine,4,6,30–33 on our own contextual knowledge of the academic and clinical practice of QI/PS in academic health centers, and on our understanding of Bourdieu’s theories.
Using Bourdieu’s theories as a framework, we concurrently gathered and analyzed data, iteratively adjusting the interview script and making decisions regarding saturation. Our analytic approach combined categorization and meaning condensation to generate a theoretically grounded interpretation of the data; that is, we coded data from the interviews into categories, and we shortened some of the participants’ longer answers into “shorter formulations” that preserved meaning.34 Our analysis was reflexively mindful of our own subject positions in the research context; we remained cognizant of the forces (e.g., our own genders, ethnic backgrounds, and social positions) that affected our analysis of the data.27 In particular, several of us (B.W., E.E., W.L., K.S.) hold local or national leadership positions in medical education and/or QI/PS, through which we actively engage in work to implement and legitimize QI/PS curricula at multiple educational levels. Additionally, two of us (A.K., W.L.) advocate for more theoretically oriented research related to medical education.
Three of us (B.W., E.H., and A.K.) primarily carried out the concurrent analysis, while others on the research team (K.S., W.L., and E.E.), who also read the transcripts, helped further refine the results by discussing coding schemes and suggesting ongoing changes to the interview guides.
We interviewed 16 individuals, 8 of whom were female, who developed QI/PS curricula targeting residents in a variety of training programs, including internal medicine (n = 9), family medicine (n = 2), pediatrics (n = 2), general surgery (n = 1), preventive medicine (n = 1), and psychiatry (n = 1), based in the United States (n = 11) and Canada (n = 5). We identified 12 participants by screening articles included in the original and updated systematic reviews.4,28 These 12 participants were either lead or coauthors on 13 (33%) of the 40 studies of QI/PS curricula that targeted residents. We identified 4 additional participants through snowball sampling.29 We specifically included 2 of the participants because they faced significant challenges when implementing their QI/PS curricula.
The fields associated with QI/PS curricular implementation
Our data revealed that academic physicians are the players that belong to, and compete for the legitimate forms of capital in, two different but interrelated fields associated with QI/PS curricular implementation: the academic field and the health care delivery field. We have organized our interpretation of the results by first characterizing each field separately, describing the different forms of legitimate capital that create the power structures within them, and explaining the way that QI/PS initiatives are positioned within each of these fields. We then turn our attention to characterizing relevant aspects of the current academic physician habitus that is shaped by, and continues to shape, these two fields. Finally, we illustrate how the educators whom we interviewed sought to exploit the prevailing forms of legitimate capital and/or to redefine what each field considers to be legitimate capital in an effort to encourage a change in the academic physician habitus and promote QI/PS as a discipline, and we describe how these efforts contributed to the sustained delivery of QI/PS training in their residency programs.
The academic field.
This field encompasses the “academy,” which for programs affiliated with a university or medical school is the “university,” whereas for programs primarily affiliated with a hospital, it is “wherever scholarly activities are carried out.” Within the academic field, physicians compete for those elements that improve their academic reputation, which would advance their position within the field and also allow them to then redefine legitimate forms of capital.
Within the academic field, the existing power structures promote the pursuit of traditional forms of bioscientific research—rather than QI/PS—as the dominant form of legitimate capital. Our participants indicated that individuals engaged in QI/PS tend not to seek grant support or publish their work, making their work less legitimate in the academic field. In contrast, individuals doing more traditional bioscientific academic work who have published a large number of peer-reviewed publications and garnered external grants tend to get more protected academic time and salary support, tend to receive awards and accolades, tend to get promoted on the basis of their research, and tend to curry favor within the academic field. To illustrate, two respondents commented:
[Our university] is a very research-oriented institution so the hiring priorities often focus on people with large research agendas and QI has not yet amassed that kind of profile. (Respondent 15)
Even though it says, for example, in many criteria … that publishing isn’t necessary, if you ask the vast majority of Promotion and Tenure Committees, they’re not going to promote somebody unless they have papers. (Respondent 14)
Our participants saw QI/PS as a new subject area within the academic field—one that has the potential of competing with traditional bioscientific research as the dominant form of symbolic capital. As a result, academic physicians who derive power within the academic field through their research productivity view the emergence of QI/PS within the field as a potential threat. One respondent observed,
There is a concern from the traditional researchers that we would teach and mandate our residents to learn QI and to practice QI—say doing a QI project—that this will steal time away from the research curriculum. (Respondent 3)
Our participants indicated that academic leaders, such as department chairs and educational directors, strongly favor the current traditional, bioscientific forms of legitimate capital because they serve to enhance their department’s reputation. This favor tends to further reinforce the existing forms of legitimate capital (e.g., research grants, publications) within the academic field. Academic departments tend not to support or sustain activities that do not align with the academic leadership’s existing views of academic legitimacy. This lack of support has important implications for the legitimacy of QI/PS in the academic field, as illustrated by one participant’s comment:
We got wonderfully passionate people interested in QI but their deans don’t buy in so they don’t give them time or they give them one hour or two hours to teach QI and that’s not going to be successful … so, it’s important to engage the leadership from the very beginning. (Respondent 9)
Exacerbating the problem, many academic departments have very few individuals who choose to pursue QI/PS academically.
It’s the same five people with the [QI] thing over and over and over again. (Respondent 14)
At the moment […] I’m a one-person show. (Respondent 4)
QI/PS, therefore, may have a lower profile than other academic pursuits. This lack of visibility further reinforces QI/PS’s lack of legitimacy and permits the dominant or more “legitimate” forms of symbolic capital (e.g., bioscientific research) to maintain their position of power within the academic field.
The health care delivery field.
This field includes the “clinical practice environments,” typically defined as a “teaching hospital” or “teaching clinic,” where academic physicians deliver clinical care and train residents and medical students. Similar to the academic field, our respondents provided suggestions for the elements that conferred greater symbolic capital in the health care delivery field: Physician practices aimed at improving the health status of individual patients dominate. Individuals regard academic physicians as “good doctors” in the health care delivery field primarily on the basis of their dedication to individual patient care rather than their efforts to improve the care delivery system at large:
All you learn about is how to deal with the patient in front of you … nobody ever talks about or very rarely talks about [a physician’s] obligation to the system … if you could just get it through people’s heads that you don’t just treat the patient in front of you, you also treat the system … it could be a valued aspect of what it is to be a physician and what it is to be an academic physician. (Respondent 14)
Our participants indicated that the approach taken to teach QI/PS concepts to residents further reinforces the greater legitimacy of caring for individual patients. Many training programs continue to favor the bioscientific model and clinical training, emphasizing this curricular content over quality and system improvement concepts. Further, residency training programs often teach QI/PS as an add-on subject, which sends a clear message that QI/PS is less important relative to other clinical topics, which are better integrated into the ongoing curriculum. This lack of integration further delegitimizes QI/PS in the health care delivery field, as illustrated by one respondent:
When you single it out as a project and get it over with or do it as a specific set of things that you’re going to accomplish in a short period of time, you don’t send the same message as if you embed it in the work as something you’re doing all the time. (Respondent 13)
According to our participants, economic capital is the other active form of capital in the health care delivery field—at the level of both the individual academic physician and the institution as a whole. Our respondents perceived that engaging in QI/PS might negatively affect the financial bottom line:
The person who’s interested in finance is not necessarily interested in giving up clinical practice time for [QI] because they’re worried about the loss of revenue. (Respondent 13)
I was taking time away from clinic visits to deal with it as quality, and because we’re always in the red, administration didn’t like giving up productivity. (Respondent 8)
Interestingly, teaching hospitals reward and recognize contributions made to advance the strategic institutional priorities, including those related to overall care delivery. Academic physiciansmight theoretically seek achievements in institutional priorities as another form of “legitimate” symbolic capital within the health care delivery field; however, few respondents identified such work as an active form of symbolic capital.
The academic physician habitus as it relates to QI/PS
Many of the respondents spoke candidly about the academic physician habitus, using phrases such as “professional identity” or “what it is to be a physician.” Although the scope of this study does not encompass an in-depth analysis of all aspects of the academic physician habitus, respondents clearly felt that it did not include an appreciation for QI/PS. Our respondents identified three different ways in which academic physicians responded to QI/PS (as illustrated respectively by the comments below): Some seem openly hostile and resistant to QI/PS; others seem to feel neutral toward QI/PS but value other priorities more highly; and finally, some academic physicians seem amenable to QI/PS but are as yet untrained in its practice and teaching.
We looked at our initial quality data and it was pretty abysmal and … [we’ve] got a faculty member that thinks it’s just everybody else’s fault. (Respondent 8)
I don’t say that they had the same priority [for QI] in their head as I did, but they were amenable, but whether they gave [QI] the energy that I did? No, I would say that’s not true. (Respondent 13)
They’re not really trying to be a jerk, it’s just that they don’t have the [QI] skills and so they get defensive or there’s an anxiety because they’re going to be exposed that they … don’t have the [QI] skills. (Respondent 8)
In Bourdieuvian terms, the specificity of the fields in which academic physicians work and the active forms of capital within these fields shape the academic physician habitus, including the overall lack of appreciation for QI/PS. In addition, although academic physicians work in two related fields (i.e., the academic and the health care delivery fields), these fields do not appear to influence their habitus as it relates to QI/PS to the same extent. Our respondents felt that the academic physician habitus tends to favor acquiring symbolic capital in the academic field over symbolic capital in the health care delivery field:
I think by and large the academic establishment … see research funding, [and] as you well know, academic promotion … as their primary driver of what they do. (Respondent 12)
As such, even if opportunities for new forms of symbolic capital (e.g., recognition for working on clinical institutional priorities) were to gain legitimacy within the health care delivery field, academic physicians are less likely to pursue these because they are too busy carrying out activities that allow them to advance themselves within the academic field.
Changing the habitus by changing the field: Strategies for legitimizing QI/PS
What I learned … was just that there’s two jobs that each of us have as physicians, the job to deliver good care, and to improve the care we deliver. (Respondent 11)
Our respondents talked a great deal about the role of faculty development in changing physician knowledge of and attitudes toward QI/PS; however, none of them had yet found success with that as their sole approach. Education has long been identified as a way of modifying the habitus,21 so educating faculty would reasonably function in this way for individual clinicians to some degree. However, in keeping with Bourdieu’s theories, most effectively bringing about changes to the academic physician habitus, and in turn, encouraging the uptake of QI/PS, would require increasing the legitimacy of QI/PS in both of the fields in which they function. Such redefining of the academic physician habitus relies heavily on the leaders of the academic and health care delivery fields who possess the necessary prestige and power to motivate these changes:
And one of the reasons [the QI curriculum] is not taken away from us or that the time isn’t reduced is because it’s clear to our leadership and to our program, this training in [QI] is just as important as the training they’re getting in seeing patients. (Respondent 6)
Our respondents provided a number of concrete suggestions, framed within Bourdieu’s concept of symbolic capital, for how they legitimized QI/PS in their own contexts (Table 2). There are two potential sets of legitimating strategies. The first set focuses on strategies that demonstrate how QI/PS can increase what players consider to be more legitimate symbolic capital, thus improving the position of an academic physician who works in QI/PS within the academic and health care delivery fields. The second set centers around changing the definition of legitimate symbolic capital to include QI/PS within the academic and health care delivery fields.
Participants provided suggestions for using QI/PS to increase legitimate symbolic capital. One recommendation within the academic field was publishing QI/PS initiatives in peer-reviewed journals. Such publications highlight the fact that QI/PS can contribute to improved research productivity in the traditional sense. Respondents recommended other strategies, such as creating research awards for QI/PS or providing funding to academic physicians to protect their time spent pursuing scholarly QI/PS work; such awards and funding would improve the reputations of academic physicians who work in QI/PS by allowing them to claim that they received an award or academic salary support for their QI/PS work. Some respondents also advocated, whenever possible, recognizing and promoting academic successes in QI/PS. This recognition could include making others aware of the scholarly output resulting from QI/PS work or public acknowledgment of important QI/PS accomplishments.
Some examples of how QI/PS could exploit the traditional forms of symbolic capital within the health care delivery field that participants mentioned include making explicit the link between effective QI/PS initiatives and local improvements in patient outcomes. Doing so would underscore the value of QI/PS. Other respondents suggested incentives to encourage participation in QI/PS, such as remunerating physicians for time spent leading QI/PS initiatives and recognizing and rewarding physician participation in QI/PS initiatives. Strategies could also focus on the health care delivery field at large, especially increasing the legitimacy of QI/PS within the clinical organization. One participant suggested highlighting the economic value of QI/PS to the clinical organization.
Participants also gave specific recommendations for changing the definition of legitimate symbolic capital within the academic and health care delivery fields to include QI/PS. Within the academic field, our respondents felt that elevating QI/PS to make it equivalent to bioscientific research was critical. In some departments, chairs promoted academic physicians on the basis of their work in QI/PS as a way to raise the profile of QI/PS.
Within the health care delivery field, many of our respondents focused on encouraging academic physicians to value their responsibility to improving the system that cares for patients. Participants felt that creating awareness around successful examples of physician engagement in QI/PS helped in this regard. Others recommended investing in a core of academic physicians that can lead and promote QI/PS initiatives by encouraging others to participate in them. This core of physicians, along with senior clinicians, must role-model QI/PS in everyday work and discuss it in everyday conversations, so that others in the field see it as “business as usual.”
For residency programs to sustain their QI/PS curricular efforts, the academic physician habitus needs to change, such that academic physicians see QI/PS as core to their identity. This identification with QI/PS should extend to both the physician’s scholarly work in the academic field as well as to his or her clinical work in the health care delivery field. This change requires an antecedent shift in these two fields, including necessarily an evolution in what are considered to be legitimate forms of capital within each of them.
This understanding of the changes that must occur, derived from Bourdieu’s theoretical framework, enables us to highlight particular issues and illuminate important challenges and opportunities to consider in implementing and working to sustain QI/PS curricula in residency training. Although not versed in Bourdieu’s theories, our respondents clearly articulated some strategies that have worked to legitimize QI/PS in their own educational contexts. Examining and expanding on these strategies using a sociological perspective like Bourdieu’s can uncover other potential approaches to accomplishing the goal of legitimizing QI/PS that might not be immediately evident to educators in their daily work. Table 3 lists examples of strategies that our participants did not mention but which derive both from our theoretical understanding and from what is known about how these fields function. This list is not intended to be exhaustive; rather, we intend it to serve as a demonstration of how academic physicians can apply Bourdieu’s theories when strategizing to increase the legitimacy of QI/PS. Explicitly applying Bourdieu, in conjunction with considering the suggestions gleaned from our participants (as presented in Table 2), will guide thinking regarding the range of possible strategies that training programs could attempt within the limits of their particular contexts and resources.
Recent examples in the literature highlightinitiatives or efforts that, viewedthrough our theoretical lens, contribute to legitimizing QI/PS. Perhaps the most notable are the emergence of journals dedicated to publishing QI/PS research and the development of the SQUIRE guidelines for publishing a QI/PS manuscript (Standards for Quality Improvement Reporting Excellence).35 Together, these guidelines and emerging journals promote QI/PS as a legitimate form of research. From an educational standpoint, some training programs now have QI leadership positions such as a resident quality and patient safety officer36 or chief resident in quality and patient safety,37 and some programs even offer financial incentives to residents to engage in QI initiatives.38 These initiatives provide concrete models of how other residency training programs can promote QI/PS through actions that, from our perspective, serve to legitimize it in the academic and/or health care delivery fields.
Although our findings derive from a case study of QI/PS and focus on, in particular, making QI/PS legitimate, members of other groups trying to foster the legitimacy of other areas within the academic medicine and/or health care delivery fields will likely find our results useful. Although more empirical work is necessary to map out the details of other fields in which members of other groups may function, medical education researchers and qualitative health researchers could benefit from consciously understanding the configurations of those fields in order to work to change the forms of capital valued therein to their advantage. This form of transferability, based in regularities of phenomena and processes, is a useful outcome of research using social theories like Bourdieu’s.
Our study has several limitations. From a methodological perspective, we have collected only interview data, which we did not triangulate with observational data or policy documents. Consequently, our findings represent the perspectives of our participants on what their specific contexts hold to be legitimate. We included educators from only residency training programs; likely, the undergraduate and continuing medical education contexts are quite different, although we would expect our theoretical framework to still hold. Finally, social theories are intended for the study of social groups rather than of individuals; therefore, our findings, although transferable through the use of theoretical generalizability, will not apply to every individual in all circumstances. The strategies we have outlined still require consideration within the individual contexts of specific programs; the degree to which they resonate with programs’ particular circumstances will dictate their usefulness.
Our case study of QI/PS curriculum implementation suggests that incorporating and sustaining an educational QI/PS program in the medical curriculum of residency training programs requires a concerted effort aimed at increasing the legitimacy of QI/PS in both the academic medicine and health care delivery fields. Situating our findings in Bourdieu’s theoretical framework allows us to articulate an approach to developing concrete strategies that can legitimize QI/PS in these two related fields. Such strategies can promote sustainable implementation of QI/PS content into the core curriculum of residency training programs in order to transform future generations of physicians.
Acknowledgments: The authors thank Dr. Elise Paradis, MA, PhD, for her helpful comments on earlier drafts of the manuscript.
Funding/Support: The authors received funding from the Education Development Fund for Innovation in Education (Faculty of Medicine, University of Toronto). The funding program had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; or preparation, review, or approval of the manuscript.
Other disclosures: None.
Ethical approval: The Sunnybrook Health Sciences Centre research ethics board (Toronto, Ontario, Canada) conducted an expedited review and approved this study.
Previous presentations: The authors presented preliminary data as oral research abstracts at the Association of American Medical Colleges Integrating Quality Meeting (June 10, 2011 in Chicago, Illinois) and the International Conference on Residency Education (September 23, 2011 in Quebec City, Quebec, Canada).