Despite years of effort to teach and enforce positive professional norms and standards, many reports of challenges to medical professionalism continue to appear, both in the medical and education literature and, often in reaction, in the lay press.1,2,3,4,5 Examples of professional lapses dot the health care landscape: regulations are thwarted, records are falsified, patients are ignored, colleagues are berated.2,4,6 The medical profession has articulated its sense of what professionalism is in a number of important position statements.7,8 These statements tend to be built upon abstracted principles and values, such as the taxonomy presented in the American Board of Internal Medicine's (ABIM's) Project Professionalism: altruism, accountability, excellence, duty, honor, integrity, and respect for others.7
These principles provide a common and coherent language for describing a basic framework of professional expectations. The community's considerable effort to precisely define these principles suggests an assumption that by providing definitions, we will be able not only to establish practice guidelines but also to impart professional principles to the next generation of practitioners. However, as with all practice guidelines, the clinician must interpret them within the context of his or her own practice. There may be instances in which deception is required (e.g., to protect a patient's confidentiality); there may be others in which two equally viable professional actions conflict (e.g., support the team's decision to wait, or immediately disclose information to a patient).9,10,11 Similarly, the learner's developing understanding of such abstract concepts depends upon the context in which she or he interprets them. Therefore, to effectively shape students' developing professionalism, we need to examine their situated perceptions of professional values and behaviors.
Students both witness others' professional decision making and encounter their own challenges. Through these experiences, they develop their “professional judgment.” Not unlike clinical judgment, professional judgment cannot be adequately learned by lecture or by rote, or confidently assessed by written examination alone. Students, as novices, acquire and apply professionalism through specific instances, and their professionalism must be taught and evaluated using the same instances. Attempts to do otherwise may lead to curricula and evaluations that are too abstract and too distant from students' real-life experience.
Efforts to understand professionalism from the position of students, while critically important, have been unable to represent the situatedness of the students' experiences. To date, the research literature has relied largely on students' recall of generalized professional and ethical dilemmas, and their depictions of the frequency and severity of these phenomena.1,2,4 While these reports provide a useful overview, they cannot offer a full understanding of what specific events students have witnessed or dilemmas they have encountered, nor have they encouraged students' reflections on these experiences. Towards these ends, this study was designed to identify and elaborate the particular contexts and dilemmas that challenge students' developing professionalism. Through such a process, we can begin to characterize the professional lapse from the point of view of students, to probe their knowledge and understanding of professionalism, and to strategize how best to bring current theories of professionalism to bear on students' experiences.
We used a convenience sample of final-year medical students rotating through internal medicine at the University of Toronto Faculty of Medicine, McMaster University Faculty of Health Sciences, and the University of Michigan Medical School. These sites (two Canadian schools with substantially different curricula, and one U.S. school) were chosen to maximize generalizability and facilitate the collection of a broad spectrum of students' contexts and experiences. Using theoretical sampling, sample sufficiency and saturation were determined through a confirming and disconfirming process alongside data collection.12 The decision to cease data collection was made when no changes to the conceptual structure were forthcoming from the data.
Six focus groups involving 29 students were conducted. Informed consent was obtained, and group participation was anonymous. Each focus group lasted 1.5 hours, followed a semi-structured format, and was facilitated by the same trained interviewer. Students were instructed to report specific instances of lapses in professional behavior they had witnessed, had knowledge of, or identified in themselves. We encouraged them to discuss professional dilemmas and discouraged them from discussing classic issues in medical ethics such as DNR (do not resuscitate) orders. Discussions were audio-recorded, transcribed, and rendered anonymous.
Transcription of the discussions yielded 120 pages of text, which were analyzed for specific instances of professional dilemmas. These instances were identified through the application of two inclusion/exclusion criteria: the description must be of a specific event, not a generalization, and the description must be of an actual occurrence, not a hypothetical one. Three researchers applied the criteria to isolate analyzable instances, and they negotiated discrepancies by referring to the instance in the transcript and comparing it with instances that had been confirmed as included.
Following the identification of analyzable instances, three researchers analyzed the data for emergent themes in the grounded-theory tradition.13 Each researcher recursively read a single transcript for recurring themes, and the group met repeatedly to discuss and negotiate preliminary analyses (approximately 30 hours per researcher). As the thematic categories in the coding structure evolved, additional instances were analyzed in the transcript to challenge, expand, and refine the categories. Categories were further detailed and subdivided, or revised and deleted, as the coding structure developed and increasing numbers of instances were included. The confirmed coding structure was then entered into NVivo qualitative-data-analysis software and applied to the entire dataset of specific instances by a research assistant, following intensive training in the codes and their definitions.14 The research assistant met with a member of the research team following the analysis of each focus group's transcript to verify the appropriateness of the coding.
NVivo facilitates axial coding, whereby instances may be cross-coded if they involve more than a single issue (e.g., “accountability” and “communicative violation”). As a result, the sum of the instances coded in all sub-categories is greater than the total sum of instances reported in the focus groups.
Students had no difficulty recalling and reporting instances of what they perceived to be professional lapses, suggesting they have a high level of awareness about professionalism as an issue. The application of inclusion/exclusion criteria yielded 48 specific instances in which students reported lapses in professional behavior and 24 nonspecific descriptions, which were not analyzed further. (Examples of the types of phrases that led to exclusion of excerpts are shown in Table 1.) Numbers and types of lapses were similar across the three sites, despite their diverse curricula and school cultures (e.g., problem-based learning versus traditional).
Through grounded-theory analysis of the 48 specific instances, a coding structure of recurring thematic categories was developed. Instances were coded along three dimensions: context, issue, and conflict. The context (see Figure 1) and conflict (see Figure 2) dimensions are described briefly to provide sufficient background; however, the focus of this paper is on the issues identified by the students (see Figure 3). It is not our intention to imply that the instances that follow are by definition professional lapses, but their perception as such by students is critical information for educators. For the purposes of this paper, we refer to student-reported lapses as “lapses.” A detailed description of the recurring issues in the reported instances follows, with supporting examples from the focus-group data.
This category contained lapses in which there was inappropriate delivery of information. In some instances, the lapse involved the content of the communication (e.g., calling a patient “a wimp”); in others, it involved inappropriate delivery of otherwise appropriate content (e.g., discussing, in the patient's presence, the fact that the patient could die). This category was further subdivided into lapses involving communication to patients (e.g., introducing a student as “doctor”), communication about patients (e.g., labeling patients or using derogatory terms), communication to other heath care professionals (e.g., a student getting yelled at by another service when asked for a consult), and communication about other health care professionals (e.g., saying members of another department are “not that bright … so just over expand … just say things so they might think it's more urgent than it is”).
Reports were included in this category if the student conveyed a sense of tension or difficulty regarding the responsibilities and constraints of either the role of the student himself or herself or the perceived roles of others (e.g., attending physicians or residents). One example is that of a student who reported being upset that “no-one” told a patient that his prognosis was poor, but when asked why he or she didn't tell the patient directly, stated “I'm the student! That's totally not my place.” In another example, one student reported that his resident had to leave the hospital while on call, and arranged for another resident to provide backup. When the student went to the backup resident for assistance, he was told “I only want to cover emergencies here. Wait till the other guy comes back.”
Objectification of Patients
This category was defined by those lapses in which patients were treated as objects. Two subdivisions emerged: lapses in which individuals ignored patients (e.g., an attending physician discussed a case as if the patient were not in the room) and those in which patients were treated as vehicles for learning (e.g., using a patient solely for teaching purposes without being explicit).
Lapses were included in this category if they involved an inappropriate lack of duty or responsibility on behalf of the student or others. This lack of responsibility could be either to colleagues (e.g., not showing up on call) or to patients. The lapses involving patients were further subdivided into those that involved avoiding patients (e.g., leaving “difficult” patients to the end of rounds when there was no time left) and those that involved failure to disclose important information to patients (e.g., not telling a patient the results of a test, despite being asked repeatedly by the patient).
This category included inappropriate or unnecessary pain or injury to patients (e.g., not providing analgesia for a procedure despite the patient's expressing pain). While only one instance clearly fit into this category, this lapse's extreme nature and the discussion it fueled in the focus group and in the analysis process warranted its inclusion in this preliminary schema.
Reports were included in this category if subordinates were placed in the middle of a struggle between superiors. For example, one student reported being put in a difficult position when the attendings from two different consulting teams had differing views about patient management. Each attending had told the student to follow his or her orders and disobey the other's, rather than speaking directly to the other and reaching a consensus.
Our study revealed six critical issues in professionalism that medical students experienced in the clinical setting. The focus-group method facilitated students' frank discussion of specific lapses that actually occurred, not generalizations or hypothetical events. Furthermore, the recurrence of findings across three sites, diverse both geographically and pedagogically, suggests the broad relevance of the dominant themes discussed below.
At first glance, the issues revealed in this study may appear as yet another list of aphorisms. Our intention, however, is not to supplant the existing frameworks, but rather to supplement them with a systematic appraisal of students' perceptions. The following example helps illustrate how the application of this supplemental framework enhances our understanding of professional lapses as students perceive them:
I saw a nurse start talking about law-suits, and suggesting that “You know, this is very important because this woman could end up dying here and then the doctor could get sued …” and starts talking about this in front of the patient. And discussing possible death in front of the patient, and it turned out that that woman's husband was an attorney who defends malpractice … and [the nurse] actually got terminated and then reinstated because he threatened to launch another lawsuit against the university.
Using the traditional set of concepts, how might one describe this lapse? The closest fit (for example in the ABIM's taxonomy) would be “respect for others,” reflecting the failure of the nurse to respect both student and patient in the instance. Such a categorization, however, does not allow us to highlight and consider the student's own emphasis on communication. The student focuses on “talking,” “suggesting,” and “discussing” in the report, directing our attention to the nature and context of the nurse's communication. Therefore, our taxonomy codes this report as a “communicative violation.” The communicative violation code highlights the role of discourse in professional encounters: in this instance, the professional lapse derives possibly from the content of the nurse's statement (warning the student about the patient's dying and potential lawsuits), but largely from its context (communicated “in front of” the patient). The student potentially learns as much about professional communication as about professional respect through this event.
The term “respect” alludes to the ideal being breached. The term communicative violation additionally emphasizes the action inherent in the lapse. The choice of term is a critical one, for terms screen our perceptions: they direct and deflect our attention. What we choose to name something has a direct impact on what we are prepared to see: if our terms are abstract, so too will our perception of the phenomenon be abstract. This is problematic, for professionalism is about what we choose to do, our enactment of values, not simply our profession of them. Therefore, if we simply call the above instance “respect,” our attention is directed to the level of abstract value and potentially deflected from the level of concrete action. Communicative violation and the other terms from our taxonomy are grounded in students' own descriptions of professional behavior, and, while they do not exclude issues of value, they purposefully highlight issues of behavior.
Consider another, more complex, example:
I had the same thing, where I didn't [tell a patient their diagnosis] because —and that's what I was asking myself too. Because what? Who's going to do it? The woman had just had [an organ] transplant, didn't have a chest x-ray for some reason, had the chest x-ray after the operation, has terrible lung cancer, and nobody would tell her. So I had to go and examine her like every single day, and her family's there, and she's had this … transplant, and I know she's got inoperable lung cancer. But my attending didn't even want to tell her, because he said to leave that to the other attending. So right, I couldn't do it. But everyone was like, why not? Would that be unprofessional to tell her? I don't think—or is it unprofessional to be like, “So … have you had any abdominal pain over night?” when it's not really the pressing issue, and … yeah.
Described using traditional terms, this example might be categorized as “accountability”: the student knows of a serious, life-altering diagnosis and with-holds this information. The label “accountability,” while it appropriately reflects the concern about disclosure, potentially deflects our attention from the student-centered action issue: choosing to enact the “student” role or the “physician” role. Our term “role resistance” highlights this act of choosing. Torn between the learner's contract with the teacher and the caregiver's contract with the patient, the student is confused, chafing against role constraints, and her professional dilemma manifests itself as an oscillation between alternative actions. As learner, there are potential repercussions to contradicting the attending physicians. As caregiver, there may be a duty to disclose the diagnosis.
This instance is not just another example of the well-documented tension between education and patient care.15 The student's learning is not the primary issue here. Rather, the issue is resisting the boundaries of the “student” role, particularly the boundaries of authority. Enacting role resistance is not necessarily a professional lapse, although it may be in certain contexts. Role resistance is, however, a necessary and complex feature of students' socialization. At some point in their development, professionalism demands that students assume authority and accept the responsibility that accompanies it. The value of accountability is certainly part of the student's report, but without attention to the action of role resistance, we would overlook the dynamic professional process this student is engaged in. This instance is part of an incremental series of contextual decisions that lead to the shedding of the student's role and the assumption of the mantle of professional accountability.
Our student-based taxonomy enhances traditional frameworks for interpreting professionalism in three ways. First, as their titles suggest, our taxonomy approaches professionalism through an emphasis on concrete behaviors. Second, our focus on behaviors directs attention towards a contextual understanding of individual professional behavior.9 Third, our taxonomy is grounded in students' own descriptions of professional lapses. These advances are important, because the behaviors that students report as salient do not map easily onto standard definitions. This approach allows us to appreciate what students perceive to be lapses, an appreciation that may be particularly vital when the lapses reported do not necessarily strike experienced physicians as such. If students are perceiving lapses where faculty may not (such as in our last example in which the physician may have had legitimate reasons for not disclosing the diagnosis), then faculty's behavior is shaping students' developing professional judgment when we may least expect it and, therefore, not address it.
Other focus-group studies have asked students to discuss “ethical dilemmas” rather than professional lapses, thus failing to capture issues such as communicative violation.9,15 Additionally, many studies of professionalism have been end-of-training surveys, which may suffer from recall bias and impede the ability for researchers to distinguish between rumors or generalizations and specific events.2,3,4 Finally, the data-analysis methods are often inadequately described in the literature on professionalism, obscuring the processes by which themes and categories were derived. This obscurity makes it difficult to ascertain whether such findings support—or merely reproduce—traditional abstract conceptions.
Our work was a preliminary effort to identify and resolve the discrepancy between abstract definitions of professionalism and students' interpretations of professionalism in context. Of course, the students who volunteered to participate in our focus groups may represent a particular segment of the medical student population at the three schools studied (i.e., those most interested in or sensitive to professional issues). In addition, the relatively public context of focus groups may have inhibited personal admissions.
Notwithstanding these issues, the focus-group method we employed facilitated the discovery of a richer set of students' stories than has been reported in other studies. Furthermore, the grounded-theory approach to data analysis helped guard against the preconceptions embedded in traditional abstractions. By allowing grounded themes to emerge from the stories themselves, we excavated unique categories of what students perceived to be professional lapses, which have not been previously described. Understanding students' perceptions in this domain is essential to the development of effective educational interventions that bridge existing theories and student experience.
1. Sheehan IH, Sheehan DV, White K, et al. A pilot study of medical student “abuse”: student perceptions of mistreatment and misconduct in medical school. JAMA. 1990;263:533–7.
2. Hicks L, Lin Y, Robertson DW, et al. Understanding the clinical dilemmas that shape medical students' ethical development: a questionnaire survey and focus group study. BMJ. 2001;322:709–10.
3. Satterwhite RC, Satterwhite III WM, Enarson C. An ethical paradox: the effect of unethical conduct on medical students' values. J Med Ethics. 2000;26:462–5.
4. Baldwin DC, Daugherty SR, Rowley BD. Unethical and unprofessional conduct observed by residents during their first year of training. Acad Med. 1998;73:1195–200.
5. McIlroy A, Taylor P. Medical schools “everywhere” rife with unethical practices. The Globe and Mail, Toronto, ON, Canada, 2001 Mar 24, Section A1.
6. Novack DH, Detering BJ, Arnold R, et al. Physicians' attitudes toward using deception to resolve difficult ethical problems. JAMA. 1989;261:2980–5.
7. American Board of Internal Medicine. Project Professionalism. Philadelphia, PA: American Board of Internal Medicine, 1995.
8. Cruess RL. Cruess SR. Professionalism: an ideal to be sustained. Lancet. 2000;356(9224):156–9.
9. Ginsburg SR, Regehr GR, Hatala R, et al. Context, conflict, and resolution: a new conceptual framework for evaluating professionalism. Acad Med. 2000;75(10 suppl):S6–S11.
10. Green MJ, Farber NJ, Ubel PA, et al. Lying to each other: when internal medicine residents use deception with their colleagues. Arch Intern Med. 2000;160:2317–23.
11. Ginsburg SR, Regehr G, Lingard L. To be and not to be: the paradox of the emerging professional stance. Med Educ. In press.
12. Strauss AL, Corbin J. Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory. 2nd ed. Thousand Oaks, CA: Sage Publications, 1998.
13. Glaser BG. Basics of Grounded Theory Analysis. Mill Valley, CA: Sociology Press, 1992.
14. Kelle U. Computer-Aided Qualitative Data Analysis: Theory, Methods, and Practice. Thousand Oaks, CA: Sage Publications, 1995.
15. Christakis DA, Feudtner MA. Ethics in a short white coat: the ethical dilemmas that medical students confront. Acad Med. 1993;68:249–54.