Cohn, Felicia G. PhD; Shapiro, Johanna PhD; Lie, Désirée A. MD, MSED; Boker, John PhD; Stephens, Frances; Leung, Lee Ann
Despite a large body of work examining the uses of reflective writing in medical education, a number of studies underscore the need for a more complete understanding of how we can best use this pedagogical modality to probe the ethical impact of medical training on students.1–21 In this regard, written reflective essays become an important means for medical educators to gain insight into their students’ values, biases, perceptions of ethical conflict, problem-solving skills, and take-home messages about the informal curriculum. As Branch3 suggests, analysis of critical incident reports and other such reflective writing can help guide medical teachers in terms of understanding students’ ethical and developmental concerns.
The primary question that we addressed in this descriptive retrospective study was, “How do third-year medical students on a required obstetrics-gynecology clerkship identify moral conflicts, recognize personal and professional values, and draw lessons for addressing future conflicts, as expressed through obligatory reflective essays?” We also wished to determine the types of narratives students favored in addressing these questions. Following the work of Branch,3 Ginsburg et al,22 and others, we performed this retrospective study of students’ values conflicts in real-life clinical situations that would allow us as researchers to learn about students’ actual behavior (at least as reported through their writing). The institutional review board for human subjects research approved this study.
During the period July 2002 to January 2006, all students (n = 299) fulfilling the third-year obstetrics-gynecology rotation at the University of California, Irvine, School of Medicine were required to complete a written assignment as part of an ethics teaching session held toward the end of the rotation. All the students completed this assignment. The obstetrics-gynecology clerkship was selected for this assignment on the basis of student and faculty reports that it evoked the most concerns about unprofessional behavior among students23 and that medical students expressed the most discomfort in challenging supervising team members about ethical issues in this and their surgery clerkships.24
Each assignment was identified by the student’s name. The students were aware that these identified essays would be read and commented on only by the instructor for the session (F.C.). Essays were not graded; completion alone satisfied the course requirement and was necessary to pass the clerkship. Students were encouraged to be frank and open in their papers and in the teaching session and were assured that both their written and verbal comments would remain confidential, as had been their experience in all of their courses with this instructor. There was no restriction on assignment word count. Assignments were turned in one week before a face-to-face small-group teaching session on reconciling personal and professional values. The session comprised primarily discussion among the 15-20 students in the rotation, facilitated by the instructor.
The reflective assignment was guided by a series of six questions (see List 1) designed to elicit students’ perceptions of a moral conflict they had encountered on the clerkship, the larger societal issues underpinning the dilemma, distinguishing personal and professional values, addressing how they handled the ethical conflict and whether they resolved it, and how they might handle similar conflicts in the future. Over time, the instructor noted a number of interesting themes and, therefore, undertook this retrospective study. Although these questions provided more structure and direction than a typical critical incident prompt,3 they bore many similarities to such prompts and produced a similar product. Most significantly, students initiated the selection of a specific ethical incident that they felt had had a profound influence on them.
We used two approaches to code the student essays. The first was a line-by-line thematic content analysis,25 an approach frequently used in textual analysis of reflective writing.26 The content analysis was based in part on the six original questions, but it also drew on grounded theory, through a detailed examination of the text produced in response to each question, to allow categories to emerge from the words and phrases of the essays. Thematic analysis is based on open coding of data—that is, the building of a set of themes to describe the focus of investigation by combining similar words, phrases, and constructs.27 This approach to analysis sometimes involves a simple listing of themes.28 For example, using the open-ended student narratives about personal and professional values, we listed, then grouped thematically, all descriptors relevant to values. The development of themes may be driven by the researcher’s a priori interests (in this case, our six questions)29 or may be derived inductively from the data.30 The general themes of this study were in large part established by the open-ended questions posed, but the specific categories emerged from the words, phrases, and constructs used by the students themselves in their essays.
We, guided by the two faculty members of our research team who had experience in teaching this class, created initial categories by reviewing a subset of students’ papers. These categories were further refined through a consensual process in which four of us iteratively suggested and tested additional hermeneutic conceptualizations of the data. Regular meetings were held during six months to refine the coding instrument. Categories for extracting demographic and background information were also added. Two raters (L.A.L. and F.C.) were then trained in the use of the content analysis coding schema.
For Question 1, we developed categories about the types or nature of the situations students described. These categories included commonly recognized bioethical issues including abortion, maternal-fetal conflicts, treatment refusal, informed consent, end-of-life decisions, family violence, and discrimination. Other categories included student role conflicts (e.g., sense of powerlessness, concern about evaluation, asked to do something the student was unprepared or not supposed to do), team dynamics (e.g., disagreement about the treatment plan, questioning a superior’s judgment), and personal belief issues (e.g., uncertain feelings, asked to do something the student considered wrong, asked not to do something the student considered right). We also recorded when the student stated he or she had not encountered an ethical conflict on the rotation and the reason for this (e.g., open minded, inexperienced, “easy” patients).
On the basis of each student’s description of the situation in Question 1, we as coders made two further determinations. First, the coder identified any contextual factors the student mentioned related to the patient/family and/or the members of the health care team. Among these were cultural issues (e.g., ethnicity, race, religion), patient insurance status, social status, personality, or drug/alcohol use. Second, the coders determined who or what the student suggested was responsible for causing the conflict, including the patient, family, health care team, the student, or society. For both of these categories, all contextual factors and agents or source of conflict mentioned by students were recorded, so the results sometimes totaled more than 100%.
For Questions 2 and 3, in which the students were asked to identify personal and professional values, we developed lists based on the students’ responses. The definitions of personal and professional were left to each student, but class discussion suggested that students generally interpreted personal values to mean those that guided their lives outside their roles as medical students, whereas professional values were those that guided their behaviors with patients and colleagues. All values identified were recorded.
In Question 4, students described sources of ethical controversy in the obstetrics-gynecology clerkship. Coders recorded whether the student viewed the source of these conflicts as primarily societal, patient centered, or educational, and they recorded this information in mutually exclusive categories totaling 100%. Whereas Question 1 generated a detailed analysis of a particular ethical conflict, in Question 4 students discussed controversial issues more generally and abstractly, even when they referenced the earlier conflict.
For Questions 5 and 6, we determined how the student addressed the ethical dilemma he or she described and how the student would address it in the future using the same categories for both what they did and what they would do. The categories generated by the students were do nothing, get help, act in the situation, defer action, educate/empower self, become disillusioned, and other. A student’s response was recorded as “doing nothing” only if the student specifically stated he or she literally did nothing or took no action to deal with the problem. Getting help included, for example, consulting a friend, family member, classmate, supervisor, or authority figure, or registering a concern (e.g., through an incident-reporting system). Acting in the situation included, for example, confronting the perceived wrongdoer, intervening in patient care or a dispute between others, refusing to act as instructed, educating or empowering the patient, or removing oneself from the situation. Deferring action was coded if the student indicated he or she would do something at a later time, such as following the clerkship, after the end of the academic year, or when he or she became a physician. These coding categories were exclusive.
Narrative scholars have expressed concern that, although offering much of value, the dissection of text can also result in overlooking or ignoring the overall sense of the story.31 In an attempt to capture this larger perspective, a global narrative coding for each essay was also developed.
We began with an inductive reading of the student essays from this global perspective. We quickly realized the similarity of thematic categories we were identifying to the narrative framework developed by medical sociologist Arthur Frank.32 Frank identified four major types of illness narratives based on stories told by patients: chaos, restitution, journey, and witnessing. Frank’s theoretical work has recently been supplemented in the literature by articles examining how such stories actually appear in patient accounts.33,34 We appropriated Frank’s four original typologies because, in our judgment, they represent archetypal narrative forms that apply (although, of course, not exclusively) not only to patients and medical students but also to human narratives in general. However, because we were applying the narrative types to narratives from medical students rather than patients, we developed operational definitions of each Frank category that specified their application in this context. Further, we developed three additional categories that we judged necessary to adequately reflect the range of narrative types found in the student essays (compromise, resistance, and no conflict/no problem). Operational definitions for all narrative categories are found in Table 1.
The same two raters were trained in the use of these global coding categories, and then they classified the narrative type each student was perceived to be using overall in his or her essay (see Table 1).
Interrater reliability was determined by computing the simple percentage agreement between the two independent readers. Raw coding of both schemas was entered into an electronic database, and percentages were computed for categories of responses on each variable to produce a descriptive report. In several instances, student responses were coded in as many categories as appropriate; therefore, the total percentage of these responses can exceed 100%. The research team worked together to interpret the possible meanings and significance of the data thus identified.
Class size (approximately 90-100 students per year) and demographics were stable during the three years of our study. The students’ mean age was 23 years. About 50% of students were female. Most students were undergraduate science majors, with biology predominating. The majority had lived and studied in California; 40-50% were white, 40% were Asian, 10% were Latino, 1% were black, and 10% did not self-categorize. Information on religious affiliation was not collected, but the majority acknowledged identity with specific religions and denominations and primarily named Christian denominations, usually Catholicism, within their narrative assignments.
A first round of coding conducted on a sample of about 30 (10%) of the essays yielded interrater agreement for the two primary coders (F.C. and L.A.L.) ranging between 72% and 100% for all but the narrative coding category. Agreement was calculated on 14 separate coded variables, and a partial-credit scoring model was used to quantify coders’ responses on questions for which multiple responses were given. For the assignment of a narrative type describing the essay content, subsequent refinement and recoding achieved 75% interrater agreement, and disagreements were adjudicated.
Nature of moral conflict
In Question 1, students focused on a broad range of conflicts, with no type predominating. Fewer than one third (88; 29.5%) of the students claimed not to have experienced a conflict, and many of those went on to describe conflicts after making this statement. The students claimed that they did not experience conflicts due to being open minded (25; 8.4%), being inexperienced (34; 11.4%), having “easy” patients (10; 3.3%), or other (19; 6.4%). Among the students who identified specific situations (211; 70.5%), the situations described were ones generally familiar to health care professionals and bioethicists. The students described issues related to (from most to least common) personal beliefs issues (102; 34.2%), maternal-fetal conflicts (35; 11.7%), abortion (24; 8%), student role issues (20; 6.7%), treatment refusal (19; 6.4%), societal issues (16; 5.3%), team dynamics issues (15; 5%), informed consent (13; 4.3%), end-of-life decisions (11; 3.7%), family violence (5; 1.7%), discrimination (2; 0.7%), and other (36; 12%).
On the basis of Question 1, the coders determined that the students most frequently indicated that the patient was responsible for the conflict (181; 60%), followed by the student himself or herself (172; 57.5%) and the health care team (100; 33.4%). The students never indicated societal factors as responsible for the conflict. For example, in reflecting on the particular ethical dilemmas that they had personally experienced, students would make statements suggesting that a patient was to blame for continuing to abuse drugs (patient), that the student himself or herself was not open minded enough (student), or that a resident should not put the student in an uncomfortable position—for example, by asking students not to reveal mistakes made to an attending physician or mention requests that the student participate in a procedure the student thought wrong for the sake of his or her education (health care team). Students did not once mention institutional, social, or cultural forces that might impinge on these and other examples of individuals’ behavior.
Similarly, coders determined that contextual issues—such as decision-making capacity, insurance status, social status, cultural issues (<15; <5%), or personality, drug-alcohol, and family (<25; <10%)—were infrequently identified as factors in the ethical conflicts described in responses to Question 1. For example, although a student described a conflict involving concerns over whether the pregnant woman or the fetus should be the primary patient, that student did not identify drug use, social status, or financial situation as contributing factors. Rather, the discussion focused on the student’s beliefs about patient rights and fetal status without addressing possible contextual or larger social issues in which such an issue is situated.
In Question 4, students themselves identified the underlying sources of ethical controversy in obstetrics- gynecology to be society (132; 44.1%) or the patient (129; 43.1%). Conflicts from society included those arising from societal issues such as the ongoing controversy regarding abortion rights or discrimination. Conflicts from the patient included situations such as those due to a patient’s refusal to stop misusing drugs during pregnancy or to accept a physician’s treatment recommendations. Few indicated the medical education system as the source of controversy (28; 9.4%).
Personal and professional values
There was little overlap when the most frequently reported personal and professional guiding values were rank ordered and compared, with one notable exception (see Table 2). Both personally and professionally, respect was among the most commonly identified guiding values (72 [24.1%] and 75 [25.1%], respectively). Otherwise, students most often appealed to religious beliefs (82; 27.4%), the Golden Rule (66; 22.1%), honesty (51; 17.1%), and tolerance (48; 16.1%) as guiding personal values. They identified professional values based mostly on traditionally taught ethical principles including beneficence (71; 23.7%), nonmaleficence (69; 23.1%), and autonomy (65; 21.7%), although justice (25; 8.4%) was rarely mentioned. Values often identified with the health care professions, including altruism, competence, cultural competence, empathy, the Hippocratic oath, integrity, privacy/confidentiality, professionalism, self-improvement, service, social responsibility, and trust, were infrequently identified personally or professionally (0-42; 0%-14%).
Students’ responses to ethical conflicts
The students reported that more than half of the dilemmas were resolved, but, of those, only about half of the students thought they were resolved satisfactorily. Students’ responses to challenging situations were largely divided between the 70 (23.4%) who reported doing nothing and the 106 (35.5%) who reported taking action within the situation. Few (15; 5%) asked for help or tried to empower or educate themselves (28; 9.4%). Very few became disillusioned (7; 2.3%).
Comparisons between what students actually did and what they speculated they would do in a similar situation in the future (see Table 3) indicated that almost none would do nothing in future situations (11; 3.7%), whereas still few (21; 7%) would ask for help. A slightly higher number would act in the situation (136; 45.5%). The largest increase, threefold, was in looking to educational strategies to empower themselves, by acquiring new knowledge or skills, enhancing communication, or self-reflection (104; 34.8%).
More than one third (113; 37.8%) of the students told restitution narratives, followed by journey (48; 16.1%), compromise (47; 15.7%), witnessing (38; 12.7%), resistance (27; 9%), chaos (20; 6.7%), and no conflict/no problem (5; 1.7%; see Table 1).
These findings suggest important insights into students’ perceptions about the moral and ethical conflicts they encounter during their medical education, their personal and professional values, and how those values inform their responses to those conflicts.
First, students seemed to perceive experiences of ethical and moral conflict in clinical situations very narrowly, without much regard to important contextual issues such as the patient’s socioeconomic status, insurance coverage, or culture, as these were rarely mentioned in student essays. Further, despite the 44% of responses (Question 4) in which the students identified conflicts in obstetrics-gynecology as based in societal issues, the actual descriptions of specific situations of conflict (Question 1) focused primarily on individual patient behaviors or individual student attitudes. It was quite telling that in considering the students’ specific ethical dilemmas, we as coders found no instances in which students explicitly identified society as responsible for the conflict and found only rare instances in which they mentioned contextual variables connecting individual patients to larger societal perspectives. Instead, students located responsibility primarily with patients and/or themselves. For example, issues that emerged regarding pregnancy-termination decisions centered on patients’ responsibility to avoid illicit drug use and students’ emotional or religious responses to patient decisions. The larger societal debates related to abortion, the child welfare system, and the role of health care professionals in criminal justice were rarely mentioned, as was the case with factors of ethnicity, religion, culture, or insurance status. Thus, it seemed that although students were aware of broader societal issues, as indicated in their responses to Question 4, that awareness had not made its way into their analyses on a practice level.
This finding, which stands in contrast to those of an earlier study—which reported that although its obstetrics-gynecology clerkship students also rarely focused primarily on social, cultural, or economic issues in writing about ethical dilemmas, they at least acknowledged them in the majority of essays35—is potentially troubling. It suggests that these third-year students had not yet learned to think in a multifaceted way about the factors that influence ethical conflicts and, therefore, may have taken a somewhat simplistic approach in their analyses. Additionally, narrow reflection may result in a perception that responsibility for these overarching societal problems and controversies does not fall within the physician’s job description. Increasingly, physicians are being expected to address the social context in which they treat patients,36 and medical education should prepare students for this role.
Second, with the exception of respect, students drew on significantly different clusters of values to guide them personally and in the professional setting. This finding suggests that these students might want to interact differently (guiding their behavior by the Golden Rule or religious values) with patients if they were not constrained by “having” to be “professional.” For example, many students concluded it was a priori “unprofessional” to explore the autonomously made decision of a patient to refuse medical treatment, although they may have wanted to talk further with the patient. Further, if a student did choose to discuss the matter with the patient, the student would justify his or her behavior as a matter of good medical practice without acknowledging the personal values that may have motivated the discussion.
Our concern here is not that one approach is “right” or “wrong” but, rather, that the students did not seem to have the reflective tools to recognize and better understand the implications of this values dichotomy. Although distinguishing personal and professional values and contexts is not necessarily a problem, we speculate that this lack of awareness of reliance on different values sets may cause confusion and moral distress resulting in unresolved professional dilemmas for the students.
Following Ginsburg and Stern,37 we formed the impression that students generally regarded the professional values they learned as vague abstractions. Their responses demonstrated familiarity with these values but suggested that the students have not internalized them or made them operational (i.e., felt comfortable acting according to the values). Specifically, although they were familiar with principals of bioethics such as autonomy, beneficence, and nonmaleficence, they seemed to regard those as values that the profession somehow required them to activate rather than as anything intrinsic to them personally.
Nor did the students seem to appreciate the complexity of balancing or prioritizing competing principles. They rarely explored how the various perspectives of physician, patient, family member, and medical student might lead to different ethically based conclusions. In fact, they preferred appealing to singular values, perceived as primary, to address the ethical situations that challenged them, but they were unable to justify either the selection of the particular value or the priority they had given it. For example, the students could not explain why respect for autonomy was considered the most important value in a given situation or how it might relate to other competing values such as beneficence or justice. Also of concern, “traditional” values associated with medicine such as altruism, social justice, and service were rarely implicated as either professional or personal values. This finding may indicate a falling away from the traditional values of other-over-self that have been a guiding foundation of medicine for centuries.38
When we compared students’ actual responses when facing ethical dilemmas with how they intended to behave in future similar situations, the two most striking findings were the almost universal conclusion that doing nothing in response to an ethical conflict was a morally bankrupt choice that they would not want to repeat, despite the many pressures they identified, either in essays or in class, that constrained their ability to act. The other difference was that students’ alternative to doing nothing seemed to be more actively cultivating empowerment strategies either through knowledge acquisition, improved communication, or reflection. This suggests that, despite frustrations, students remained hopeful that they could acquire additional knowledge to deal more successfully with the ethical dilemmas they observed. Of concern, however, is that this strategy also suggests a reliance on education to help them deal with conflict, when what they may need is enhanced moral fortitude or more encouraging role models (i.e., knowing more may not always be helpful in working through an ethical dilemma). It was disappointing that students rarely mentioned asking for help when facing future moral conflicts.
One puzzling discrepancy to emerge was the difference between the small number of essays attributing ethical conflicts to the medical education process and anecdotal reports from the instructor regarding the number of times this issue emerged in class discussion. Furthermore, in the specific examples offered by students, we as coders determined that the health care team was responsible for the conflict in approximately one third of the papers, although, as noted in their more general discussions, students only rarely identified the team as the source of the controversy. Faculty experience with class discussions (“venting” by students) suggested that student conflicts involving supervisors might be more frequent than was reported in their essays. Perhaps this type of conflict was underrepresented because it was censored by students when choosing an encounter for the written assignment. We speculate that students immersed within the educational system understandably were reluctant to criticize that system directly in a written format, especially when they could be identified by name. In the safety of informal classroom discussion and assured of confidentiality, they may have expressed more honest opinions.
The narrative typology findings also are thought provoking. We believe this study is the first to expand Frank’s narrative typologies to medical student narratives. Whitehead34 found that narratives of patients with chronic fatigue syndrome showed a trajectory from restitution to chaos back to restitution, subsequently often proceeding on to journey. Although often we found elements of more than one narrative type in the student essays, generally speaking they fell comfortably within a single typology (including the additions we made that emerged from the essays themselves). This may have been an artifact of the assignment in that students were asked to describe a single incident, rather than overviewing, for example, their experiences across three years of medical school.
There was no one “majority” or universal story being told, so we do not believe there was one totalizing, completely dominant influence constraining student stories. The most common story told was the restitution narrative, judged to constitute more than one third of the stories. The frequency of these stories makes sense because it remains the dominant story in modern medicine as well: the “find it-and-fix-it model.”32 The restitution story is certain, predictable, and reassuring. It can also be understood as related to Ginsburg et al’s39 concept of restorying, in that restitution stories take the moral dilemma and smooth it in a way that minimizes or eliminates the problem. Yet, the fact that other types of stories emerged, such as journey, witnessing, and resistance, suggest some cracks in the students’ desire to resolve their ethical conflicts too quickly and neatly. These cracks were further in evidence in the fact that only about one quarter of students reported actual successful resolution of the conflicts they reported.
Interpretation of results
The results must be interpreted in the context of the study design. It focused on a single school at which students were homogenous in age, gender distribution, religious identification, and science major backgrounds; these characteristics may not be representative of students at other schools. We did not compare third-year students with students from other years, and we did not track cohorts longitudinally. In addition, the papers represent only a single snapshot into students’ ways of thinking. The narrative typology coding method has not otherwise been tested, and its internal consistency and appropriateness to the essays are yet to be established.
Further, several aspects of the assignment itself may have influenced our results. Specifically, it is possible that demand characteristics were operating in the sense that student essays were identified by name, and it was known that the ethics professor would be reviewing them. Such knowledge may have encouraged students to write comments that would reflect on them as “good students.” On the other hand, because the essays were written before the ethics discussion, the students could only guess at the professor’s expectations rather than specifically tailor their responses to incorporate material previously presented. Further, coders were surprised at the frequency of “nonprosocial” statements in the essays, which seemed to ignore or contravene material presented in lecture and discussion offered in previous ethics sessions about the nature of ethical dilemmas, their meaning, and how they can be approached. This observation suggests that the operation of demand characteristics was somewhat mitigated.
Strengths of the study
The strengths of the study are notable. A large number of responses were analyzed, the data set was complete, the researchers represented different academic disciplines, and the coders achieved sufficient levels of interrater agreement for meaningful interpretation of categories.
Implications for curricular change
Taken as a whole, these findings can be helpful in informing the development of bioethics/professionalism curricula. They suggest clues to the hidden curriculum,40 indicating that students are learning to tolerate ethical conflicts that are either unresolved or unsatisfactorily resolved in their minds, avoid complexity and uncertainty in problem solving, accept the current state of medical practice as a given, and resist broader perspectives on health care.41 Armed with this awareness, medical educators, and ethicists in particular, can more vigorously challenge such views and pose viable alternatives. Such challenges and alternatives cannot, however, arise only out of ethics courses but must be incorporated into medical education more broadly and role modeled professionally. Some efforts are being made in this regard, such as the implementation of patient safety curricula focused on systems change.42 Although medical students and physicians are entitled to hold varying personal and professional values, we as medical educators can play a role in instilling or at least suggesting the kinds of values we and society deem preferable and, more particularly, in increasing students’ awareness of how their personal conflicts influence day-to-day patient-care decisions.
Further, the study suggests a new conceptual lens for viewing student narratives. Although applying labels to individual experiences runs the risk of minimizing important variations, it also offers a mechanism for examining those experiences for significant commonalities. The themes introduced can suggest patterns of perception and behavior in need of attention and can reveal both problems and areas for encouragement. For example, the prevalence of restitution narratives suggests that the modernist, find-it-and-fix-it medical model is regularly applied to moral and ethical dilemmas. To the extent that such a model is simplistic and reductive, ethics education could focus on more nuanced critical analysis skills and making students aware of other kinds of narrative types, such as journey and witnessing, and encourage their acceptance as legitimate ways of responding to ethical dilemmas.
The professional acculturation of physicians-in-training involves helping would-be physicians to identify their values, sort them out, and manage conflicts when they occur. Acknowledging the tensions between our personal and professional values is a first step that, at least, allows us to openly confront those tensions.43 Our findings suggest that permitting students to identify conflicts arising from discordant personal and professional values may be an important function of the bioethics and professionalism curriculum.
The authors wish to thank Clarence Braddock of Stanford University and Mark Kuczewski of Loyola University Chicago for helpful feedback on the manuscript of this report.
Dr. Lie is supported in part by K07 award HL079256-01, 2004-2009, from the National Institutes of Health.
1 Schoen DA. Educating the Reflective Practitioner. San Francisco, Calif: Jossey-Bass; 1987.
2 Spalding E, Wilson A. Demystifying reflection: A study of pedagogical strategies that encourage reflective journal writing. Teach Coll Rec. 2002;104:1393–1421.
3 Branch WT Jr. Use of critical incident reports in medical education: A perspective. J Gen Intern Med. 2005;20:1063–1067.
4 Wallace D. Using reflective diaries to assess students. Nurs Stand. 1996;10:44–47.
5 Tryssenaar J. Interactive journals: An educational strategy to promote reflection. Am J Occup Ther. 1995;49:695–702.
6 Westberg J. Helping learners become reflective practitioners. Educ Health (Abingdon). 2001;14:313–321.
7 Charon R. Narrative and medicine. N Engl J Med. 2004;350:862–864.
8 Shapiro J, Kasman D, Shafer A. Words and wards: A model of reflective writing and its uses in medical education. J Med Humanit. 2006;27:231–244.
9 Baernstein A, Fryer-Edwards K. Promoting reflection on professionalism: A comparison trial of educational interventions for medical students. Acad Med. 2003;78:742–747.
10 Rucker L, Shapiro J. Becoming a physician: Students’ creative projects in a third-year IM clerkship. Acad Med. 2003;78:391–397.
11 Levett-Jones TL. Facilitating reflective practice and self-assessment of competence through the use of narratives. Nurse Educ Pract. 2007;7:112–119.
12 DasGupta S, Charon R. Personal illness narratives: Using reflective writing to teach empathy. Acad Med. 2004;79:351–356.
13 Christakis DA, Feudtner C. Ethics in a short white coat: The ethical dilemmas that medical students confront. Acad Med. 1993;68:249–254.
14 Caldicott CV, Faber-Langendoen K. Deception, discrimination, and fear of reprisal: Lessons in ethics from third-year medical students. Acad Med. 2005;80:866–873.
15 Lindau ST, Bernacki RE, Leeds HS, Tawose OM. Medical students as witness to justice and injustice in patient care: Essays from the ob/gyn clerkship. Paper presented at: American Society for Bioethics and Humanities Annual Meeting; October 2005; Washington, DC.
16 Feudtner C, Christakis DA, Christakis NA. Do clinical clerks suffer ethical erosion? Students’ perceptions of their ethical environment and personal development. Acad Med. 1994;69:670–679.
17 Sandor MK, Sierpina VS, Vanderpool HV, Owen SV. Spirituality and clinical care: Exploring developmental changes in nursing and medical students. Explore (NY). 2006;2:37–42.
18 Mamede S, Schmidt HG. The structure of reflective practice in medicine. Med Educ. 2004;38:1302–1308.
19 Wong FK, Kember D, Chung LY, Yan L. Assessing the level of student reflection from reflective journals. J Adv Nurs. 1995;22:48–57.
20 Hatem D, Ferrara E. Becoming a doctor: Fostering humane caregivers through creative writing. Patient Educ Couns. 2001;45:13–22.
21 Strauss A. Qualitative Analysis for Social Scientists. Cambridge, UK: Cambridge University Press; 1987.
22 Ginsburg S, Regehr G, Hatala R, et al. Context, conflict, and resolution: A new conceptual framework for evaluating professionalism. Acad Med. 2000;75(10 suppl):S6–S11.
23 Papadakis MA, Osborn EH, Cooke M, Healy K. A strategy for the detection and evaluation of unprofessional behavior in medical students. Acad Med. 1999;74:980–990.
24 Clever SL, Edwards KA, Feudtner C, Braddock CH 3rd. Ethics and communication. Does students’ comfort addressing: Ethical issues vary by specialty team? J Gen Intern Med. 2001;16:559–563.
25 Riley T, Hawe P. Researching practice: The methodological case for narrative inquiry. Health Educ Res. 2005;20:226–236.
26 Pitkälä KH, Mäntyranta T. Feelings related to first patient experiences in medical school. A qualitative study on students’ personal portfolios. Patient Educ Couns. 2004;54:171–177.
27 Morse JM, Field PA. Qualitative Research Methods for Health Professionals. Thousand Oaks, Calif: Sage; 1995.
28 Gordon K, Turner KM. Ifs, maybes and butts: Factors influencing staff enforcement of pupil smoking restrictions. Health Educ Res. 2003;18:329–340.
29 Crabtree BF, Miller WL. Using codes and code manuals: A template organizing style of interpretation. In: Crabtree BF, Miller WL, eds. Doing Qualitative Research. 2nd ed. Newbury Park, Calif: Sage; 1999.
30 Kalnins I, Hart C, Ballantyne P, et al. Children’s perceptions of strategies for resolving community health problems. Health Promot Int. 2002;17:223–233.
31 Kuper A. Literature and medicine: A problem of assessment. Acad Med. 2006;81(10 suppl):S128–S137.
32 Frank A. The Wounded Storyteller: Body, Illness, and Ethics. Chicago, Ill: University of Chicago Press; 1995.
33 Auxiliadora C, De Benedetto M, Blasco PG, de Castro AG, de Carvalho E. Once upon a time … at the Tenth SOBRAMFA International and Academic Meeting. J Learn Arts [serial online]. 2006;2:7. Available at: (http://repositories.cdlib.org
/clta/lta/vol2/iss1/art7). Accessed January 29, 2009.
34 Whitehead LC. Quest, chaos and restitution: Living with chronic fatigue syndrome/ myalgic encephalomyelitis. Soc Sci Med. 2006;62:2236–2245.
35 Belling C, Coulehan J. A window of opportunity: Ethics and professionalism in the obstetrics and gynecology clerkship. Teach Learn Med. 2006;18:326–329.
36 ABIM Foundation; American Board of Internal Medicine; ACP-ASIM Foundation; American College of Physicians-American Society of Internal Medicine; European Federation of Internal Medicine. Medical professionalism in the new millennium: A physician charter. Ann Intern Med. 2002;136:243–246.
37 Ginsburg S, Stern D. The professionalism movement: Behaviors are the key to progress. Am J Bioeth. 2004;4:14–15.
38 American Medical Association Council on Ethical and Judicial Affairs. Code of Medical Ethics: Current Opinions With Annotations. Chicago, Ill: American Medical Association, 1994. Available at: (http://www.ama-assn.org/ama/pub/category/2503.html
). Accessed April 1, 2008.
39 Ginsburg S, Regehr G, Lingard L. The disavowed curriculum: Understanding students’ reasoning in professionally challenging situations. J Gen Intern Med. 2003;18:1015–1022.
40 Hafferty FW, Franks R. The hidden curriculum, ethics teaching and the structure of medical education. Acad Med. 1994;69:861–871.
41 Haidet P, Stein HF. The role of the student-teacher relationship in the formation of physicians. The hidden curriculum as process. J Gen Intern Med. 2006;21(suppl 1):S16–S20.
42 Committee on Quality Health Care in America, Institute of Medicine. Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, DC: National Academy Press; 2001.
43 Goldberg JL. Humanism or professionalism? The white coat ceremony and medical education. Acad Med. 2008;83:715–722.