As third-year medical students are immersed in clinical work, they begin to apply the knowledge, skills, and values they have acquired in the classroom, develop professional norms, and confront the behaviors and attitudes of superiors in the medical hierarchy. Students may have little difficulty identifying exemplary or substandard clinical role models. However, the majority of their superiors will likely function in a middle ground where most ethical issues are subtle and answers not self-evident. Indeed, medical students gain much of their clinical experience in arenas replete with unintended or mixed messages from faculty and housestaff about what constitutes ethically acceptable behavior.
The medical ethics education literature documents a persistent mismatch between the issues taught to medical students in required bioethics curricula and texts and the dilemmas students typically face in their clinical experiences.1–4 Nevertheless, some investigators have attempted to rectify this mismatch. For example, recommendations to reform curricula have included addressing specific issues such as whistle-blowing and fostering specific attitudes, habits, and rationales for professional behavior.5–7 One report frames elements of professionalism in terms of student-specific dilemmas in order to teach developmentally appropriate ethics to medical students.8 A recent medical ethics education text presents archetypal dilemmas confronted by medical students and residents to assist readers in learning moral decision-making skills, either for their own individual purposes or in the context of a formal curriculum.9 Such skills include recognizing moral dilemmas as they arise, generating possible courses of action for the moral stakeholders, and grounding rationales in ethical principles and concepts.
A good curriculum will allow educators to elicit from students what they think are the bread-and-butter issues and then respond to them. Borrowing a metaphor from the cultural competence literature, we aimed to “go where [our students] live”10 by systematically examining the ethical conflicts they report encountering in the context of their clinical work. We aimed to compare the ethical conflicts reported by the students we studied to those reported in the literature. Finally, we aimed to gain insight into the bioethical knowledge, skills, or attitudes acquired by third-year students that might compel new or renewed emphasis in formulating national educational objectives, standard curricula, and texts.
We examined consecutive case write-ups submitted as individual course assignments, in which students analyzed ethical issues encountered during their clinical clerkships. These write-ups were a component of the State University of New York (SUNY) Upstate Medical University’s required third-year bioethics course during the academic years 1999–2000, 2000–01, and 2001–02. We devised our analytic methods from the first two years of the data (1999–2001, or cohort 1) and then tested the same methods on data from the subsequent year (2001–02, or cohort 2). Both analyses yielded similar results across multiple domains (e.g., distribution of patient characteristics, specialties involved, and primary ethical issues identified), validating our methods devised for cohort 1 and allowing us to pool the data for this report.
The bioethics course we studied follows a small-group format, where each group is led by a member of the ethics faculty. All assignments are handed in to each student’s small-group faculty leader. In the first study year, students attended the course weekly during their medicine clerkships. In the second and third study years, the course met monthly during students’ entire third years, such that in any given meeting students might have been on any of their clinical clerkships. To examine whether the different course schedule in the first study year might confound the results, we conducted separate analyses with and without the first year’s data; results of both analyses were similar, again allowing us to pool the data.
The course requires students to write two or three brief papers (assigned in the fall and spring) describing and analyzing a clinical case they are part of that presented an ethical issue. All students receive the same instructions to state the primary ethical issue, identify the moral stakeholders anonymously, identify the values involved, describe alternative positions or outcomes, and cite supporting relevant literature.
We scanned cases in their entirety into a computer database. Data tabulated from each case included the patient’s age and gender, the specialty of primary service, and location of service (ward, intensive care unit [ICU], outpatient setting, urgent/emergency setting).
To develop a typology of student-identified ethical issues, we devised a coding process for cohort 1 cases. By referring to standard bioethics texts,11,12 studies of medical students’ ethics education,1–4 and national learning objectives,13–15 we derived a comprehensive checklist of ethical issues against which we tallied the student-identified ethical issue from each case. We refined this checklist in an iterative process by piloting randomly selected subsets of our database. Multiple readers (CVC, KFL, and others) coded the pilot cases against the checklist, adding new issues to the checklist as necessary, until no further issues were identified. Since we were specifically interested in students’ understanding of the ethical import of clinical situations they encountered personally, we coded the ethical issues as they were perceived and identified by each student, even if coders perceived other issues in the cases. Coders’ perceptions of ethical issues did not match students’ in 16 (2.3%) of the cases.
One of us (CVC) and two research assistants trained by CVC coded all cases independently. Each case was ascribed to a particular code only after consensus was achieved among the three readers, who discussed areas of disagreement until all discrepancies were resolved. In 22 cases (3%) encompassing a wide variety of primary ethical issues, disagreements could not be resolved. One of us (KFL) arbitrated these disagreements. The checklist we devised for cases in cohort 1 fit virtually all of the cases in cohort 2. Five new issues from the cases in cohort 2 were added to the checklist, representing 18 cases, or 2.6% of the entire data set. None of these issues retroactively applied to any cohort 1 cases. The issues for cohorts 1 and 2 are shown in Table 1.
The case assignment is a core component of the third-year bioethics course curriculum; its purpose has been (and continues to be) educational, not research driven. Students were instructed to mask the identities of all moral stakeholders and clinical sites. Cases in which some identifiers inadvertently slip through were rendered anonymous by a research assistant prior to scanning the papers into the database. Each student author name on each case was replaced by a code number; after the cases were scanned in, the record of student names was destroyed. SUNY Upstate Medical University’s Institutional Review Board deemed our study exempt from committee review, based on the secondary use of existing educational data and removal of all traceable identifiers from the cases.
From July 1999 to June 2002, 327 third-year medical students submitted 688 case write-ups as course assignments, 675 (98%) of which presented patient-related ethical dilemmas. We eliminated two cases, one plagiarized and the other an essay on a theoretical topic, from further analysis. The 686 cases we analyzed occurred on wards (69%), in outpatient settings (21%), ICUs (6%), and urgent/emergency settings (4%); these various settings were located in a state university hospital, a private hospital, a Veterans’ Administration hospital, hospital-affiliated clinics, and in private offices. In the 675 patient-related cases, 297 (44%) involved female patients; ages of all patients were evenly distributed between birth and more than 80 years. Figure 1 presents the distribution of cases by students’ clerkships as they corresponded to the patient’s primary service and the study year. Clerkship durations varied from four to 12 weeks.
Primary ethical issues
Our checklist comprised the 40 specific primary ethical issues depicted in the cases, which we organized into seven general categories (see Table 1). The highest proportion of ethical issues fell into the general category of decision making regarding treatment (n = 294 [43%]). In addition, three areas were particularly interesting to us in terms of the implications they have for education in ethics: deliberate lies and deception in the context of medical care (communication), discriminatory treatment of patients (justice), and students’ reluctance to speak up about ethically troubling behaviors for fear of reprisal (which cut across all categories)—e.g., receiving a poor grade or evaluation, or being ostracized. We discuss examples of these three areas, as well as other illustrations of our data set, below. We have edited students’ case write-ups for length.
Decisions regarding treatment.
Within this category, the most common specific issues were the patient’s/surrogate’s right to refuse treatment (n = 48) and the patient’s/surrogate’s insistence on futile treatment (n = 46). One student offered the following case, which exemplifies a treatment dilemma regarding a patient’s right to refuse:
Mr. A is a 52-year-old with diabetes, vascular disease, and neuropathy admitted for left leg ulcers. Two years ago, Mr. A had a right above-the-knee amputation because of ulcerations unresponsive to antibiotics. Unfortunately, the ulcers on his left leg have progressed and amputation is deemed necessary. Mr. A believes that once his fever goes away, amputation will be unnecessary. Despite days of IV antibiotics, Mr. A remains febrile. If they amputate now, the surgeons believe the infection can be halted and a prosthesis fitted. However, the longer Mr. A waits, the greater the risk of infection or even death. The patient’s medical attending and team agree with the surgeons. They have strongly urged Mr. A to consent to the amputation, but he is unwilling. The fallacy of his reasoning has been discussed with him, but he still refuses.
Communication, lies, and deceptions.
Deliberate lies and deceptions was the single-most commonly described issue out of the 40 primary ethical issues we identified. Subsumed under the general category of communication, it accounted for 10% of the entire sample (n = 68). The general category of communication also included breaches of confidentiality (n = 44) and poor communication with patients, families, and other health care providers (n = 39). One student gave an example of the negative impact of a physician’s deception:
Mr. B comes to see Dr. A to have a mole removed from his face. Mr. B has no objection to having a medical student, Ms. MS, watch, but he specifically says that he would like Dr. A to do the procedure himself. Dr. A does not respond to this directly. Dr. A believes that Ms. MS is perfectly capable of removing the mole under his direction and that Mr. B won’t suffer any harm. He places the drape over Mr. B’s face and then motions to Ms. MS to pick up the scalpel and remove the mole.
By contrast, some students perceived deliberate lies and deceptions in a positive light, as this student’s write-up illustrates:
Baby C is a 12-month-old African American girl who presents to the clinic with her mother for a round rash, which her mother thought was ringworm. The doctor confirmed that Baby C had ringworm and prescribed a topical antifungal. The mother asked the doctor to write the prescription in her own name because she had insurance and her daughter did not. If the mother had to pay out of pocket, she could not afford it. The doctor knows she should write the script in the name of the patient with the diagnosis. But if she does, there will be noncompliance because the family cannot afford the medication. If she does not help this family out, it may cause tension between them and make future treatment more complicated. The doctor wanted the baby treated, and wrote the prescription under the mother’s name. I think that this was the best decision. The doctor acted out of duty to serve her patient, even if she needed to bend the rules.
Justice and discrimination.
The general category of justice (n = 61 [9%]) included cases in which patients received discriminatory treatment based on their personal characteristics, such as socioeconomic status, self-induced illness, and prisoner status (n = 39), as this student’s write-up illustrates:
I was part of a team treating a prisoner with his “worst headache ever.” He was an HIV positive IV drug user. His CT showed ring-enhancing lesions and his serum was positive for antitoxoplasma IgG. While rounding one morning our resident appeared to be a little rougher than usual while palpating this patient’s abdomen. The prisoner yelled, “Hey, take it easy, man!” The resident stopped but made no apologies. While leaving the room the resident said, “Prisoners are all the same. They’re all looking for an extended vacation in the hospital.” He later commented that prisoners prevent patients who truly need beds from getting admitted to the hospital.
Other cases we assigned to this category described a different sort of justice issue—excessive treatment given to a family member of medical personnel—or even more subtle forms of discrimination that demonstrated inconsistencies in care. For example, a medical student participating in the abdominal surgery of a boy with cerebral palsy was instructed to close the incision with staples. One week later, the same student assisted the same attending physician in a similar operation on an otherwise healthy, pretty young girl. The student reported being told to close the patient with sutures so as to avoid scarring:
I followed [the surgeon’s] instructions but regrettably never asked why we had stapled the other patient. Was it acceptable for that [cerebral palsy] patient to have scars?
Various ethical issues relating to one’s role or status as a student accounted for 69 (10%) cases. Examples included students learning to treat patients over the patients’ objections or without consent when patients were deceived specifically about the student’s status (rather than deception more broadly), and learning on patients when students were inadequately skilled or supervised. Students also described uncertainties about a variety of other less well-understood responsibilities, as this write-up exemplifies:
I was the third-year medical student on rotation in an outpatient office. The attending had a very full schedule with several emergency add-on patients. As a result, he became rushed and very rude towards the office staff, often swearing and shouting. Nonetheless, he was very pleasant to me and took time to evaluate my assessments and to discuss several cases. The attending [at first] remained polite with the patients but eventually became short with them and did not address all of their concerns. In my opinion, these concerns seemed trivial to him but worrisome to several patients. Should I have intervened on the patients’ behalf? I felt I should show respect towards the attending and felt obligated to recognize my role in the office. I also felt obligated to assist the patients. This scenario made me question allocation of precious time and personnel resources.
Fear of speaking up.
One theme cut across all ethical issues and specialty services: students’ avoidance of speaking up for fear of reprisal (see Table 2). In contrast to how they discussed other types of issues, students mentioned this fear as a rationale for their action (or inaction) in response to the ethical dilemmas they experienced, not as the primary ethical issue in the case. Explicitly cited in 81 papers (12%), fear of speaking up was mentioned in cases depicting 18 (45%) of the 40 specific ethical issues on our checklist, including 36 (52%) of all cases depicting student-specific issues and seven (24%) of all cases depicting quality of care. Fear of speaking up was invoked most frequently in cases about patients on the surgery [20 (19%), p < .025] and obstetrics/gynecology [15 (22%), p < .01] services, and least frequently on the pediatrics [zero (0%), p < .05] and internal medicine [23 (8%), p < .025] services.
Some students did not speak up when situations troubled them even when a patient’s safety hung in the balance, as the following excerpt shows:
My surgical team was rounding, rushing in order to make it to morbidity and mortality conference on time. I was in the hall recording vitals when my resident dashed into a room, palpated the patient’s abdomen, and moments later emerged announcing “Type I exam.” She mumbled a plan and signed halfway down the page. I knew what she expected me to write in: Heart RRR, Lungs CTAP, Abdomen BS+, etc., followed by the plan and my signature. I also knew that she had lost her stethoscope a week before, so she could not have listened to the patient’s heart, lungs, or abdomen. The clock was ticking. Should I take the time to examine the patient myself? What were the consequences for me? What were the consequences for our patient?
Students invoked justifications for not speaking up such as their rudimentary clinical knowledge or judgment, possible negative effects on their evaluations and grades, and their subordinate position in the medical hierarchy. In the following excerpts, two students describe their differing views on speaking up:
As students, we become overly concerned with grades. Students should speak up when it concerns the patient. If we do not, we are as guilty as the individuals who violate patients’ values.
It is hilarious and bizarre to expect any subordinate in the medical hierarchy to uphold ethical standards in defiance of their superiors. The repercussions will be too great and the benefits are too uncertain.
Representative phrases used by students to describe this fear included wanting to be “a team player” rather than appearing to “rock the boat”; citing the “ethics versus evaluation” dilemma; and valuing the “relationship with [a] preceptor” over the relationship with a patient.
Variation by specialty.
The distribution of leading ethical issues according to the specialty service caring for the patient is shown in Table 3. This table lists issues in the 620 cases that occurred only on the required clerkships over the three years of data collection. We excluded 66 cases in which the student authors classified the specialty as “other” (e.g., chemical dependency clinic, pain clinic, physical medicine and rehabilitation, radiation oncology) or “unknown.” Some issues were specialty specific, such as decision making for minors in pediatrics and treatment over objection in psychiatry.
By virtue of its large sample size, our study constitutes a robust addition to the extant descriptive literature on the ethical dilemmas that arise for medical students during their clinical clerkships. Many cases present readily apparent bioethical issues, confirming the relevance of objectives from professional organizations13–15 and widely respected bioethics texts 11,12 such as respect for patient privacy and confidentiality altruism, and maintenance of one’s own professional competence. Our data mirror many of the ethical dilemmas of third-year students as reported in the medical education literature, such as a patient’s right to refuse treatment, students practicing procedures on patients, and students striving to be “team players.”1–4,6,9 Importantly, our students discerned ethical dilemmas not only in “usual and customary” situations such as how to respond to a patient refusing treatment, but in small, seemingly incidental situations—a physician closing a disabled child’s skin incision with staples instead of sutures, or a student removing a mole at the behest of the attending and against the patient’s wishes. Additionally, our study presents an explicit analytic method devised for evaluating student cases and potentially suitable for analyzing ethical dilemmas in terms of clinical rotations.
Soliciting students’ specific impressions of their clinical experiences remains worthwhile even though the literature provides thoughtful presentation and discussion of many archetypal cases. The pedagogic process of working through students’ own cases transforms ethics from a theoretical ideal into an actual code of professional conduct2 capable of cultivating moral habits of action. Only in asking students to recount their actual experiences will medical educators be able to answer students’ particular questions and guide them through their conflicts. For example, we were surprised to learn that a morally problematic issue as basic as lying would emerge with such salience for our students. We could not have made this discovery without having asked for their experiences. Through this process we gained insight into three particular areas that compel further educational emphasis: deliberate lies and deception, discriminatory treatment of patients, and students’ fear of speaking up.
Deliberate lies and deception
We believe that the emergence of deliberate lies and deception as the most commonly cited individual issue in our data distinguishes our work from prior similar reports. For example, Huijer et al.4 did not list deception per se in their analysis of 522 student ethics cases. Other research reports have presented issues derived from student ethics cases, including physician behaviors perceived by students as deceptive.1–3 However, in these reports, the deceptions have been presented as embedded in other issues (e.g., withholding diagnostic or prognostic information) and did not clearly constitute the most commonly encountered dilemmas. A content analysis of required student papers on an ethical issue of students’ own personal interest did not cover deceptions in the context of medical care.16
By almost any measure, a physician’s or medical student’s deliberate lies or deceptions to patients is unethical. However, a consequentialist view of physician deception, such that commitment to patient welfare or confidentiality justifies a lapse in honesty, is not new to the medical literature.17 Students may feel forced to choose between competing, albeit worthwhile, values. For example, in the facial mole case, the student weighed her need to learn and her desire to establish a positive relationship with her preceptor against her intention to respect the patient’s wishes. In the ringworm case, the student observed the attending defraud the insurer by falsifying the prescription rather than risk losing the patient’s trust. Both students learned firsthand that in clinical situations the truth often competes with other compelling interests. Simply asserting that truthfulness must be maintained is insufficient; these cases demonstrate that students continue to lack the necessary tools to navigate these dilemmas with moral sensitivity and commitment.18
Discriminatory treatment of patients
We found cases in which patients were treated unfairly. These examples of overt or subtle discrimination against certain types of patients violate the letter and spirit of physicians’ codes of ethics and conduct.13–15 That discrimination persists and is tolerated in medical educational settings despite these codes remains largely unexamined in the medical literature. Our students’ cases may alert the profession to a type of physician behavior insufficiently acknowledged publicly, behavior that may be emulated by students just as they adopt other types of behavior from their mentors and educators.
Before designing strategies to encourage and equip students to handle discriminatory attitudes and behaviors successfully, faculty must first help students distinguish between situations in which differences between patients may indeed matter and those in which differences should not be considered. For example, defensible medical, patient-related, or safety reasons may have motivated the difference in incision closure between the boy with cerebral palsy and the girl. Faculty and housestaff regularly provide medical students with explicit descriptions of their therapeutic rationales and decision analyses; students understand they can expect or request such explanations. Our findings reinforce the relevance of analogous educational explanations—an “ethics differential”—to ethical choices in patient care.19
Students’ fear of speaking up
Students’ fear of speaking up appears in other published reports17,20,21; unsolicited, it was overtly described in 12% of the cases we evaluated. Our data suggest that some students justified not speaking up because they believed they were in a training environment in which it was not only acceptable to subordinate patient welfare to one’s own academic success, but also a matter of academic survival.20 Our study cannot distinguish whether students’ fears were more imaginary than real. However, students who chose not to speak up described an academic climate that seemed unsupportive of demonstrations of moral agency—such as the willingness to risk a potentially higher grade or better evaluation for the good of a patient. We consider such moral agency to exemplify altruism, a core objective of medical students’ and physicians’ behavior.13,14 Yet these students felt that their academic milieu discouraged them—explicitly or tacitly—from adhering to this core objective.
While administrative, instructional, and curricular interventions may facilitate students’ speaking up, all personal risk cannot be eliminated: courage is required. The essence of courage is action in the face of fear.20 Based on our data and those of others,17,20,22 we assert that moral courage is an essential ingredient in medical training and practice. We submit that learning and displaying moral courage is well within the realm of professional expectations for medical students and is an appropriate explicit addition to the formal objectives for medical education.
As we did not specifically ask students if they spoke up about their case’s ethical issue, it is possible that we underestimated students’ frequency of voicing their concerns. In order to gauge this frequency more accurately, identify reasons for speaking up or not, and articulate possible courses of action, future students must be asked explicitly about their fears of speaking up. Our analysis showed that the percentage of cases in obstetrics–gynecology and surgery in which students mentioned fear of speaking up was statistically significantly greater than the percentage in the other clerkships, raising questions about the nonrandom distribution of students feeling intimidated. Although it is tempting to postulate potential explanations, future investigations should also be based not just on our work but also on the empirical work of others.21
Although our students’ clinical experiences occurred at multiple sites, the students were all from a single medical school, possibly limiting our data’s relevance to other schools. The nature of the assignment required students to draw from cases they experienced and, consequently, may have encouraged overrepresentation of “problem” cases, the most obvious ethical dilemmas, or cases for which relevant literature could be cited. While we do not take the student cases we evaluated to represent all of clinical behavior, “problem” and obvious cases do exist. The cases were written up while students actively participated in clinical rotations, suggesting that they represented actual events rather than descriptions of recalled or generalized events.
Normalizing for actual weeks spent per clerkship greatly diminished the uneven raw distribution of cases across specialties (data available on request). Although the number of cases reported per specialty may have been a function of the length of the student’s exposure to that clerkship, we cannot draw conclusions about the frequency of ethical dilemmas per specialty. Other specialty-related explanations for the frequency of dilemmas include the nature of the particular medical issues, the proportion of time students spent in acute care settings, and the moral sensitivity and skills of attendings. Our data suggest that the direction of a doctor–patient conflict over treatment decisions may be somehow related to specialty, although our sample did not have enough power to achieve statistical significance in this matter.
Our cases possess value as descriptions of the range of experiences and practices regarding ethical decision making that remain in students’ minds at the end of the day. It has been suggested that students collectively and passively absorb and adopt normative behaviors and attitudes from their clinical role models.23,24 If so, then efforts at reforming the mismatch between what students are taught and what they experience may be more effective if applied to the entire medical hierarchy. Our findings highlight an insistent need to examine the moral climate of clinical practice, a climate that makes it allowable, easy, and even preferable for medical students to place their relationships with a superior, peer, or team over their duties to patients.
We suspect that many faculty are unaware of everyday situations or clinical practices that may be perceived by medical students as ethically problematic. No matter how shrewdly observant, students may have received incorrect impressions or incomplete information, or simply misunderstood. Nonetheless, students perceive their understanding of the cases as the truth. A standard bioethics course curriculum is unlikely to bridge these differences in perception, but may encourage students to speak up about ethical concerns.
Medical educators could ask themselves what they can do to cultivate students’ courage to speak up. Faculty can commit themselves to solicit and discuss ethical issues regularly on rounds. Addressing student-specific issues can be integrated as a core element within ethics curricula, with teaching moral courage an essential component of this element. Medical educators can expect students to assume responsibility for courageously upholding their own moral standards and their patients’ welfare when safe, confidential structures and mechanisms for reporting perceived ethical infractions, faculty evaluations, and feedback become readily and widely available. Learning objectives and standards specific to professional behavior can be articulated for all in the medical hierarchy so that students can see that even deans, program directors, faculty, and housestaff—along with students—will be held accountable for the courage to implement moral behavior. In the words of Clare Boothe Luce, “Courage is the ladder on which all the other virtues mount.”25
The authors gratefully acknowledge Ms. Melissa Freeman for her assistance with coding and manuscript preparation and two anonymous reviewers for their comments. This work was accomplished without external funding.