You are in a busy clinic seeing a 15-year-old girl for a “sore throat.” Her physical examination findings are normal, and as you are ready to leave the room she blurts out that she thinks she is pregnant. Her mother is in the waiting room.
Physicians routinely face difficult ethical situations such as the scenario above as they practice medicine. Developing a systematic approach to problems in clinical ethics may enhance the physicians’ effectiveness when confronting these challenges.1,2 Medical students and residents most commonly learn how to approach ethical dilemmas by observing faculty and through hallway discussions about what they could have done differently after a difficult case.3–5 Carving out time for formal ethics education is one step toward teaching specific skills and demonstrating an institutional commitment to this aspect of professional education.
Education about ethical issues should, at a minimum, enhance the learner's ability to (1) recognize ethical issues when they arise, (2) reason through ethical dilemmas and determine a justified response, (3) have an awareness of professional responsibility to know that one must act, and (4) act in the face of ethical conflict.1 Once these basic skills are developed, the learner can further refine those skills through a variety of methods.6,7 Most formal ethics education occurs during the preclinical years, before students become immersed in their clinical experiences. However, novel ways of addressing ethics education in the clinical setting, covering a wide variety of topics, have been developed.8–11A barrier to implementing ethics training in the clinical setting is the perception that educators must have formal training in ethics to be able to teach this topic effectively.12 Because of this perception, and because time for formal teaching is limited during a typical medical school clinical clerkship, it is important that educational interventions be supported by evidence of effectiveness.
We recently implemented a case-based ethics curriculum for third-year students at the University of Washington School of Medicine during their pediatrics rotation. Our goal was to develop a standardized curriculum to help clinician–educators without ethics training teach an approach to solving ethical dilemmas. The objective of this project was to compare the effectiveness of two teaching interventions on students’ recognition, knowledge, and ability to justify actions in common ethical dilemmas.
All third-year medical students at the University of Washington School of Medicine participated in a pediatrics clerkship that included a new ethics component in the academic year 1999–2000. Approximately half of the students completed the six-week pediatrics clerkship at hospitals and clinics in Seattle (discussion group, 80), and the remaining students were located at hospitals and clinics outside of Seattle (case-analysis group, 66). Students were assigned to rotation sites according to personal preference and lottery. The sites outside of Seattle are the WWAMI (Washington, Wyoming, Alaska, Montana, and Idaho) sites, which include hospitals and clinics in Tacoma, Washington; Spokane, Washington; Pocatello, Idaho; Great Falls, Montana; and Anchorage, Alaska. Each clerkship site has designated pediatrics faculty who coordinate the medical students’ education program and use the same curriculum and evaluation criteria. Students evaluate similar numbers and types of patients’ problems and receive individualized instruction from pediatrics faculty. Although the patients’ socioeconomic characteristics are similar among all sites, students in Seattle are more likely to see patients from diverse ethnic populations than are WWAMI students.13,14 The demographics are similar for students rotating in Seattle compared with WWAMI sites with respect to career choice, overall academic performance, and performance during the pediatric clerkship.15 All students receive instruction on ethical issues during their preclinical years as a part of a course on performing histories and physical examinations. On average, 15% to 18% of students at the University of Washington School of Medicine participate in preclinical electives in medical ethics each year.16
The University of Washington Institutional Review Board approved the study.
All students analyzed four ethics cases as a requirement of the clerkship. The case content focused on common issues for pediatricians or scenarios that students are likely to face (see Table 1). Although some topics may have been covered previously in other courses or clerkships (e.g., end-of-life decisions), topics such as adolescent decision making and parental rights to guide therapy were unique to pediatrics (see Table 1). We provided all students with a list of suggested readings, the addresses for several bioethics Web sites, and sample case analyses to illustrate the format and degree of detail appropriate for their responses.17
We compared two teaching methods: written ethics case analyses (case-analysis group) and written case analyses followed by group discussion (discussion group). All students completed responses to the first three ethics cases by the end of the third week of the rotation. All students received written feedback on each of the three cases analyses by the end of the fifth week of the rotation using a standardized evaluation form and individualized written comments (described below). During the final week of the rotation, all students completed a final “take-home” ethics case analysis. Between submission of the first three written case responses and the final case analysis, students in the discussion group participated in a one-hour, one-time group discussion of the first three cases facilitated by one of the pediatrics attendings (SS). An outside reviewer with formal training in medical ethics observed the facilitated session (CHB) and gave specific feedback to the facilitator.
Evaluation and Feedback
The written evaluation form and the facilitated group discussion highlighted the same features of the case analyses. We assessed four components: (1) ability to identify ethical dilemmas, (2) ability to see multiple viewpoints, (3) ability to formulate a plan, and (4) ability to justify their actions (see Figure 1). Content experts (KFE, CHB, DSD) reviewed the evaluation tool with particular attention to completeness and face validity of the ethical topics suggested by each case and the arguments included to assess the completeness of the students’ discussion. Serial group discussions were used to derive a scale for the evaluation of the students’ plans. We developed specific coding criteria to increase interrater reliability. For example, to quantify the completeness of the students’ discussion of the ethical issues, we developed the scale shown in Figure 1. Assessing the students’ plans was particularly difficult, so we assigned incremental credit (one point each) for forming a plan, the degree of detail and feasibility of the plan, and whether the plan addressed multiple ethical issues, which improved interrater reliability (assessed through coding 5% of the sample). This tool was used to give feedback to all students in the study.
One evaluator (KFE) then reviewed and coded a randomly selected, masked subset of the students’ case analyses (48 of 146) using the written evaluation form. This subset of case analyses was selected from students who rotated throughout the academic year and were evenly distributed between the two groups. An initial score was determined by averaging the scores of the initial three case analyses. Both the total score and the scores for each of the four components were determined. These initial scores were then compared with the scores on the final case analysis. Absolute numerical change in scores as well as the percentage of students that performed better, worse, or the same were determined.
Students’ Evaluation of the Exercise
Students’ satisfaction with the ethics exercises was elicited in the form of an open-ended question as a part of the anonymous end-of-clerkship evaluation. All students in the clerkship were asked to complete the evaluation (146).
The Mann-Whitney test was used to examine differences between individual components of the initial case analyses of the two groups of students. Analysis of variance and Mann-Whitney tests were used to examine the change in performance between the initial and final case analyses. Interrater reliability of the evaluation tool was evaluated by using the Pearson correlation coefficient. A value of p < .05 was considered statistically significant. SSPS version 10.05 (SPSS, Inc., Chicago, Illinois) and Statview version 5.0 (SAS Institute Inc., Cary, North Carolina) were used for analyses.
Interrater reliability correlation determined on a random sample of student case analyses ranged from 0.79–0.922 (p < .0001) for all combinations of raters by the Pearson correlation coefficient. The two evaluators with the most experience scoring (KFE and SS) had the highest correlation coefficient of 0.922. The ratings by KFE were used in this analysis because she had used masked data.
Performance on all of the components of the initial case analyses was similar between the two groups (p > .2–.8; see Figure 2A). On the final case analysis, students who participated in the group discussion had a higher absolute increase in scores on total score and the component “ability to formulate a plan” (p = .017 and p = .013, respectively; see Figure 2B). Overall, most students, improved their performance over the course of the rotation (see Table 2). Students who participated in the group discussion had higher total scores (p = .029) on the final case analysis. Additionally, these students’ performances were also more likely to remain unchanged or to improve in the components of identifying ethical issues, seeing multiple viewpoints, and justifying their actions (although these differences did not reach statistical significance). The case-analysis group's performances on the final case analysis declined in some cases, particularly in the component of seeing multiple viewpoints of the ethical issues (p = .078). Common errors that occurred in the final case analysis by all students included not identifying the ethical issues (e.g., failing to recognize that confidentiality was an important issue), or failing to formulate a plan (e.g., resolving the clinical situation without addressing the underlying ethical dilemma).
Of those responding to the open-ended anonymous evaluation at the end of the clinical rotation (134), students in the group discussion (74) expressed more satisfaction with the experience than did students in the case-analysis group. Eighty percent of the group discussion students had positive comments about the exercise, whereas only 25% of the case analysis students gave positive comments. In general, specific negative comments focused on two areas: (1) a desire to have face-to-face discussion about the cases, and (2) interpreting written feedback as a grade. Finally, a large proportion of the case analysis students (40% versus 15% of the discussion students) did not comment on the value of the exercise.
Assessing differences in students’ performances following two different educational interventions allowed us to determine whether group discussions added educational value when teaching ethics in a clinical setting. Although the literature suggests discussion is important for increasing understanding, we needed concrete data in this clinical setting to motivate busy clinical faculty to make discussion a priority in their teaching of ethics. We found that students’ ability to identify and assess ethical problems improved following exposure to the pediatrics ethics topics, regardless of the educational modality used. Including a facilitated discussion session added incremental benefit to students’ performances in the final case analysis and contributed to their improved satisfaction with the educational experience. This latter finding parallels the experience of others teaching medical ethics to students in the preclinical years.18,19
Literature regarding the evaluation of the effectiveness of a structured educational intervention in ethics teaching is limited. The modest increases in performance in both groups in our study can be explained in part by the sensitivity of our evaluation tool. Because the evaluation scale was not large (13 possible points per case analysis), students could not improve significantly because of the high performance on the initial case analyses (average score was 10 in both groups). Additionally, we balanced the degree of detail expected in each analysis with the time students would need to spend writing and researching the case, thus further limiting our ability to determine differences. Alternative measures of moral reasoning such as the Defining Issues Test and Sociomoral Reflection Measure have been used in the preclinical setting, but give no formative feedback to students regarding content and areas for improvement,18,19 thus limiting their usefulness in our setting. Others have proposed using structured assessments (objective structured clinical examinations) to determine the effects of longitudinal ethics education curriculum. However, the high costs of development and implementation limit their application to most clerkship settings.20,21 Our evaluation tool could be revised to include more scoring elements and criteria to distinguish differences in performance, particularly if students had more time to devote to the exercise.
Although most students improved their scores over the course of the rotation, we also saw a significant number of students perform less well on the final case analysis in both groups. Most notable was the decline in the case-analysis groups’ ability to see multiple viewpoints. The absolute decline in performance was small, likely reflecting both the small sample size of our study and sensitivity of the evaluation tool. Additionally, the decline in performance may be case specific (e.g., pregnant teenager), and students may have failed to recognize the ethical complexities in a situation in which the law is clearly articulated. Alternatively, the difference in performances could also reflect the benefit of additional reinforcement of major teaching points in a facilitated group discussion. Furthermore, exploring multiple viewpoints of an ethical issue may be taught best by using additional examples in group discussions.
While teaching these topics during a busy pediatrics clerkship, we had to overcome several obstacles. One of the major barriers to teaching clinical ethics is the faculty's perception that they lack expertise.12 The collaborative model described here capitalized on the interest and teaching abilities of a pediatrician with limited formal training in ethics and the content expertise of several other faculty members in the ethics department. We developed a teaching method that builds on a medical decision-making model—a method of case analysis familiar to all clinicians (e.g., identification of a problem, obtaining additional data, formulating an assessment, and implementing a plan). The method was easy to apply to cases familiar to clinicians and differed from other commonly used approaches7 in that it focused the students not only on identifying ethical dilemmas but also on moving their contemplation toward a plan of action.
An important component of teaching this curriculum was developing a consistent evaluation tool to give feedback to students. We used a checklist because it proved easy to use, and it provided a mechanism to suggest alternative topics to students that could be included in their discussions. The process of developing the feedback tool also facilitated the faculty development of WWAMI clinic site coordinators. In the year following initial implementation, these clinicians took over the teaching and administration of ethics. They were provided with the checklist evaluation tool, texts on clinical ethics, and the background research that we performed while developing the evaluation tool. Five of the six WWAMI sites incorporated active discussion sessions and use the evaluation tool for assessing students’ essays. Three of these sites also incorporated discussion about current cases seen on the ward. In subsequent years, a majority of WWAMI students reported a high degree of satisfaction, and sites with extensive faculty discussions reported > 90% satisfaction as a result of the changes.
This study suggests that formal teaching about ethics enhanced students’ ability to think through common ethical situations, despite its limitations. Although students in the discussion group performed better on the final case analysis than did the case-analysis students, it is difficult to gauge whether the enhanced performance was solely due to participation in the discussion session. Also, we did not determine whether students understood the feedback given to them in written form, thus qualifying its utility as a teaching tool. However, a substantial number of students in the case-analysis group improved their performance, thus making it less likely that they did not understand the feedback or the purpose of the exercise. Personal interaction with interested faculty about the topic may have provided the additional motivation students needed to perform well on the final case analysis, thus confounding our findings. Finally, the improvements in students’ performances may reflect only a better understanding of the evaluation criteria, having seen it modeled in a class discussion.
Others have indicated that students with an interest in primary care may be better able to identify ethical issues during patient care.22 Students who rotate through the WWAMI sites (the case-analysis group) during the third-year pediatrics clerkship enter the fields of pediatrics, internal medicine, or family medicine at the same rate as those who rotate in the Seattle clinics,15 thus making this an unlikely contributor to the differences we found. Similarly, these students were as likely to have taken ethics electives in the preclinical years (18% versus 20%) as students in the discussion group.16
In conclusion, we have shown that implementing a formal ethics curriculum had a positive impact on the ability of medical students to approach these problems. Furthermore, we demonstrated the additional value of a facilitated discussion in improving the ethical analysis skills of third-year medical students. Providing a forum for small-group discussion is a well-received and effective method of teaching. These findings support the growing view that clinical ethics involves skills that can and should be taught to medical students.
A portion of this research was presented at the Ambulatory Pediatric Association/Pediatric Academic Societies/American Academy of Pediatrics meeting in Boston, Massachusetts, May 12–16, 2000.
1. Rest J, Narveaz D (eds). Moral Development in the Professions: Psychology and Applied Ethics. Hilldale, NJ: Lawrence Erlbaum Associates, 1994.
2. Perkins H, Geppert C, Hazuda H. Challenges in teaching ethics in medical schools. Am J Med Sci
3. Stern D. Related values on call: a method for assessing the teaching of professionalism. Acad Med
. 1996;71(10 suppl):S37–S39.
4. Hundert EM, Hafferty F, Christakis D. Characteristics of the informal curriculum and trainees’ ethical choices. Acad Med
5. Hafferty FW, Franks R. The hidden curriculum, ethics teaching, and the structure of medical education. Acad Med
6. Dubose E, Hamel R, O'Connell L (eds). A Matter of Principles? Valley Forge, PA: Trinity Press International, 1994.
7. Jonsen A, Siegler M, Winslade W. Clinical Ethics: A Practical Approach to Ethical Decision Making in Clinical Medicine. 4th ed. New York: McGraw-Hill Health Professions Division, 1998.
8. Braunack-Mayer A, Gillam L, Vance E, et al. An ethics core curriculum for Australian medical schools. Med J Aust
9. Carter B, Roberts A, Martin R, Fincher R. A longitudinal ethics curriculum for medical students and generalist residents at the Medical College of Georgia. Acad Med
. 1999;74(10 suppl):S102–S103.
10. Epstein R, Hundert E. Defining and assessing professional competence. JAMA
11. Roberts A, Fincher R. Teaching third-year medical students how to handle ethical dilemmas. J Med Assoc Georgia
12. Diekema DS, Shugarman R. An ethics curriculum for the pediatric residency program. Confronting barriers to implementation. Arch Pediatr Adolesc Med
13. Ramsey PG, Coombs J, Hunt D, Marshall S, Wenrich M. From concept to culture: the WWAMI program at the University of Washington School of Medicine. Acad Med
14. Stapleton F, Pendegrass T. WWAMI: a regional approach to pediatric education. J Pediatr
15. Longitudinal Database of the University of Washington Department of Medical Education and Biomedical Information. Accessed 15 February 2002.
16. University of Washington School of Medicine Course Registration Database. Accessed 30 June 2003.
17. Smith S. University of Washington Pediatric Clerkship Ethics Curriculum 〈http://eduserv.hscer.washington.edu/pedPrograms/ethics.html
〉. Accessed 1 July 2003. University of Washington, 2003.
18. Self D, Olivarez M, Balwin D Jr. The amount of small-group case-study discussion needed to improved moral reasoning skills of medical students. Acad Med
19. Self D, Balwin DJr, Wolinsky F. Evaluation of teaching medical ethics by an assessment of moral reasoning. Med Educ
20. Singer P, Robb A, Norman G, Turnbull J. Performance-based assessment of clinical ethics using an objective structured clinical examination. Acad Med
21. Singer P, Robb A, Cohen R, Norman G, Turnbull J. Evaluation of a multicenter ethics objective structured clinical examination. J Gen Intern Med
22. Hebert P, Meslin E, Dunn E. Measuring the ethical sensitivity of medical students: a study at the University of Toronto. J Med Ethics