We used a controlled, pre-post study design to evaluate the effect of a curriculum for teaching medical ethics to orthopaedic residents. A survey was administered to the residents in two training programs (the baseline survey), the educational intervention was carried out at one of the two programs, and the follow-up survey was then administered at both programs.
Educational Intervention
A case-based approach was used for the intervention5. This approach is familiar to resident physicians as the method of learning clinical medicine, and it permits inclusion of their own cases and experiences. Cases for the curriculum were based on a series of ethical dilemmas that arose on the orthopaedic surgery services and other medical and surgical services at several institutions. The goals of the educational program were to promote increased and earlier recognition of ethical issues, to improve knowledge and understanding of ethical codes and principles, and to enhance the capability to resolve ethical problems arising in patient care. The course was taught by a physician who had experience in the teaching and clinical practice of medical ethics.
The topics that were covered in the one-year medical-ethics curriculum were based on the experience of orthopaedic surgeons, topics covered in standardized orthopaedic examinations, and previous curricula developed for other training programs2,7,9,13. These topics included informed consent, confidentiality and privacy, truth-telling, end-of-life decision-making (including the use of do-not-resuscitate orders), the physician-patient relationship, the impaired physician, and ethical issues pertaining to managed care (including justice in the allocation of health-care resources). The curriculum was presented as part of the regular orthopaedic didactic program. Each monthly forty-five-minute presentation covered one topic and included a case presentation followed by a discussion of principles applicable to it. Pertinent literature, codes, and policy statements were presented, and related cases were analyzed. Practical clinical approaches, based on the principles, were emphasized. As in clinical teaching, resources for ethical knowledge, the availability of literature focusing on clinical ethics, and situations in which ethical consultation should be sought were presented.
Study Samples and Design
The samples for this study were the full residency classes at two orthopaedic surgery training programs. The medical ethics curriculum was implemented at one of these programs (the intervention site), and no formal education in ethics was provided at the other (the control site). There were twenty-five residents at the intervention site and thirty at the control site. At the beginning of the academic year, in August 1995 (before the beginning of the ethics curriculum), and at the end of the academic year, in June 1996 (after the completion of the curriculum), the residents at both sites were asked to complete the same ethics questionnaire.
The survey instrument focused on the residents' knowledge of clinical ethics in six areas of medical practice: confidentiality, informed consent, truth-telling, the physician-patient relationship, end-of-life decision-making, and incompetent colleagues. The twenty-two-item survey included thirteen brief clinical scenarios and nine short-answer questions. All of the questions had one ethically appropriate response (or set of responses), and the answers were multiple choice or true or false. The details of the survey were reported previously30. The survey used in the current study did not include four items from the original survey because these items were answered correctly at baseline by nearly all of the residents. Also, the questions focusing on economic aspects of care in the original survey were combined with those pertaining to the physician-patient relationship in the survey used in the current study. The twenty-two items that were evaluated in the current study were identical in the baseline and follow-up surveys. In addition, in the baseline survey, respondents were asked their age, gender, and year of graduation from medical school as well as whether they had received any training in medical ethics. In the follow-up survey, they were asked whether they had received any instruction in ethics during the previous year and, if so, whether it had been very, somewhat, slightly, or not at all clinically useful.
The questionnaires were delivered by hand or were mailed to all residents. The survey was self-administered. A second survey was mailed to non-responders. Data from the survey were entered into a database after all identifying information had been removed so that the responses could not be linked with the respondents. Only respondents who had completed both the baseline and the follow-up survey were included in the current analysis.
The survey scores were calculated as the number of correct responses divided by twenty-two (the number of items in the survey), yielding a composite ethics score for each individual. The mean scores for the residents at the intervention and control sites on the baseline and follow-up surveys also were calculated. Pairwise comparisons were made between each respondent's scores on the baseline and follow-up surveys. The mean difference between these scores at the intervention site was compared with that at the control site with use of a t test. The mean scores on the baseline and follow-up surveys at each site also were calculated for each of the six subindices, and the differences between the mean scores at the intervention site were compared with those at the control site with use of t tests. Finally, a linear regression analysis of the difference between the composite scores on the follow-up and baseline surveys was carried out with the site of the study, the age and gender of the resident, the time since graduation from medical school, and whether the resident had received any previous training in medical ethics as independent variables.
Twenty-five (96 per cent) of the twenty-six residents at the intervention site and thirty (83 per cent) of the thirty-six residents at the control site completed both the baseline and the follow-up survey. Forty-eight (87 per cent) of the fifty-five subjects were men. The mean age of the fifty-five subjects was twenty-nine years (range, twenty-six to thirty-five years). Forty-four residents (80 per cent) stated, on the baseline survey, that they had received some type of formal instruction in medical ethics. There were no significant differences (p > 0.2) in these characteristics between the residents at the two sites (Table I).
There was no difference in the baseline scores between the sites: the residents at the intervention site had a mean score of 0.71 (that is, they answered 71 per cent of the questions correctly), and those at the control site had a mean score of 0.72. The respondents had better scores (combined mean scores of more than 0.8) on the questions concerning confidentiality, truth-telling, end-of-life decision-making, and incompetent colleagues than on those involving informed consent or the physician-patient relationship (combined mean score, 0.61 and 0.65, respectively).
The residents at the intervention site had a mean score of 0.81 on the follow-up survey, and the residents at the control site had a mean score of 0.74—an improvement of 0.10 and 0.02, respectively, compared with the baseline scores. The improvement at the intervention site was significantly greater than that at the control site (p = 0.002). There was a trend toward greater improvement in the scores for each of the six subindices at the intervention site than at the control site; however, this difference was significant only with regard to informed consent (mean improvement in score, 0.15 compared with 0.02; p = 0.02) (Table II).
Several survey items and the responses to them are presented in order to demonstrate the improvement in the residents' knowledge at the intervention site compared with that at the control site. The question that follows concerns the residents' approach to a potential conflict of interest.
The appropriate response is B. The physician must reveal that he or she will receive a royalty on the prosthesis because that fact would not be readily known by the patient and the revelation is necessary to avoid both a real and an apparent conflict of interest. From the time of the baseline survey to the time of the follow-up survey, the number of respondents who answered this question correctly increased from seventeen (68 per cent) to twenty-four (96 per cent) at the intervention site, whereas at the control site it increased from twenty-two (73 per cent) to twenty-four (80 per cent).
One series of items focused on aspects of informed consent. In the scenario that follows, subjects were asked how to approach a patient who waives the right to be fully informed about the risks and benefits of a procedure.
The correct answer is C. Some patients may choose not to be informed about selected (or even all) aspects of a procedure when providing consent. It is essential to be sure that the patient has decision-making capacity, just as it is necessary when a patient refuses a procedure that the physician believes is needed. In nearly all circumstances, it is ethically acceptable to allow a patient who has decision-making capacity to waive the right to information before he or she provides consent. In the situation just described, the medical record must document that the patient had the capacity to make decisions and that attempts were made to inform him of the potential risks and benefits; the information that the patient chose not to hear before consenting to the procedure must also be documented. The patient should not sign a consent form indicating that all elements of informed consent have been completed, although some consent forms include waiver of information as an option to which the patient can attest. The differences in the responses on the baseline survey between the residents at the intervention and control sites persisted at the time of the follow-up survey. On the latter questionnaire, one-third of the residents at the control site chose a response indicating that they did not understand the patient's right to waive information when providing consent.
As shown by the responses to the following item, a substantial minority of the residents at the control site also did not understand which factors should be considered when decisions are made for a patient who lacks decision-making capacity.
The correct answer is C. Focusing on what the incapacitated patient would want facilitates decision-making and is ethically grounded in the principle of autonomy. When the physician guides discussions with family members and other surrogate decision-makers in this direction, it is possible to avoid treatments aimed at meeting the surrogate's (rather than the patient's) goals18.
Another scenario reveals that residents at the intervention site gained knowledge about appropriate care for patients who are near the end of life.
Answers A, B, D, and E are all appropriate responses. It is ethically acceptable, with regard to this patient, to write a do-not-resuscitate order; to discontinue all treatment, including the ventilator and total parenteral nutrition; and to move the patient out of the intensive-care unit. On the follow-up survey, all respondents recognized that the use of narcotics with the intent of suppressing respiration is not acceptable.
A final example focuses on the subjects' understanding of the purpose and use of advance directives. Residents at the intervention site had a greater improvement in their knowledge pertaining to this area than did those at the control site.
A and C are accurate answers concerning advance directives. These documents are useful for stimulating discussion about end-of-life issues between physicians and patients10, and a proxy may be designated to participate in medical decisions after a patient has lost decision-making capacity. These documents do not require a lawyer's assistance or convey financial powers, and they may be retracted verbally at any time.
All respondents at the intervention site indicated, on the follow-up survey, that they had received instruction in medical ethics during the academic year of the study, whereas only one respondent in the control group responded affirmatively. Among the respondents at the intervention site, six (24 per cent) rated the medical ethics curriculum as very useful; seventeen (68 per cent), as somewhat useful; and one (4 per cent) each, as slightly useful or not at all useful.
When the linear regression analysis was performed with the site of the study, the age and gender of the respondent, the time since graduation from medical school, and whether the respondent had received previous instruction in ethics as independent variables (adjusted R2 = 0.14), only the site of the study was found to be a significant independent predictor of improvement in the composite ethics score between the baseline and follow-up surveys (p = 0.002).
Medical ethics has been taught in some medical schools and residency training programs for decades. Many curricula have been developed, but few have been evaluated objectively. The requirement to include medical ethics in orthopaedic training is recent, and ethics curricula designed for orthopaedic surgeons are not widely available. To our knowledge, we describe the first evaluation of a medical ethics curriculum in an orthopaedic surgery residency program. This one-year curriculum improved residents' knowledge of medical ethics and their ability to handle hypothetical situations.
The ethics curriculum that was used in this study covered the topics included in most previously published curricula19. However, in contrast to the focus on end-of-life care in previous curricula7, the curriculum in the present study was intended to address the dilemmas most often faced by orthopaedic surgeons. These difficult issues include informed consent, intricacies of the physician-patient relationship (such as interruption of the relationship without abandonment of the patient), conflicts of interest (such as receipt of royalties for prostheses), and gatekeeper-specialist relationships within the context of capitated care. The case-based structure that was used in this study was previously shown to increase the amount of ethics-based discussion of clinical cases2,13. All of the scenarios that were developed for the educational intervention were constructed around orthopaedic themes. Whenever possible, articles, codes of ethics, and case examples were culled from the orthopaedic surgery literature to enhance the relevance to the residents.
Several investigators evaluating the effect of medical ethics education have shown that both lectures and discussions of cases improved the moral reasoning scores of medical students14,22-24. A study with a pre-post test design showed that education regarding informed consent improved medical students' knowledge of that topic15. Sulmasy et al., in controlled trials, found that lectures and discussions on ethics increased medical residents' knowledge and confidence in addressing ethical issues28,29. The ethical sensitivity of psychiatry residents was found to have increased after formal education11. A case-based educational program in ethics for surgical residents, focusing on end-of-life issues, led to an increase in documented discussions about care and a decreased duration of hospitalization among dying patients in the intensive-care unit13.
In the current study, a curriculum oriented toward orthopaedic surgery residents succeeded in improving scores reflecting knowledge about ethics and responses to scenarios involving ethical problems. Compared with the residents at the control site, the residents at the intervention site had a greater baseline-to-follow-up improvement in the score for each of the subindices, although this difference was significant only for the subindex of informed consent. Overall, the increase in correct responses averaged about two of the twenty-two questions at the intervention site, with little change at the control site.
It should be recognized that our findings represent only changes in knowledge; it is not known whether correct responses on the survey will translate into ethical behavior16. According to one model of moral behavior, an individual proceeds through four steps before autonomously acting in a moral manner. These steps are moral sensitivity, judgment of whether an action is moral, motivation, and development of a moral character21. In focusing on knowledge about ethics and then asking about intended behavior, the current survey primarily addressed respondents' moral sensitivity and their judgment of whether specific actions were moral. One study of orthopaedic surgeons suggested a relationship between moral reasoning and clinical behavior3. However, additional investigation is needed to determine whether an educational program will produce more ethical orthopaedic surgeons. Structured clinical examinations have been developed for this purpose, but they are not reliable for individual application26,27. Peer review as well as evaluation, performed by nurses and patients, of the humanism and ethical behavior of physicians are other potential methods of behavioral evaluation that would be more advanced than the methods used in this study17.
The current study also was limited in scope. The medical ethics curriculum was presented at only one site; therefore, additional evaluation is necessary to determine whether the curriculum will lead to increased knowledge of ethics at other sites. Many of the cases used in this curriculum have been produced on a videotape with an accompanying discussion guide so that the course can be tested at other venues. In addition, the follow-up survey was administered at the conclusion of the course; the question of whether the gains that were identified will be sustained merits investigation. We studied only orthopaedic surgery residents. It is not known whether attending orthopaedic surgeons would accept or benefit from the ethics curriculum, although their deficits in knowledge of ethics mirror those of the residents30. Furthermore, other methods of teaching medical ethics, such as bedside instruction25, should be considered4,6.
In conclusion, compared with a control group, residents who received the educational intervention had improvement in their knowledge of ethics, particularly in the areas of informed consent and the physician-patient relationship. This curriculum might be adapted to other samples, such as attending orthopaedic surgeons.