We conclude that a 4-day, 32-hour curriculum in end-of-life care leads to significant improvements in knowledge, skills, and attitudes that are sustained. Baseline assessments were stable across rotations and academic years, suggesting that the effects are not due to other changes in the medical school curriculum or in the larger social context. In addition, this also means students do not learn this material elsewhere in the clinical curriculum of the third year or the fourth year.
We chose the self-reported measurement of confidence to perform various skills because it had been used for the large comparative group of 10,000 internal medicine residents and faculty. In that setting, the choice is obvious because of the size of the group. Our need of a comparison group, and the size of our intervention, also favored the use of self-report. In further research, more focused evaluation of skills in a representative subset of students would be feasible.
Some who look at this data might be discouraged by the size of the absolute differences. Therefore, the statistical test of Effect Size is designed for situations like this. The Effect Size varies from 0–-1 where an effect less than 0.3 is small, 0,4–0.6 is moderate, and 0.7 to 1 is large. In the national sample, the effect size for change was 0.18. In contrast, the effect size for this intervention is 0.56—a moderately large effect.
This illustrates several important points about the evaluation instrument. First, the evaluation instruments were designed to cover all significant domains of palliative care—they were not designed to measure the achievement of specific learning objectives from a specific course. Consequently the instruments can be used across a variety of curricula, and an assessment of gain in the broad domain of palliative care can be discerned. For example, in our experience, only highly experienced faculty in the specialty of hospice and palliative medicine score 100%. Fellows studying in hospice and palliative medicine begin at the same level as medical students and rarely get out of the 70%–80% range despite an entire year of training. Therefore, the analogy to the thermometer is apt—a small change on the thermometer (from 37°C to 38.5°C on a 1–100 scale is tiny, but it is highly significant. The same is true for the instruments used in this study.
This curriculum is similar to that reported by the University of Maryland School of Medicine where they tested a required rotation in hospice and palliative medicine in the junior year. This module was received very positively by students and was ultimately made a mandatory part of the curriculum.31
At the University of Rochester,32
the introduction of a major curricular reform curriculum integrating basic science and clinical training over 4 years of medical school, provided an opportunity to develop and implement a fully integrated, comprehensive palliative care curriculum. Dr. David Weissman has developed a comprehensive program of hospice and palliative medicine education at the Medical College of Wisconsin over the past 20 years, which includes a required course for second- and third-year medical students and clinical electives for fourth-year medical students on the palliative medicine consultation service in the University Hospital and with affiliated hospice programs.18
The importance of clinical training in end-of-life care is reflected in the 2006 decision of the American Board of Medical Specialties (ABMS) to approve hospice and palliative medicine as a subspecialty. A unique and precedent setting event for ABMS is that 10 members of the ABMS agreed to implement certification in hospice and palliative medicine as a cooperative effort among 10 cosponsoring boards, representing anesthesiology, emergency medicine, family medicine, internal medicine, obstetrics and gynecology, pediatrics, physical medicine and rehabilitation, psychiatry and neurology, radiology, and surgery. The scope of the sponsoring Boards speaks strongly to the recognition that end-of-life care is highly valued across medical specialties.33–34
This study drew on several principles of best practices. For students to acquire the necessary attitudes, knowledge and skills of hospice and palliative medicine, such education should be longitudinal, a mixture of didactic and experiential learning opportunities, contain opportunities for self reflection, provide opportunities to practice the skills they are learning, and be interdisciplinary.
We postulated that students learn best when they are exposed to the direct care of patients who are being treated with the knowledge, skills, and attitudes the student needs to develop. When family members of patients who died are asked about quality of end-of-life care, hospice programs perform better than hospitals, nursing homes, and home care (without hospice care).6,35
Thus, we chose to imbed training in end-of-life care in a hospice setting within a required core internal medicine rotation. our results demonstrate that this approach successfully increases core knowledge and skills and decreases the level of concerns of learners who deal with the challenging issues surrounding death. It also demonstrates that a modest amount of instruction in the third year raises students' levels of knowledge to that of U.S. faculty.
Our approach to educational reform reflects the understanding that curricular change requires “buy-in” from educational leaders as well as provision of resources.28,36–43
When deans and faculty recognize the value of instruction, finding time in the curriculum becomes easier.
Limitations of our study include the inclusion of a single medical school and the lack of random assignment of trainees to the educational intervention. To address such threats to internal validity frequently confronting medical education research, we incorporated design elements to mitigate these limitations.44
In our study, this included the use of benchmark data from a national study of residents and faculty, providing us with an empirical context from which to interpret the effect of our curricular training. In addition, we drew on the results of the Association of American Medical College's Graduation Questionnaire, to place our study's findings in the context of medical students' perceptions of end-of-life care education in other medical schools.
Another potential limitation is reflected in the extent of palliative care resources present in the study institution, for we recognize that the number of full time board-certified subspecialist palliative medicine physicians and subspecialty fellows and a dedicated hospice-based center for education and research are not broadly available in the United States. However, viewed another way, this is a strength. The study results were achieved with more than 40 different physician faculty suggesting that the results are not dependent on a single charismatic physician faculty member. Consequently, this is germane to the many hospice programs that host medical students as part of clinical clerkships.
The development of hospital-based palliative care teams can be seen as an effort to try to bring the skills developed in hospice programs into hospitals where they can be applied more broadly. Efforts to demonstrate patient-centered outcomes of such innovations are underway. As a way to ensure medical students are exposed to appropriate clinical care as part of a hospice and palliative medicine education curriculum, collaboration with a hospice program or palliative care team can be an important element.
Although developed with many physicians, our curriculum does not require hospice-based physicians to teach it. This offers encouraging evidence that the curriculum could be adopted effectively by other schools. Dedicated inpatient consultation services and units are rapidly multiplying in the United States. Clinical medical student training can effectively occur in this environment. These factors suggest that the curriculum and its results are “portable,” i.e., they could be extended to other training settings and populations.
For this curriculum a 50% time coordinator assured the students knew where to come and assembled the course materials for them. The syllabus was printed each year. Since the time of this study, it is now given to them on a “memory stick” The medical school covered the cost of developing the standardized patient for breaking bad news. The 16 hours of physician classroom time is required, which is the most expensive aspect of the course.