The CAD is a brief self-report questionnaire to assess general concerns about death in health care providers .12 The CAD measures three factors through 10 statements: general concern about death, spirituality and patient-related concern about death. Each item is scored using a Likert scale format with five anchored rating choices from disagree completely to agree completely. In a validation study with third year medical students, test-retest reliability of the three scores was good (r = 0.84, 0.89, 0.83, respectively), internal consistency was adequate (alpha = 0.73, 0.76, 0.85, respectively), and correlations with related measures supported its construct. 15

PEAS, a self reported instrument, measures the willingness of medical students /residents to interact with dying patients and their families, and was developed as an outcome measure for palliative care education. PEAS developers now recommend use of a 35 Likert-type item form version of PEAS that are matched into two subscales (Personal Communication, Hayslip, 2005). These subscales assess unique communication concerns of physician trainees in their interactions with dying patients. The personal preparation subscale with 16 items and the professional role subscale with 19 items are scored to indicate higher communication concerns with higher scores. The personal preparation subscale assesses personal feelings about interacting with dying persons and families including personal touch, conversation about end-of-life topics, feelings and comfort when speaking or being with dying patients or their families. The professional role subscale assesses the trainee’s difficulties in disclosing a poor prognosis to a patient; communicating with the patient about the projected future course of his or her illness; sharing or withholding information from the patient; discussing advanced directives with the patient; emotional boundaries in caring for dying patients; breaking bad news; and feeling inadequate as a professional when a patient dies. Excellent internal consistency and reliability are reported on the two subscales; validity studies with CAD subscales are also reported. 13

This survey measures students/residents/fellows’ self-assessed competencies and concerns about end-of-life care. 14 We selected and modified three of the four domains in the survey: competence associated with six end-of-life communication skills, 10 end-of-life medical care topics and topics to be included in future end-of-life care education.

Two tools assessed Death Rounds: a post DRQ and a Facilitators’ Log of observed student learning issues. The questionnaire was developed based on findings reported by Rhodes-Kropf et al, 2005 about medical students’ reactions to their most memorable deaths. 16 Specifically, the students choose adjectives that best describe their most recent death, with whom they discussed these deaths, their level of comfort with death, the value of Death Rounds, learning issues they encountered and how they addressed them, and their assessment of their palliative care skills before and after participation in Death Rounds. The facilitators for Death Rounds recorded the student identified learning issues in a compiled log.

Students evaluate the quantity and quality of instruction and training they receive in 15 topics identified as cross cutting topics included in the UNMSOM integrated curriculum during Phase II and Phase III (Years 3 and 4). Students rate the quantity and quality of instruction for each topic on a 1-5 scale. Topics associated with palliative medicine include: end of life care, management of pain, communication skills and ethics and professionalism.

The AAMC GQ is completed by graduating students at Liaison Committee in Medical Education approved medical schools to assess the overall curricular experience. An annual report is prepared for each school aggregating the responses from its school as well as responses in all schools. Questions included in the 2006 and 2007 AAMC GQ address palliative medicine/pain management and end –of- life care. There are two different question types used. In one format, students are asked “Do you believe time devoted to your instruction in the following areas was inadequate, appropriate, or excessive?” In a second format, students are asked to “Indicate your level of agreement with the following statements” using a Likert scale format with 5 anchors from strongly agree to strongly Disagree. The questions under Communication Skills begin with the following lead: “I am confident that I have the appropriate knowledge and skills to” “discuss treatment options with a patient with terminal illness” (2007 only) and “initiate discussion of DNR orders with a patient or family member.” An individual range, mean and respondent count data are reported to the individual school for the past three years and the current year.

The questionnaire was given to students to complete at the end of Year 2 (Classes of 2007-2009), 185 students responded, and at the end of Year 3 (Classes of 2006-2008), 191 students responded, which is a 75-85% return rate from each administration.

Overall demographics and response rate showed fifteen percent of survey respondents did not provide information on their age, gender and ethnicity. Of those who responded to the demographic questions, 64% were female, 50% ranged in age from 21 to 26, 30% ranged in age from 27 to 31, 63% identified as Anglo, 24% identified as Hispanic , 3% identified as Native American , and 0.5% identified as African American . Eighty to ninety percent of each class responded to the ten items on the CAD survey at each administration. More than 75% of the students were able to agree or disagree with each statement rather than choosing “neutral” for a response. Subscale scores spanned the entire range of possible scores from 1 to 5. In Results are shown in Table 2. Students’ general and spiritual concerns about death did not change over the year they participated in the palliative medicine curriculum. However, their concerns about working with dying patients decreased.

Sixty-three students from the Class of 2006, and sixty-two students from the Class of 2007 responded to the PEAS questionnaire (80-85% ) during their third year. We did not include analysis from respondents in the first two years of medical school because they used the “not applicable” (NA) option for several items and their total and subscale scores gave incomplete information. As seen in Table 3, average scores on questions in both the Professional Role and Personal Preparation subscales suggest that at the end of the third-year clerkships, students’ communication apprehension is low, with no significant difference reported between the two classes. Third year UNMSOM students in Class of 2006 and 2007 participated in the fully integrated palliative medicine curriculum.

Sixty-seven of 132 (51%) graduating medical students in the Classes of 2005 – 2006 filled out the survey. Forty-one students (53%) in the Class of 2007 responded to a modification of the survey. A significantly lower percentage of students in the Class of 2005 (42%) rated themselves competent to perform, independently or with minimal supervision, the 6 end-of-life communication skills listed in Table 4 than did students in the combined Classes of 2006 and 2007 (67%). A significantly lower percentage of students in the Class of 2005 (33%) also rated themselves competent to perform, independently or with minimal supervision, care related to the 10 end-of-life topics listed in Table 4 than did students in Classes of 2006 and 2007 combined (60%). In addition, in Table 5, over 85% of students, feel the need to learn more about 10 topics in future educational programs.

Ninety percent (90%) of the students experienced a recent patient death during a clerkship. They selected diverse adjectives to describe the impact of the experience in personal terms. Our findings about how students describe their encounters with a patient’s death are consistent with previous results reinforcing that students have strong emotional responses to patient deaths. About one quarter of our students spoke about the death with an attending physician. A majority of students found Death Rounds to be an educational model that increased their comfort level about death and management of the dying process. 17 One student who had not experienced a patient death commented, “ I think it’s [Death Rounds] a good idea. I only wish I had known patients that died so I could deal with these important issues. I had several “bad news” scenarios, but no deaths in any patient that I met.”

Evidence in both the DRQ and Facilitators’ Logs demonstrate that multiple topics were addressed and were associated with the School of Medicine’s six core competencies. Students’ self ratings of knowledge, skills and competence after Death Rounds supported this evidence. 17

Ninety to ninety-five percent of the students in Phase II and Phase III given the CCS in the Classes of 2006- 2008 completed it. As seen in Table 6, all mean scores for both quantity and quality of training on Cross Cutting issues associated with palliative care were at or above a mean of 3.0 on a five point scale except for Management of Pain.

Ninety-three percent of the students in Class of 2006 and 96 percent of the students in the Class of 2007 from UNMSOM completed the AAMC GQ.

In response to the question, “Do you believe time devoted to your instruction in the following areas was inadequate, appropriate, or excessive?”, 65.6% in 2006 and 82.9%

in 2007 of the UNMSOM completing the questionnaire rated instruction in palliative care/ pain management as appropriate. Included in this report is the percentage from all

respondents across AAMC sites. In the All School 2007 rating for this item, 75.8% of students reported receiving appropriate instruction in palliative care/pain management. In response to a similar question about time devoted to instruction in End of life care, 63.9% in 2006 and 76.3% in 2007 of the UNMSOM students rated instruction as adequate, and 78% of 2007 students from All Schools rated instruction as appropriate.

Eighty-eight percent of 2007 UNMSOM students and an identical percent of 2007 students from All Schools agree or strongly agree with the statement, “I am confident that I have the appropriate knowledge and skills to discuss treatment options with a patient with terminal illness”. Eighty-seven percent of 2006 UNMSOM students and 88 percent of 2007 UNMSOM students agree or strongly agree with the statement, “I am confident that I have the appropriate knowledge and skills to initiate discussion of DNR orders with a patient or family member.” Eighty-five percent of 2007 students from All School agree or strongly agree with this statement as well. These findings echo the significant improvements documented throughout the United States from 1998-2006. (18)

Figure 1

Table 1: Administration Schedule of Assessment Tools

Figure 2

Table 2: Concerns About Dying Survey Summary (CAD)

Each item was rated on a 1-5 scale where a rating of 4 or 5 indicates concerns about death or discomfort working with patients who are dying. Subscale scores are averages of the ratings of all items in the subscale.

Figure 3

Table 3: Physicians’ End-of-Life Care Attitude Scale (PEAS): Summary of Third- Year Medical Student Responses

Figure 4

Table 4: Competencies and Concerns in End-of-Life Care (CCEOL)

Figure 5

Table 5: Percentage of Respondents Requesting Inclusion of Topic in Future Education Programs CCEOLC Survey

Figure 6

Table 6: UNMSOM Cross Cutting Themes Survey Classes of 2006-2008