Preliminary Results on ASCO End-of-Life (EOL) Survey

[ Conrad Notes Index | ASCO Index ]

a timely medical meeting newsletter

PRELIMINARY RESULTS ON ASCO END-OF-LIFE (EOL) SURVEY

E.J. Emanuel, MD, PhD, Department of Clinical Bioethics, National Institutes of Health, Bethesda, Md, highlighted the 1997-98 mail survey on some 8,700 members of the American Society of Clinical Oncologists (ASCO). There were 149 multiple-choice questions and 4 patient vignettes on EOL issues of cancer patients. About 3000 (35%) of the ASCO members participated. The respondents were primarily urban, males, oncologists in academia or group practice, and extremely busy. Oncologists, according to the survey, need an improved comfort level in communicating with dying patients and in shifting to palliative care. ASCO faces the challenge of providing members with training to master the required skills.

Characteristics of surveyed members

Emanuel described the sample as:

Middle-aged (ave = 47y) and male (80%)
Primarily urban (>500,000 population)
Judeo-Christian (83%) with religion rated as important or fairly important (66%)
Extremely busy with 85% spending more than 50% of available time caring for oncology patients

This database includes information on highly experienced oncologists.

Learning about EOL care

According to the survey results, the top four ways that oncologists learn about EOL are:

Experience from patients through trial and error
Fellowship training
Colleagues
Residency role models

Least helpful in learning about EOL care were continuing medical education, ASCO and medical school courses, and palliative care rotation.

More recent medical school graduates received some EOL care training as a formal course or during clinical rotation. Nevertheless, fewer than 20% of the oncologists who graduated in the 1990s reported formal EOL training in medical school and only 50% received such information during postgraduate training.

Self evaluation of competency

This survey showed >90% of the responding oncologists perceive themselves as having "a lot" of expertise in managing pain, constipation, nausea/vomiting, fever, and neutropenia in EOL cancer patients. They report much less ability in treating depression especially and loss of appetite. In Emanuel's opinion, one barrier to learning new EOL skills may be the high sense of competency in pain management. (The unpublished Foley survey at NY Memorial Hospital found some 25% of EOL cancer patients still in pain).

Communicating with terminally ill patients

The Emanuel survey tabulated responses to such important questions as:

Percent	Item
>75	Competent in talking about EOL issues
10	Experienced a sense of failure when patients became terminally ill
20-25	EOL care among the worst parts of being an oncologist

Oncologists less likely to agree with the last finding were good communicators, female, practice includes more patients who die, less religious, urbanites, and never had a traumatic experience with a dying patient.

Barriers to caring for the dying

The data show that oncologists believe very important barriers to good EOL care are unrealistic patient/family expectations and family conflicts. Oncologists anxious about telling patients they will die reported having unrealistic expectations compared with good communicators. Only 10% of the US-surveyed oncologists found patient resistance to taking opioids or federal/state regulations of opioids as problematic.

Healthcare delivery represents a serious barrier in the US, according to questionnaire responses:

Percent	Item
20	No available pain service
40	Insufficient reimbursement for EOL care
50	Palliative care consultations not easily available

Other serious barriers include lack of time to discuss EOL concerns and limited availability of unskilled home care or hospice services.

Chemotherapy and EOL

Which oncologists prescribe chemotherapy for terminally patients? The present data point to practitioners >50 years old and persons who experience a sense of failure when the patient becomes terminally ill. Surgical oncologists seemed most likely to prescribe chemotherapy rather than palliative care. Only 20% of the respondents provide hospice care after second-line chemotherapy.

Vignette: patient with unremitting pain

One of 4 vignettes covered a breast cancer patient in pain although on iv morphine. What should be done? Here are the survey responses:

Percent	Procedure
75	Increase the drip
10	Add morphine antagonist with increased drip
10	Use nonopioid analgesic
3	Consult with family

Oncologists less likely to increase the morphine dose included: >50 years of age, surgical or radiation oncology practice, opposition to euthanasia, or not a follower of a major US religion.

Euthanasia and physician-assisted suicide (PAS)

Twenty-two percent of responding oncologists found PAS ethically acceptable and 15% would be willing to do it. Corresponding figures for euthanasia were lower, 6.5% and 2.5%, respectively. Others reporting more support for PAS were less religious or surgeons. Oncologists who have more time to talk to the patient, Catholic, or possess poor pain management skills found PAS and euthanasia less acceptable.

Presenter conclusions

Emanuel's conclusions are based on a preliminary analysis of the data which will be refined for publication. The survey results show a need to improve the comfort level of oncologists in communicating with dying patients, especially in shifting to palliative care. ASCO should provide members with training to master the required skills identified by this survey. One benefit could be a decreased use of third- and fourth-line chemotherapy in the terminally ill.

For professional correspondence, please contact Dr. Emanuel by Fax at: (301) 496-0760 or by E-mail at: eemanuel@nih.gov

Eugene A. Conrad

Presented at the American Society of Clinical Oncology (ASCO) Meeting, May 16-19, 1998
Copyright © 1998 Conrad Group Inc. All Rights Reserved
Eugene A. Conrad, PhD, MPH / ISSN 1078-2230 / June 1998
Send Comments to: ConradNote@aol.com

[ Top | Conrad Notes Index | ASCO Index ]