JOURNAL OF PALLIATIVE MEDICINE Volume 10, Number 5, 2007 © Mary Ann Liebert, Inc. DOI: 10.1089/jpm.2006.0264
Do-Not-Resuscitate Orders for Terminal Patients with Cancer in Teaching Hospitals of Korea DO YEUN KIM, M.D.,1 KYOUNG EUN LEE, M.D.,2 EUN MI NAM, M.D.,2 HYE RAN LEE, M.D.,3 KEUN-WOOK LEE, M.D.,4 JEE HYUN KIM, M.D.,4 JONG SEOK LEE, M.D.,4 and SOON NAM LEE, M.D.2
ABSTRACT Objectives: To examine the current practices relating to do-not-resuscitate (DNR) orders for terminal patients with cancer at teaching hospitals in Korea. Methods: The records of 387 deaths from January 1 to December 31, 2005 at four cancer centers were identified and reviewed to assess the DNR delineation. Basic demographics, circumstances surrounding the establishment of the DNR directive, the percentage of orders for identified populations, and the time interval between DNR consent and death were evaluated. Results: An order of DNR consent was obtained from 296 patients (76%) of a total of 387 patients. All DNR consents were made between the physician and family, without involving the patient. Written preprinted DNR consent forms were used in 169 (57%) cases and 127 patients (43%) had verbal DNR permission. DNR consent was interpreted in two ways: one forbade resuscitation in two hospitals and the other implied limited care in two other hospitals. A unilateral physician decision to withhold cardiopulmonary resuscitation (CPR) was decided for 62 (16%) patients. Terminal CPR was performed on 29 (7%) patients. DNR discussion was made within 7 days of the day of death on 228 (77%) patient among the 296 DNR consenting patients. Conclusion: From our teaching-hospital–based analysis of terminal cancer patients in Korea, consent for a DNR order was common. However, DNR order forms were not standardized and lacked room to document patient involvement in the decision. Usually the DNR decision was made within last days of the patient’s life. Our results reflect the need for the improvement of end-of-life care decisions in Korea. INTRODUCTION
T
HE DO-NOT-RESUSCITATE (DNR) order is an example of end-of-life care needs for a patient with terminal cancer’s best interest. One of the goals of completing DNR orders is to encourage the physician to consult with the patient or the family to determine their wishes concerning further treatment, so as to allow patients the pursuit of death with dignity. Therefore, it
is necessary that a DNR directive be clarified to the patient or the surrogate decision-maker. DNR orders are frequently used for terminal patients with cancer. However, many studies have shown that DNR orders have wide cultural differences in their implementation.1,2 Who decides about the order, how it is noted, and what it really means in practice are often questions to which there are no clear answers.
1Department
of Medical Oncology, Dongguk University International Hospitals, Goyang, Gyeonggi, Republic of Korea. of Medical Oncology, Department of Internal Medicine, School of Medicine Ewha Womans University, Seoul, Republic of Korea. 3Department of Medical Oncology, Inje University Ilsanpaik Hospital, Goyang, Gyeonggi, Republic of Korea. 4Department of Medical Oncology, Seoul National University Hospital, Seongnam, Gyeonggi, Republic of Korea. 2Section
1153
1154
KIM ET AL.
In Korea, cultural factors make issues relating to the DNR dilemma more confusing. First, not informing a patient with cancer of the true diagnosis is common in our society, even though patients have the right to be informed about their illness and its treatment. These rights, for example, include being informed about their prospects for recovery, the available treatments, and the current extent of the disease. However, many patients with cancer in Korea do not receive accurate information. It is estimated that only 14.4 % of South Korean family members permit physicians’ access to the patient when end-of-life decisions are being discussed.3 However, because the patient’s proxies underestimated or overestimated the patient’s preference regarding resuscitation in 44% of cases,4 patients themselves must take part in deciding on their treatment. If the patient with terminal cancer has insight about his/her illness, he/she may want to have the right to decide to their own treatment. In the United States, hospitals have been required to inform patients of their right to refuse medical treatment. In contrast to the United States, most Korean patients with cancer do not have this right; therefore they might receive aggressive intervention that they do not desire. Second, the DNR is not a legal medical practice in Korea. Legislation has governed the use of DNR decision in the United States since the 1970s and later on in Canada and Australia.5 In Korea, DNR practices exist almost exclusively at the institutional level. Therefore, no resources are available regarding about DNR policy or guidelines for systemic standardization. Because DNR practices are not guided by health practitioners’ consensus or protected by law, it is not surprising to observe that Korean physicians frequently misunderstand the scope of the level of care of DNR orders. In Korea, there is little information on how a DNR order works or whether a DNR order meets its goals for terminal patients with cancer under the above-described cultural background. The aim of this paper was to describe the results of a study of 387 deaths of terminal patients with cancer at four teaching hospitals to reveal current practices relating to DNR. For each patient, we collected data on whether they had a DNR status, who was involved in the discussion about their DNR status, how the DNR consent was noted, what DNR consent really meant in practice, and when DNR was discussed prior to death.
MATERIALS AND METHODS Hospitals Four major teaching hospitals in urban areas of Korea were invited and agreed to participate in a volun-
tary survey study. The mean hospital bed size was 675 (350 in Ewha Womans University Dongdaemun Hospital, 650 in Inje University IlsanPaik Hospital, 800 in Ewha Womans University Mokdong Hospital, 900 in Seoul National University Bundang Hospital), which consisted of a mean of 55 (30, 50, 60, and 80, respectively) oncology beds.
Patients A total of 387 consecutive terminal patients with cancer at four major teaching hospitals were enrolled. A mean of 62% (range, 51%–75%) of patient deaths resulting from cancer that occurred in each of the four hospitals between January 1 and December 31, 2005 were retrospectively reviewed. Patients were required to meet these criteria: admission of at least 1-week duration; patient age above 18 years; and the primary attending staff was a medical oncologist. The data extracted consisted of patient demographics, the primary cancer site, and the circumstances surrounding the establishment of resuscitation status, with special attention paid to the DNR delineation. Consent for a DNR order is defined as present if the patient has a preprinted DNR form or recorded verbal communication on a medical note. A unilateral decision was recognized if DNR was performed without consent for a DNR order.
Data The project staff provided a data collection form and instructions for completing a medical chart review. A medical oncologist at the participating hospitals first conducted the medical chart reviews, and then project staff again reviewed the chart, with discrepancies resolved by consensus between the researchers. Data were entered into a database (SPSS 13.0, SPSS Inc., Chicago, IL) and descriptive statistics were used to summarize the data. Ethics approval was obtained from all of the Institutional Ethics Review Boards.
RESULTS Patients’ characteristics Selected patient characteristics are shown in Table 1. The mean age was 61 years (range, 20–90). Patients had a broad range of primary diagnoses. For most patients, there was evidence in the records prior to death that they were considered to be dying by the clinicians responsible for their care.
1155
DNR PRACTICES FOR TERMINAL PATIENTS WITH CANCER IN KOREA TABLE 2.
DNR order DNR consent was noted on the charts of 296 patients at the four hospitals (Table 2). Preprinted written documents were present for 169 patients. Verbal consent on medical records was given to 127 patients. Even though the majority of patients were able to communicate and participated, all physicians discussed the DNR decision with family members. Generally, the primary attending physicians who were in charge during the patients’ death discussed the DNR with the patients’ surrogate (92%). Preprinted DNR consent forms were available at three hospitals. On all consent forms from these three hospitals, the time, date and nature of the discussion with the patient’s family were recorded. These forms provided a coding space for the formal confirmation of a DNR decision by the treating physician, which included patient and/or family member’s signatures. We did not find the following details on the DNR order sheet: patient diagnosis, an assessment that attempts at resuscitation would not benefit in addition to notes on consultation with others before a DNR decision was made. Institutional DNR order forms were examined. One hospital used the fixed DNR decision consent form and the other two hospitals possibly used no treatment decisions that could be regarded as a limited care. Preprinted categories to order no treatment decisions (antibiotics, hemodialysis, and intensive care unit transfer) were present on the limited care form DNR consent form. There was no place to enter a request to withhold or withdraw artificial administration of food and fluids. Verbal DNR consent was obtained in one other hospital. All verbal DNR consent was recorded as simply writing down “do not resuscitate” on a progress note
TABLE 1. Characteristics Gender Male Female Mean age (range) Site of cancer Lung Stomach Leukemia Colorectal Lymphoma Hepatobiliary Breast Pancreas Others
CIRCUMSTANCES SURROUNDING DNR ORDER n (296)
%
296 169 127
100 57 43
296 0 0
100 0 0
135 161
46 54
251 45
85 15
DNR consent Preprinted written form Verbal consent Presence at DNR discussion Family alone Patient alone Patient family Purpose of DNR consent Forbid resuscitation Imply palliative care Place of DNR discussion General ward Intensive care unit DNR, do not resuscitate.
from the primary attending physician after a family meeting regarding DNR directive. Unilateral physician decision to withhold CPR was made in 62 (16%) cases. Cardiopulmonary resuscitation was performed on 29 (7%) patients. The most frequent reason to use implement terminal CPR was deterioration of the patient’s condition without prior discussion about DNR (Table 3).
Time to death from DNR Figure 1 describes the time interval between the signing of the DNR consent and the eventual death of the patient. The mean date of DNR consent before death was 6 days (range, 0–69; median, 3). DNR discussion was held to 228 patient’s surrogates (77%) within 1 week of their death. On the day of a patient’s death, 57 patients (19%) were determined to have met the DNR status according to the wishes of the family.
DISCUSSION
PATIENT CHARACTERISTICS n (387)
%
231 156 61 (20–90)
60 40
79 67 42 37 30 24 15 14 79
20 17 11 10 8 6 4 4 20
This observational study found that 296 terminal patients with cancer among 387 admitted to large academic hospitals had a consent of DNR order at some point during their last admission. All DNR discussion was made between the physician and family without involving the patient. The DNR implementation was different across individual institutions. In most cases, DNR discussion was made at a time close to death. Terminal patients with cancer have a right to be provided with information so that they can take part in decision making about their treatment. Because many physicians and family do not disclosure the truth to the patient, many terminal patients with cancer do not
1156
KIM ET AL. TABLE 3. AS
CARDIOPULMONARY RESUSCITATION PERCENT OF ALL PATIENTS
Terminal CPRa Reason for CPR Family request Clinical deterioration without prior DNR discussion Misunderstanding of patient’s DNR status
n
%
29
100
4 24
14 83
1
3
aPatient who died while undergoing futile CPR CPR, cardiopulmonary resuscitation; DNR, do not resuscitate.
have enough information to decide on their treatment. Disclosing the truth to a patient with terminal cancer is traumatic and stressful for both patients and physicians. Many international studies have demonstrated this fact. In the United Kingdom and Italy, physicians are apt to withhold information from the patient at the request of the family.6,7 In China, physicians believe that causing patients to lose hope by telling them about the illness might hasten their death.8 Also, our study showed that a DNR discussion with patients with refractory cancer in Korea always took place between the physician and family members, irrespective of the ability of the patient to communicate. Unfortunately, surrogate decision-making does not ensure an accurate reflection of the patient’s wishes. In a study by Zweibel et al.,4 the proxies underestimated or overestimated patient preferences about resuscitation in 44% of cases. In a recent Korean study about treatment preferences
FIG. 1.
for resuscitation, 16.7% of physicians and 53.1% of family members favored the use of CPR.3 The proportion of family members who favor the use of CPR was relatively high. Although there are no reports comparing the treatment preferences between patients and their family members in Korea, this finding reflects that family members may consent to more aggressive treatment than patients would want for themselves. Fortunately, current studies in Korea indicate an improvement. In 1982, only 18% of physicians in Korea were likely to disclose bad news to patients but in 1990 81.8% agreed to do so.9,10 From a patient viewpoint, the recent study shows evidence that patients prefer to be involved in end-of-life care discussions; 96.1% of Korean patients with cancer wanted to be told if they had a terminal illness.11 It seems a consensus is emerging in Korea that end-of-life decisions should be the shared responsibility of physicians and patients and their families. The original purpose of a DNR order was to forbid resuscitation and this has remained its primary purpose as the use of such orders has developed. However, the concept was expanded to include not performing invasive medical intervention. In the United States, the consent for DNR order is often interpreted as the first step toward palliative care.12,13 Because of the broad scope of DNR orders, many institutions in Western countries have implemented different levels of care with respect to resuscitation. Our study found that several patients or family members refused any medical intervention, such as blood tests, administration of an-
Timing of do-not-resuscitate (DNR) issue before death.
DNR PRACTICES FOR TERMINAL PATIENTS WITH CANCER IN KOREA
tibiotics or fluids, transfusion, hemodialysis, or ICU transfer prior to the patient’s death irrespective of the patient’s resuscitation status. We suggest that different levels of care, beyond the resuscitation status, should be set for terminal cancer patients in Korea, instead of simply writing down the abbreviation “DNR” or using vague terms such as “minimal supportive care” or “best palliative care.” If the people participating in patient care interpret DNR orders differently, the patient may not be treated optimally. We suggest to clearly documenting DNR in a standardized format, because it is a definite medical decision and procedure. Are there circumstances in which it is acceptable not to discuss resuscitative measures with patients or families? The British Medical Association recently released its updated guideline, stating “for a person who almost certainly will not benefit from CPR, it should not be presented as a treatment option. Whether this is discussed is a matter of judgment.”14 Despite these guidelines, many health care professionals still argue that a unilateral DNR decision is justifiable. Because our study was carried out retrospectively, there is no way to tell why a physician made a DNR decision without involving the family. Further investigation is required to clarify this matter in Korea. The issue of DNR discussion at a late stage that was observed in our finding is similar to a study that was recently under taken at a single center in Korea.15 According to that study, DNR was discussed with half of the patients prior to within 1 week of death. When is the appropriate time to discuss DNR? The time required for the patient to emotionally and rationally come to terms with the finality of his/her condition seems to be the answer to this question. We think truth disclosure about terminal illness should proceed smoothly in solving the problem related to the late DNR decision. Advanced discussion of DNR with terminal patients with cancer earlier might prevent the medical futility of performing unnecessary treatments.16,17 In our study, the most frequent reason for administering CPR was “not having a prior DNR discussion.” It may be related to a late DNR decision. Another reason was the family’s request for a resuscitation attempt. A physician’s judgment of when to say “no” or to do “everything that has to be done” as a result of demands by the patient’s family is complex. Two possibilities come to mind. First, “do not resuscitate” practice is not permitted by law in Korea, as mentioned previously. In 1997, there was a sensational legal case in which a jury convicted doctors of murder. These doctors had discharged an intracerebral hemorrhage pa-
1157
tient with a low possibility of recovery. The doctors had accepted the request of the spouse of the patient to discharge the patient after a discussion of an expected grave prognosis. Later, the patient’s sibling accused the patient’s spouse and doctors of negligence. Given this event, many Korean physicians tend to accept the family’s resuscitation request if a definite DNR decision was not made between families. Therefore, a legal initiative may be necessary to achieve full implementation of DNR orders. Second, medical futility is an unfamiliar term to many physicians in Korea. Even in Western countries, where the patient’s autonomy is relatively well understood, studies regarding “medical futility” appear to be on the increase.18–20 We observed a rare treatment withdrawal, including treatment with intravenous antibiotics, ICU care, inotropics use, hemodialysis, etc., between the DNR-status patients and the CPR-status patients who were enrolled in this paper (unpublished data). Several limitations are noted in our study. First, measurement bias likely exists given a study designed on chart review. Medical charts may not have identified events that occurred but were not recorded. That being the case, there is no way we can identify the circumstances surrounding the DNR orders. Second, it is known that there are variations in DNR practices among institutions, ethnic groups and social levels. Therefore, our study cannot generalize about DNR decisions for all terminal patients with cancer in Korea. In summary, this is the first Korean study to evaluate the Do Not Resuscitate Order at multi-center teaching hospitals. In order to protect the autonomous right of a patient to make health care decisions, physicians in Korea should try to explore and understand the wishes of a patient regarding terminal care. This could prepare physicians and patients for the international evolution toward paying more attention to setting up advance care planning. Systemic standardization of DNR orders in Korea would benefit end-of-life decisions toward a patient-centered maximal comfort care in the earlier stage of terminal cancer.
REFERENCES 1. Josson PV, McNamee M, Campion EW: The ‘do not resuscitate’ order. Arch Intern Med 1988;148:2373–2375. 2. Rabkin MT, Gillerman G, Rice NR: Orders not to resuscitate. N Engl J Med 1976;295:364–366. 3. Oh DY, Kim JE, Lee CH, Lim JS, Jung KH, Heo DS, Bang YJ, Kim NK: Discrepancies among patients, family members, and physicians in Korea in terms of values regarding the withholding of treatment from patients with terminal malignancies. Cancer 2004;100:1961–1966.
1158 4. Zweibel NR, Casset CK: Treatment choices at the end of life: A comparison of decisions by older patients and their physician-selected proxies. Gerontologist 1989;29:615– 621. 5. Hayes S, Stewart K: The role of audit in making do not resuscitate decisions. J Eval Clin Pract 1999;5:305–312. 6. Jenkins V, Fallowfiled L, Saul J: Information needs of patients with cancer: Results from a large study in UK cancer centers. Br J Cancer 2001;84:48–51. 7. Gordon DR: Culture, cancer, and communication in Italy. In: Pfleider B, Bibeau G, eds. Anthropoalgies of Medicine. Heidelberg: Vieweg, 1991, pp. 137–156. 8. Muller JH, Desmond B: Ethical dilemmas in a cross-cultural context: A Chinese example. West J Med 1992;157: 323–327. 9. Cho DY: Attitudes of Korean physicians toward telling truth to the dying patient. J Korean Neuropsychiar Assoc 1981;20:1–16. 10. Han SW, Chung HY, Han SH: A study on the attitudes toward dying patients. J Korean Neuropsychiar Assoc 1990; 29:1408–1425. 11. Yun YH, Lee CG, Kim SY, Lee SW, Heo DS, Kim JS, Lee KS, Hong YS, Lee JS, You CH: The attitude of cancer patients and their families toward the disclosure of terminal illness. J Clin Oncol 2004;22:307–314. 12. Sulmasy DP, Sood JR, Ury WA: The quality of care plans for patients with do not resuscitate orders. Arch Intern Med 2004;164:1573–1578. 13. Fins JJ, Miller FG, Acres CA, Bacchetta MD, Huzzard LL, Rapkin BD: End of life decision making in the hospital: Current practice and future prospects. J Pain Symptom Manage 1999;17:6–15. 14. British Medical Association: Withholding and Withdrawing Life Prolonging Medical Treatment: Guideline for Decision Making. London: BMJ Books, 2001.
KIM ET AL. 15. Oh DY, Kim JH, Kim DW, Im SA, Kim TY, Heo DS, Bang YJ, Kim NK: CPR or DNR ? End of life decision in Korean cancer patients: A single center’s experience. Support Care Cancer 2006;14:103–108. 16. Curtis JR, Park DR, Krone MR, Pearlman RA: Use of the medical futility rationale in do-not-attempt-resuscitation orders. JAMA 1995;11:124–128. 17. Marik PE, Craft M: An outcome analysis of in hospital cardiopulmonary resuscitation: The futility rational for do not resuscitate orders. J Crit Care 1997;12:142–146. 18. Tomlinson T, Michalski AJ, Pentz RD, Kuuppelomaki M: Futility care in oncology: When to stop trying. Lancet Oncol 2001;12:759–764. 19. Cantor MD, Braddock CH 3rd, Derse AR, Edwards DM, Logue GL, Nelson W, Prudhomme AM, Pearlman RA, Reagan JE, Wlody GS, Fox E; Veterans Health Administration National Ethics Committee: Do-not-resuscitate orders and medical futility. Arch Intern Med 2003;163:2689–2694. 20. von Gruenigen VE, Daly BJ: Futility: Clinical decisions at the end of life in women with ovarian cancer. Gynecol Oncol 2005;97:638–644.
Address reprint requests to: Soon Nam Lee, M.D. Section of Medical Oncology Department of Internal Medicine School of Medicine Ewha Womans University 911-1 Mok-6-dong Yangcheon-gu, Seoul, 158-713 Republic of Korea E-mail:
[email protected]
