NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Holzheimer RG, Mannick JA, editors. Surgical Treatment: Evidence-Based and Problem-Oriented. Munich: Zuckschwerdt; 2001.

Cover of Surgical Treatment

Surgical Treatment: Evidence-Based and Problem-Oriented.

Show details

Ethical considerations

, M.D.H. and , M.D., Ph.D.

Author Information

The introduction of cardiopulmonary resuscitation (CPR) in 1960 was followed by its widespread administration to all patients suffering an in-hospital cardiopulmonary arrest. As clearer ideas about the ultimate likelihood of success emerged, the concept of withholding resuscitation began. Currently, the majority of patients that die in the intensive care unit (ICU) have had some form of treatment withheld or withdrawn. In a 1995 survey, 96% of intensivists withheld or withdrew treatment in the previous year (Asch 1995). The most common reasons for withholding treatment are poor prognosis or irreversible condition, followed by estimation of poor quality of life and alleviation of pain and suffering. Approximately 70% of in-hospital deaths are preceded by a do not resuscitate (DNR) order, and 9–13% of all patients admitted to an ICU will have a DNR order written at some point during their stay (Rapoport 1996).

The decision to withdraw or withhold life-sustaining treatment is a difficult one, as it is contrary to preserving life, a basic principle of medical ethics. There are some ethical and legal precedents that have been established to guide decisions regarding limitation of care. The ethical duty to alleviate suffering is now viewed as important as preserving life. Withholding and withdrawing treatment are ethically and legally equivalent. Furthermore, nutrition and hydration are equivalent to more aggressive modalities such as vasopressors and mechanical ventilation, and can be refused by both patients and surrogates.

The principle of patient autonomy governs most decisions in American medicine. It holds that each person has the right to make decisions with respect to which treatments will be accepted or rejected, and was the cornerstone of the President's Commission on decisions to forego life-sustaining treatment (President's Commission, 1983). In order for the patient's wishes to be carried out, the physician must know those patient's wishes. This can happen by direct communication with the patient, or with someone in whom the patient has confided. Unfortunately, physicians know patient preferences about life-sustaining treatment less than 40% of the time (Bedell, 1984;SUPPORT 1995). Inability or unwillingness of patients to express their wishes, inability or unwillingness of physicians to discuss DNR preferences, lack of a primary care physician, and lack of a supportive family may contribute to the communication barrier. Advance directives in the form of a living will, health proxy, and durable health care power of attorney are ways in which the patient can communicate his or her wishes, although they are not always recognized as legally binding documents. Unfortunately, even when advance directives are available and known to the physician, they are not followed in a significant number of patients (Danis, 1991). Since the Patient Self-Determination act of 1990, all hospitals in the United States have been required to provide patients with the opportunity to complete these forms.

Without clear knowledge of a patient's wishes, decisions regarding life-sustaining treatment may be made by a surrogate. Spouse, adult children, parents, more distant relatives and friends can function as surrogates, usually in that order. Discussions with surrogates about resuscitation should emphasize what the patient would want for him or herself, not necessarily what the surrogate would want for the patient. In rare cases where no surrogate is available and the patient has a clearly terminal condition, input from an ethics committee may be necessary.

Consensus statements regarding end of life decisions have focused on some of the above ethical, legal, and medical considerations (references 7–10). Although each decision concerning limitation of care must be individualized, there are some broad questions that have been further studied.

Does the presence or type of hospital DNR policy influence DNR orders or discussions about limiting care in any way?

Data support a grade III recommendation for a DNR policy in all hospitals.

The presence and composition of DNR policies varies from hospital to hospital, although guidelines from the American Hospital Association suggest policies should be humanistic, practical, and disseminated to those who will be using them. Compliance with a DNR policy increases in the years after its implementation. The presence of a DNR policy may increase the number of patients with DNR orders and increase the amount of consultation done by the responsible physician (level III and IV). More data support that the documentation behind DNR orders and clarity of specific interventions increases when a policy is implemented without increasing the overall number of patients with DNR orders (levels II-V). The patient self-determination act appears to have had no influence on the number of hospitalized patents with DNR orders, perhaps because only 41–65% of patients receive information about advance directives (level II and IV). Interestingly, the ICU environment and specifics of the DNR policy do not appear to affect the number of DNR orders (level III). In a recent survey, over 90% of hospitals had some type of policy for DNR orders.

Does the presence of a DNR order influence resource use or nursing care?

There are not enough quality data to provide an evidence-based recommendation about the use of DNR orders to influence resource use or nursing care. Further study might address whether earlier use of DNR orders would result in decreased resource consumption and level of nursing care.

It has been the assumption that patients with a DNR order will consume fewer resources and require less intensive nursing care. This assumption is supported by several studies (level V). Other studies do not support this contention, and find that patients with DNR orders consume an equivalent or a greater number of resources, and require either the same or increased level of nursing care (level II-IV). This apparent conflict may be due to a heterogeneous population of patients with DNR orders written both early and late in their ICU course. Once patients have a DNR order, resource consumption as measured by length of stay or the therapeutic intervention scoring system (TISS) may decrease (level III), although level of nursing care may remain the same (level III) or decrease (level V).

Can education change physician practices regarding end of life decisions?

Data support a grade II recommendation to provide ethics education for physicians in training, and education about appropriate forms to facilitate advance directives.

Level V surveys support the fact that the majority of patients want to discuss resuscitation preferences with their physicians, although only a minority have done so. Level II and III data document physician discussion with patient/surrogate in a minority of cases. General information about patient prognosis and family wishes, conveyed by nurse clinicians to practicing physicians, is not effective in increasing patient-physician communication regarding DNR preferences or changing the incidence or timing of DNR orders (level II), although level V data indicate that providing information about patient prognosis will influence whether or not hey opt for resuscitation. Ethics education that includes specifics of the hospital DNR policy and case studies of difficult ethical dilemmas, in some cases accompanied by a new form designed to augment the traditional DNR order increases the number of care concerns addressed with each DNR order (i.e. wishes regarding vasoactive drugs, antibiotics, withdrawing mechanical ventilation, nutrition) (level II).

Do patients want to discuss DNR preferences?

Data support a grade III recommendation for discussion about resuscitation practices. Discussion early in hospital course may allow more patients to be involved and decrease the need for surrogate decision-making.

The majority of patients want to discuss resuscitation options with their physicians, and prefer to do so early in the course of disease (level V). More DNR discussions held with family/surrogate that with patient, often secondary to the declining mental status of the patient throughout his or her hospital course (level II-V). Problems identifying and contacting a surrogate are among the major problems implementing a DNR order (level V). Surrogates also overestimate patient desire for resuscitation (level V). Early discussion may increase the number of DNR orders by appropriately determining the wishes of the patient.

There are many considerations that go into each decision about limiting or withdrawing care. There are some guiding principles; respect for patient autonomy, compassion, balance between preserving life and relieving suffering. There are also practical guidelines; early establishment of open communication, use of clear language, allowing time for all decisions to be reached. Effective and compassionate care at the end of life requires not only adherence to the minimal recommendations that are supported by evidence, but recognition of these principles as well.


Asch D A, Hansen-Flaschen J, Lanken P N. Decisions to limit or continue life-sustaining treatment by critical care physicians in the United States: Conflicts between physicians practices and patients wishes. Am J Respir Crit Care Med. (1995);151:288–292. [PubMed: 7842181]
Rapoport J, Teres D, Lemeshow S. Resource use implications of do not resuscitate orders for intensive care unit patients. Am J Respir Crit Care Med. (1996);153:185–190. [PubMed: 8542114]
President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research (1983) Deciding to forego life-sustaining treatment: Ethical, medical, and legal issues in treatment decisions. US Government Printing Office, Washington DC .
Bedell S E, Delbanco T L. Choices about cardiopulmonary resuscitation in the hospital. N Engl J Med. (1984);310:1089–1093. [PubMed: 6708990]
SUPPORT investigators. A controlled trial to improve care for seriously ill hospitalized patients. JAMA. (1995);274:1591–1598. [PubMed: 7474243]
Danies M. et al. A prospective study of advance directives for life-sustaining care. N Engl J Med. (1991);324:882–888. [PubMed: 2000110]
ACCP/SCCM Consensus Panel. Ethical and moral guidelines for the initiation, continuation, and withdrawal of intensive care. Chest. (1990);97:949–958. [PubMed: 2182302]
Council on Ethical and Judicial Affaires, American Medical Association. Guidelines for the appropriate use of do-not-resuscitate orders. JAMA. (1991);265:1868–1871. [PubMed: 2005737]
Rubenfeld G D. Do-not-resuscitate orders: A critical review of the literature. Respiratory Care. (1995);40:528–537. [PubMed: 10142888]
Task force on ethics of the Society of Critical Care Medicine. Consensus report on the ethics of foregoing life-sustaining treatments in the critically ill. Crit Care Med. (1990);18:1435–1439. [PubMed: 2245622]
Copyright © 2001, W. Zuckschwerdt Verlag GmbH.
Bookshelf ID: NBK6980


  • PubReader
  • Print View
  • Cite this Page

Related information

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...