Writing DNAR Orders: A Frequently Missed Opportunity

Abstract & Commentary

Synopsis: Among medical patients admitted to an acute-care ward who were judged by an expert panel to be appropriate for do-not-attempt resuscitation (DNAR) orders, more than half did not have such orders written. Attending physicians offered several reasons for not writing DNAR orders, although they claimed not to be uncomfortable with or morally opposed to them.

Source: Eliasson AH, et al. Arch Intern Med 1999;159:2213-2218.

In this study from walter reed army medical center, Eliasson and colleagues sought to determine whether orders not to attempt resuscitation (DNAR) were written when appropriate and, when they were not, the reasons for not writing them. Eliasson et al reviewed the records for all admissions to the general medicine service during a four-month period. After the patients were discharged from the hospital, a five-member panel determined whether DNAR orders were indicated using a screening tool consisting of features of different diagnoses that suggested a poor prognosis. When the panel’s assessment was that a DNAR order was indicated, and no such order had been written during the hospital stay, one of the investigators interviewed the patient’s attending physician to determine the factors that had interfered with the writing of the DNAR order.

There were 613 admissions to the medical service during the study period. Of these, 149 (24%) were judged to merit a DNAR order, but in 88 of these patients (59%) DNAR orders were lacking. Absence of a DNAR order did not correlate with patient age, gender, or race.

The three most common reasons given by attending physicians for not having written a DNAR order during the patient’s hospitalization were: (1) the belief by the attending physician that the patient was unlikely to die during this hospitalization (in 56% of patients); (2) the belief that the patient’s primary physician should discuss CNAR issues with the patient (46%); and (3) the lack of an appropriate opportunity to discuss end-of-life issues with the patients or their family (43%). In 12% of the cases, patients or their families did not accept a physician’s recommendation for a DNAR order. None of the attending physicians stated that moral objections to a DNAR order, medical-legal considerations, or personal discomfort with discussing the topic played a role in their failure to write a DNAR order during their patient’s stay.

COMMENT BY DAVID J. PIERSON, MD, FACP, FCCP

This study shows that, at Eliasson et al’s institution, DNAR orders were often not written when an expert panel felt in retrospect that they should have been. It also shows that physicians cite several different reasons for not writing DNAR orders, including the belief that the illness "wasn’t serious enough," not being familiar enough with the patient to broach the issue, and the conviction that discussing end-of-life issues with the patient was "someone else’s job." I suspect that these findings are representative of physician behavior at other institutions as well.

Although the Patient Self-Determination Act mandates that everyone admitted to the hospital be offered the opportunity to make a living will and to make known their wishes for attempted resuscitation, this act has had little affect, and the majority of patients do not make such wishes known to their physicians. The present study and others have also shown that physicians tend not to seize the opportunity to discuss end-of-life issues with their patients or, when appropriate, to write DNAR orders. This was a study of general medical ward patients, not of patients admitted to an ICU. Because the majority of patients who die in ICUs have life support withheld or withdrawn, it appears that physicians are less reluctant to discuss these issues with patients or their families at times of more obviously life-threatening illness.

Nonetheless, for patients with medical illnesses serious enough for a DNAR order to be "indicated," according to the five-member expert panel in this study, such orders were written less than half the time. It is symptomatic of the fragmentation of contemporary medicine that inpatient attending physicians often felt that they did not know their patients well enough to discuss DNAR status, or considered such discussions the job of others in the health care system. Finding ways around these and the other obstacles to appropriate discussions of end-of-life issues that were identified in this study is a difficult but important challenge for the future.

Reasons given by attending physicians for not having written a do-not-resuscitate order during the admission of a patient with serious medical disease include:

a. the belief by the attending physician that the patient was unlikely to die during this hospitalization.

b. the belief that the patient’s primary physician should discuss CNAR issues with the patient.

c. the lack of an appropriate opportunity to discuss end-of-life issues with the patient or their family.

d. All of the above

e. None of the above