The Wrongful Resuscitation
An 80-year-old man with diabetes, peripheral vascular disease, bilateral below-the-knee amputations, and poor quality of life had previously been resuscitated from sudden death. After his recovery, he completed a DNR (do not resuscitate) form signifying his desire to avoid such treatment in the future.
The patient presented to the emergency department (ED) in extreme pain and was found to have a ruptured abdominal aortic aneurysm (AAA). Although his DNR form was with him, neither the ED staff nor the consulting surgeon looked at it. The patient was rushed to the operating room (OR), where his AAA was repaired. Postoperatively, an internist came upon the DNR form in the patient's chart and discussed resuscitation preferences with the patient and the family. The patient reconfirmed his desire to avoid resuscitation and heroic procedures, expressing anger that he had been taken to the OR for the AAA repair. The family agreed with the patient's choice. The internist wrote a DNR order in the chart, but the surgeon—having just completed major surgery on this patient—was furious, changing the code status back to "full code." Ultimately, the internist consulted with the hospital ethicist, who convinced the surgeon to honor the patient's and family's wishes. The DNR order was reinstated, and the patient later died of a cardiac arrest during the hospitalization.
The fear of older, dying persons having life prolonged against their wishes has in part driven the current focus on completing written advance directives in the United States. This focus may be somewhat surprising, because the published evidence indicates that patients being resuscitated against their wishes is a rare event. In a review of 17 narratives from the Study to Understand Prognoses and Preferences for Risks of Treatments (SUPPORT), there were no cases in which a physician unilaterally ignored patient preferences to provide life-sustaining treatment.(1) At least in SUPPORT, the predominant concern was the converse: that physicians did not write a Do Not Resuscitate (DNR) order in nearly 1 in 2 cases when there was evidence (based on patient or family member interview) of that patient's preference to forgo resuscitation.(2) Although this failure to document a DNR preference is concerning, the absence of a DNR order did not result in resuscitation being performed against a person's wishes in SUPPORT.(3) Moreover, a 1995 study found that only 26% of hospitalized nursing home residents who had previously executed an advance directive had these preferences recognized during that hospitalization. However, when the advanced directives were recognized, they influenced treatment decisions in 86% of cases.(4) Thus, "near misses" do not frequently translate into cases where a patient is wrongfully resuscitated.
Although the rarity of the error of physicians "trumping" or ignoring a patient preference by mistake is somewhat reassuring, there do appear to be many cases in which patients are at risk of receiving treatment inconsistent with their previously stated wishes, and patients seem to harbor significant concerns regarding the lack of communication.
While case reports of physicians unilaterally trumping patient preferences exist, as the surgeon did in this case by reversing the DNR order against a patient's preferences, fortunately they are rare. In this example, an ethics committee was utilized to resolve the conflict with a surgeon who was unilaterally making a decision against a patient's informed wishes. An ethics committee is an appropriate vehicle to help resolve these conflicts. It should be noted that the case as presented does not question the authenticity of the patient's decision to complete a DNR order. Indeed, the patient signed a DNR order in the nursing home, and the family agreed with the patient's wishes and decision. This leads to a very important question regarding the actions of the surgeon. Should that physician be sanctioned for making a decision that flagrantly ignored a patient's preference? I believe that such behavior should receive the same scrutiny as operating on the wrong side of the brain—each is a bodily assault that provides harmful care to which the patient did not consent. Such a case should be reported to the appropriate state authorities, and the resulting sanctions and corrective actions should ensure that informed preferences by future patients are honored.
Although the actions of the surgeon in this case raise important concerns, two other issues in this case also deserve greater discussion. A critically ill nursing home resident is transferred to the ED with severe abdominal pain and hypotension. In the midst of crisis, the physician did not realize that the patient was DNR and successfully resuscitated this elderly patient. This is an important error. Transitions of care involving frail, older persons are often problematic. The error of resuscitating a patient because health care providers were unaware of his code status reflects inadequate communication and coordination of care between the nursing home and the ED. Steps can be undertaken to prevent this error, including the following:
- The use of a bracelet identifying the patient’s DNR status is increasingly being adopted by states to ensure that patients are not resuscitated against their wishes by emergency medical services (EMS) staff and other health care providers. The majority of states have adopted an out-of-hospital DNR policy. Physicians being aware of their state law and following the state protocol for out-of-hospital DNR orders are critical steps to ensure that patient preferences are honored.(5)
- When a frail, older nursing home resident is transferred to an acute care setting, not only should an interagency form be completed, but a verbal report should be called to the ED. Each ED must develop a system that ensures that information from that verbal report becomes part of that patient’s treatment record. Electronic medical records can include appropriate "flags" that make sure that important information such as DNR orders and drug allergies is immediately accessible to treating physicians. The primary care physician should follow up with a phone call 24 hours after admission to review the patient’s care and ensure that the treating team at the hospital has all the information needed to ensure continuity of the care plan.
The second issue, which may not seem like an obvious error, reflects one of the predominant concerns with decision making at the close of life. Too often, the timing of discussions to clarify patient preferences only occurs in the last days of life. As the previously cited data illustrate, the error of "wrongful" resuscitation is exceedingly rare. Rather, the more common error occurs when the physician has not written a DNR order because the patient's end-of-life wishes have not been clarified. It is this delayed communication that can lead to higher health care costs and higher utilization of the intensive care unit (ICU) for the seriously ill.(6) For the frail, older nursing home resident, the discussion of CPR is somewhat of a moot point given the poor outcomes of resuscitation. Murphy and colleagues (7) found that residents age 70 and older are resuscitated successfully in fewer than 1 in 100 cases, matching a threshold that has been proposed for a treatment to be deemed "futile." In this 80-year-old man with multiple organ impairment, the most appropriate focus of advance care planning should have been on the role of the acute care hospital. If the physician and nursing home staff addressed the appropriate use of hospitalization and prospectively discussed the goals of care with the patient and family, I strongly suspect that the high costs of terminal hospitalization could have been avoided.
Increasingly, nursing homes are utilizing "Do Not Hospitalize (DNH) Unless for Comfort" orders in frail, older nursing home residents. In 2000, nearly 1 in 10 nursing home residents with advanced dementia had a DNH order.(8) The frequency of such orders varied substantially across the United States, ranging from 0.7% in Oklahoma to 25.9% in Rhode Island. Regions with higher rates of DNH orders for persons with advanced dementia had fewer patients admitted to the ICU during a terminal hospitalization. In cases like this one, clarifying patient preferences regarding the use of the acute care hospital in the nursing home setting is an important way to prevent inappropriate terminal hospitalizations. Too often, the question asked about the use of life-sustaining treatment is, "Do you want me to resuscitate your mother?" or "If she gets sick, should we send her to the hospital?" Rather, health care providers need to be trained in a goal-based approach to advance care planning in the nursing home setting.(9,10) There are three key steps in this approach.(10)
First, listen to the patient and family. A key step in advance care planning is to listen and understand where the patient and the family are in living with multiple chronic illnesses. Has the patient reached a point where continued existence is burdensome? Does the patient have preferences that should help guide his or her treatment decisions?
Second, based on understanding where the patient is in his or her disease trajectory, the health care provider should clarify misconceptions, state what they heard the patient and/or family tell them about where they are in the disease process, and formulate goals of care. For example, "Given what you have told Mrs. Smith, your goal is to focus on your comfort, not interventions that would prolong your life." For the nursing home resident, typical goals could include treatment to focus on extending life regardless of their current state of health or efforts to restore them to their current state of health. However, if that is not possible, care should focus on comfort even if it shortens life. Such a focus on comfort may lead to a decision to forego further hospitalizations.
Third, working with the nursing home staff, the physician formulates a care plan that ensures that those goals are met. Often, this includes a referral to hospice that will work with the staff in the nursing home to formulate a plan for palliation of symptoms and avoiding hospitalization against the wishes of the nursing home residents and/or family.
Increasingly, decisions regarding medical care for frail, older persons involve making choices that involve a trade off between quality and quantity of life. Eliciting and respecting an informed patient's preferences are key to making this decision. Communication needs to occur early in the disease course. A 3-step goal-based approach to advance care planning listens to where the patient is in living with multiple chronic illnesses, formulates goals of care, and then develops a plan of care that outlines how those preferences will be honored. Key parts of this plan include taking steps to ensure that the goals of care are communicated across settings of care and considering whether transfers from setting to setting (including nursing home to hospital) are appropriate in the first place.
- A physician unilaterally trumping an informed patient's choice should be treated as a sentinel event.
- Good communication (including telephone calls from the primary care physician to the ED staff and the admitting physician 24 hours after admission) is important to ensure continuity of information about a patient’s advance care plans and medical condition.
- Physicians should be familiar with the out-of-hospital DNR laws in their state and counsel patients to wear the appropriate bracelet or other means of signifying that the patient is DNR as prescribed by state law.
- For frail, older nursing home residents, a key point of clarification as part of advance care planning surrounds the role of hospitalization. Such clarification may result in the use of a "Do Not Hospitalize Unless for Comfort" order.
Joan M. Teno, MD, MS Professor of Community Health and Medicine The Warren Alpert School of Medicine at Brown University Associate Medical Director, Home and Hospice Care of Rhode Island
1. Teno JM, Stevens M, Spernak S, Lynn J. Role of written advance directives in decision making: insights from qualitative and quantitative data. J Gen Intern Med. 1998;13:439-446. [go to PubMed]
2. Teno JM, Mitchell S, Rhodes R, Intrator O, Brostrup-Jenson C, Mor V. Health Care Transitions and Incident Feeding Tube Insertion Among Persons with Advanced Cognitive Impairment: Lost in Transition. 2007.
3. A controlled trial to improve care for seriously ill hospitalized patients. The study to understand prognoses and preferences for outcomes and risks of treatments (SUPPORT). JAMA. 1995;274:1591-1598. [go to PubMed]
4. Morrison RS, Olson E, Mertz KR, Meier DE. The inaccessibility of advance directives on transfer from ambulatory to acute care settings. JAMA. 1995;274:478-482. [go to PubMed]
5. Sabatino CP. Survey of state EMS-DNR laws and protocols. J Law Med Ethics. 1999;27:297-315, 294. [go to PubMed]
6. Teno JM, Fisher ES, Hamel MB, Coppola K, Dawson NV. Medical care inconsistent with patients' treatment goals: association with 1-year Medicare resource use and survival. J Am Geriatr Soc. 2002;50:496-500. [go to PubMed]
7. Murphy DJ, Murray AM, Robinson BE, Campion EW. Outcomes of cardiopulmonary resuscitation in the elderly. Ann Intern Med. 1989;111:199-205. [go to PubMed]
8. Mitchell SL, Teno JM, Intrator O, Feng Z, Mor V. Decisions to forgo hospitalization in advanced dementia: a nationwide study. J Am Geriatr Soc. 2007;55:432-438. [go to PubMed]
9. Gillick MR. Adapting advance medical planning for the nursing home. J Palliat Med. 2004;7:357-361. [go to PubMed]
10. Teno JM. Do-not-resuscitate orders and hospitalization of nursing home residents: trumping, neglect, or shared decision-making at the eleventh hour. J Am Geriatr Soc. 2004;52:159-160. [go to PubMed]