End-of-Life Care and Decisions About Life Sustaining Treatments

Introduction
Older adults who are dying in the critical care setting need the same quality care as is given to any patient who is facing his or her final days in critical care. Where the care of older adults differs is often in decision making, the consequences of care decisions, and in modifications to symptom management to account for the physiologic changes seen with aging. Good end-of-life care in any setting requires effective communication with the patient and the family, facilitation of decision making, management of complications of the treatment and the disease, symptom control, psychosocial– spiritual care of the patient and the family, and holistic care at the time of death (Foley & Gelband, 2001).

Decision making is often a key issue in ICU. Health care professionals who may not expect young adults to have planned for future medical needs are frequently surprised when older adults, who are generally in declining health, have not addressed future medical decision making. Families of older adults may not be prepared to make medical decisions when the patient loses decision-making capacity.
For older adults facing their final days, the consequences of their health care decisions may include facing death in the hospital or ICU setting or making the transition to another setting to die. Families may be poorly prepared for these consequences and transitions. Proactive discussion of the consequences of medical decision making, along with excellent management that allows transitions to be made as smoothly and seamlessly as possible, can improve the experience for patients and families.
For older adults who receive end-of-life care in the critical care setting, often a shift occurs from curative care or care that extends life, to comfort or palliative care. The focus of palliative care is on making the person as comfortable as possible by good symptom management, attention to spiritual and psychosocial needs, and providing support to the family. Good palliative care includes preparing both the patient and family for the death. Nurses also need to notify and support families immediately after the death and facilitate decisions about autopsy and organ donation.
Decision Making in Critically Ill Older Adults
The importance of medical decision making has been accentuated because of recent cases such as the Terri Schiavo case, the Patient Self-Determination Act, and research about whether patient preferences are followed in care decisions. A sentinel report, the Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments (SUPPORT) (SUPPORT Investigators, 1995), was a major impetus to the recent increase in research about end-of-life care. SUPPORT was conducted in the early 1990s and had two phases, an observational phase, which took place over 2 years and an interventional phase. Over 9,000 patients with one or more of nine life-threatening illnesses were enrolled from five medical centers. The intervention used was a nurse who interacted with the patient, family, and staff to provide information to the attending physicians on prognoses and patient/surrogate information and preferences for care. The mean patient age was 63 years, 44% of patients were female, and 16% were Black. Overall survival to 6 months was 53%. In risk analyses that adjusted for sex, ethnicity, income, baseline functional status, severity of illness, and aggressiveness of care, each additional year of age increased the risk of death by 1.0% for patients 18 to 70 years of age, and by 2.0% for patients older than 70 years of age. Estimates of 6-month mortality rates were 44% for 55-year-old patients, 48% for 65-year-old patients, 53% for 75-year-old patients, and 60% for 85-year-old patients. Surprisingly, age contributed less to the risk of death than aggressiveness of care, acute physiology, and diagnosis (Hamel et al., 1999). Based on this information, age should not be as important in decision making as comorbidities and the patient’s clinical condition. It is important not to make decisions solely on the patient’s age, but to carefully consider these other elements to determine the most appropriate treatment.
The Hospitalized Elderly Longitudinal Project (HELP) was a subset of the SUPPORT sample and consisted of adults 80 years or older. A total of 417 HELP patients died within a year of enrollment. Most of the patients (70%) said they preferred comfort care rather than care aimed at extending life (Somogyi-Zalud, Zhong, Lynn, & Hamel, 2000). A total of 72 HELP patients died during their enrollment hospitalization. Of these patients, 70% wanted comfort-focused care and 80% had a Do Not Resuscitate (DNR) order. Yet, of the 72 who died, 63% received one or more life-saving treatments, 54% were admitted to an intensive care unit, and 43% were on a ventilator before they died (Somogyi-Zalud, Zhong, Hamel, & Lynn, 2002). In general, these older patients preferred to have their families and the physician make decisions for them rather than strictly follow their communicated preferences at the end of life (Puchalski et al., 2000).
Although these patients preferred family and physicians making decisions, they frequently did not have the conversations required to inform physicians about their wishes. The SUPPORT study also revealed that even though most patients who had long (more than 14 days) ICU stays had a high likelihood of dying in the next 6 months (55% died), fewer than 40% of the patients reported that their physicians talked to them about their prognosis or preferences for life-sustaining treatment. Only 29% of the patients who preferred a palliative approach felt that the care they received was consistent with the goal of comfort care (Teno, Fisher, et al., 2000).
Although older patients were less likely to want aggressive treatments than younger patients, the majority of older patients wanted cardiopulmonary resuscitation, and many wanted care that emphasized life extension rather than comfort. Families and physicians underestimated the elders’ desire for aggressive treatment. An examination of age-related differences in hospital care for SUPPORT patients revealed that after adjustment for sociodemographic factors, severity of illness, and patients’ preferences for aggressive care, decisions to withhold life-sustaining treatments were more frequent for older patients (Hamel et al., 2000). These factors all contribute to a discrepancy between older adults’ preferences and the care they receive at the end of life.
Survival of Older Adults in Critical Care
There is conflicting information about the risk of death and chance of survival of older adults. The findings of the SUPPORT study indicate that as age increases, people have a greater risk of dying and a lower chance of survival. Even when adjustments were made for the use of less aggressive care in the older adult, this increased mortality risk because of age persisted (Hamel et al., 2000). Yet, for those dying following withdrawal of ventilatory support, there is no relationship between age and time to death (Campbell, 2007). Furthermore, older adult ICU patients have survival similar to younger patients when severity of illness is similar (Kleinpell & Ferrans, 1998) unless the patient is very old (85+) (Van Den Noortgate, Vogelaers, Afschrift, & Colardyn, 1999). Because the relationship of aging to mortality risk is not clear, care decisions should be made on the likelihood of survival of each individual rather than on the chronological age of the patient.
The likelihood of survival, or the prognosis in terms of approximate length of life of the older hospitalized adult, can be estimated from the clinical information in the medical chart and a brief interview with the patient or surrogate. By adding information about prognosis from the physician, and patient preferences in care, accuracy of the estimate can be improved (Teno, Harrell, et al., 2000). When physician estimates of intensive care survival are less than 10% it is more likely that life-sustaining treatment will be limited and this is more predictive of ICU death than illness severity, organ dysfunction, or the use of inotropes or vasopressors (Rocker et al., 2004).
Advance Care Planning
Flow diagram of possible outcomes of treatment.
Flow diagram of possible outcomes of treatment.

Under the Patient Self-Determination Act, hospitals are mandated to ask whether a person has an Advance Directive (AD). Despite this legislation and the wide promotion of ADs, the completion rates have not met recommended levels (Ramsey & Mitty, 2003). When ADs are present, they do not consistently guide care in ICUs (Goodman, Tarnoff, & Slotman, 1998). Even when the older adults have completed a living will or health care power of attorney, they might not have understood the form (Jezewski & Meeker, 2005) or the impact of these decisions. Chronically ill older adults tend to live life a day at a time, not looking too far into the future, and plan to “cross the bridge” when they get to it (Carrese, Mullaney, Faden, & Finucane, 2002) so they may not have completed an AD or discussed their preferences with the person they have designated as a surrogate decision maker or health care agent.
One approach that has been more consistently effective is the use of orders such as Physician Orders for Life-Sustaining Treatment (POLST). In one study of the older adults in an Oregon program, POLST was followed consistently for cardiopulmonary resuscitation (CPR) (91%), antibiotics (86%), IV fluids (84%), and feeding tubes (94%) (Lee, Brummel-Smith, Meyer, Drew, & London, 2000).
One challenge in advance care planning with older adults is that their preferences often shift over time (Ditto, Jacobson, Smucker, Danks, &Fagerlin, 2006), or the patients are not clear about their preferences (Schneiderman, Pearlman, Kaplan, Anderson, & Rosenberg, 1992). Further, documented preferences may not reflect the patient’s current wishes. When older adults were asked about end-of-life care situations before, soon after, and several months after a hospitalization, participants reported less desire to receive life-sustaining treatment at the posthospitalization interview than they did at the interview conducted prior to hospitalization. However, desire for life-sustaining treatment returned to near prehospitalization levels at the annual interview conducted several months after hospitalization (Lloyd, Nietert, & Silvestri, 2004). The timing of such discussions seems to be a factor in choices; recent memories of aggressive treatment tend to reduce the desire for such treatment.
Even when older adults are sure of their preferences concerning end-of-life care, they rarely have discussions with their physicians about their preferences. A recent study examined 80 patients older than 64 years to address preferences about end-oflife care and CPR as their disease progressed. Although 40% did not wish to have CPR, only two had previously discussed their CPR preferences with their physicians (Formiga et al., 2004). In another study of 115 patients with oxygen-dependent chronic obstructive pulmonary disease (COPD), less than one third reported having a discussion about end-of-life care with their physicians (Knauft, Nielsen, Engelberg, Patrick, & Curtis, 2005). It seems that neither patients nor health care professionals are raising the topic. A recent study showed that even in the oldest-old veterans, those greater that 85 years, only 50 of 149 (34%) had documented care preferences (Wu, Lorenz, & Chodosh, 2008). In this study only age and number of outpatient visits were associated with advance-directive completion, so perhaps as we age the issue is raised more often.
In addition to the POLST, there are other suggestions about improving advance directives or advance care planning. Recommendations include improving documentation of discussions, putting a short AD on Medicare cards (Pollack, 2000), simplifying completion (reducing requirements for witnessing), and emphasizing the need for discussion with the physician over completion of documents (Lo & Steinbrook, 2004). The use of gerontological advance practice nurses for advance care planning as well as coordination of care and symptom management has also been recommended (Henderson, 2004).
In the ICU, where patient conditions are unstable, it is often critical to assess patient preferences for care, but difficult to do so. The treatments and medications used in ICUs often prevent patients from participating in communication about wishes. Information provided by family members, especially when a consensus forms, is often the basis for decision making (Tonelli, 2005). This can be difficult on family members who may be elderly themselves and for adult children who have never discussed with their parents what their wishes for care would be. Another complication concerns the dynamics of the family. Older adults may have had multiple marriages and may have extended family who are scattered across the country. Reaching family consensus when the family includes the current spouse and adult children from a previous marriage can be difficult at best. State laws vary concerning who has the legal right to make decisions for an incapacitated patient without an AD.
Families may not understand the implications of the choices that they are asked to make. For example, in the case study that opened the chapter the family readily agreed to a perataneous endoscopic gastrostomy (PEG) tube for nutrition. Although the family was informed of the risks of PEG placement, they were not well informed of the risks and benefits of initiating tube feedings, including the risk of diarrhea, aspiration pneumonia, infection, ulcers, or the difficulties in stopping tube feedings. Figure 8.1 has examples of how such a scenario might occur. How information is presented can clarify or confuse the decision makers. When family members are asked questions such as, “If your family member is not able to breathe on her/his own do you want a tube put into help her/him?” or “If your family member’s heart stops, do you want to us to try to restart it?” they may feel that they are choosing death for the patient if they say no. As a result, families may choose more aggressive care for even the very old patient. In doing so, however, they might be choosing a longer and more painful road to death (Kaufman, 1998).
Careful communication is needed to present true options to families without encouraging medically futile treatment. Families often have unreasonable expectations about what medical treatment can offer or the seriousness of their family member’s condition (Winzelberg, Patrick, Rhodes, & Deyo, 2005). Communication about treatment options is complicated by the fact that physicians have a different view and different values than the family concerning what is best for the patient. Communication with family about the choices at the end of life requires careful consideration of patient and family values and ethical principles (Gordon, 2002). The final outcome should be a plan of care that is tailored to the patient’s and family’s values, beliefs, and wishes, and that offers the most appropriate treatment to meet the jointly determined goals of care. These goals may include returning to the prehospitalization state of functioning or palliation of symptoms.
Using words that are easily understandable by the family yet as clear as possible is recommended. Often families will use a euphemism for death such as “passing on” or “expire” and it is appropriate to use this term when speaking with a family. If they do not give such verbal clues, use the words “death” and “dying” to avoid misunderstandings. There are a few phrases that can be greatly clarified or improved with slight changes. For example, it is not appropriate to tell a family that you “know exactly how they feel.” This takes the focus off of the family and puts it onto you. It is better to use this as an opening for the family to express their feelings by saying, “Often families are very upset when facing these changes. Are you having a hard time with this?” Another term that is often used by professionals that can be misinterpreted by families is “failed.” We may say that the patient has failed chemotherapy or failed weaning from the ventilator. This implies to the family that the patient had some control over the situation. It is better to say that the treatment is no longer effective or that the ventilator was not able to be removed. Particularly disturbing to families are statements about “withdrawing care” and that there is “nothing more we can do.” Kirchhoff (2005) recommends that critical care staff stop using the phrase “withdrawing care” because it implies that if life-sustaining therapies are withdrawn, the patient will not receive any care. It is more appropriate to talk about changing to a comfort or palliative focus and to be clear that when we stop life-sustaining efforts, the patient will still get the best symptom management and comfort care possible.
When a conversation with the patient is not possible, having an informed surrogate who has had a recent discussion with the patient about end-of-life wishes is the best substitute. Recent pilot work suggests that surrogates can better understand and communicate the patient’s wishes when a trained facilitator interviews the patient in the presence of the surrogate (Briggs, Kirchhoff, Hammes, Song, & Colvin, 2004). In this study the interview group was compared to a group receiving standard advance directive care. Greater understanding on the part of the surrogate, such as occurs with this intervention, leads to a number of positive outcomes: increased congruence between the choices made by the patient and the surrogate, greater satisfaction with the process of decision making, and less decisional conflict.
Because most patients and their surrogates will not have had a facilitated discussion, it is important to find other ways to assist the decision-making surrogate. The first is to carefully review any advance-directive documents or documentation of recent discussions with the patient about his or her care preferences. In the absence of a clear, applicable advance directive or written documentation of the patient’s wishes, one of two principles is adopted: substituted judgment or best interests (Krohm & Summers, 2002).
The principle of substituted judgment is designed to guide the role of the decisionmaking surrogate. In using substituted judgment, the surrogate decision maker, whether or not that person is the legal power of attorney for health care, tries to determine what the patient would have wanted had he or she understood the circumstances under which treatment or procedures would be provided. One way to assist the surrogate to think this way is to ask, “If [the patient] had known what would happen and wrote you a note to read right now, what would he/she have said he/ she wanted done?” Although there are some ethical issues and controversy associate d with the use of substituted judgment (Brostrom, Johansson, & Nielsen, 2007; Nagasawa, 2008; Shapiro, 2007), it is still the approach that most courts favor (Krohm & Summers, 2002).
The principle of best interest is often the default of health care professionals and the principle that allows treatments to be withheld if they are medically futile. In using the principle of best interest, the surrogate decision maker must assess the risks and benefits of the available treatment options and try to choose the best option for the welfare of the patient. In choosing the best option, issues such as pain and suffering, as well as prognosis, and anticipated success of procedures must be considered. The principle of best interest does not necessarily recognize or conform to any desires that the patient may have expressed. If legal guardianship is sought, the principle of best interest is usually the standard used to guide the guardian’s decisions for the patient.
Decisions About Individual Treatments
Although many people think of end-of-life decision making as one grand, sweeping decision, the reality is that there are many separate decisions about individual treatments that must be made. Making decisions about individual interventions for the critically ill older adult is especially difficult as the results of one treatment choice may lead to complications and/or side effects that will prompt additional choices. Individual treatments that may be offered include the use of antibiotics for pneumonia, nutritional support such as tube feedings or parenteral nutrition, hydration support through intravenous fluids or hypodermoclysis, surgical procedures, use of medications for blood pressure support, use of pacemakers or internal cardiac defibrillators, and ventilatory support. Even family members who are well prepared to act as a health care decision agent may be confused and unsure of what the patient would choose in these complex situations.
Pneumonia in the older adult is a common reason for the use of antibiotics. Pneumonia is a frequent cause of hospital admission and death in older adults, and the risk of incidence and mortality increases with age (Kaplan et al., 2002; Loeb, McGeer, McArthur, Walter, & Simor, 1999). The symptom pattern and expression of pneumonia differs from that of younger patients, with delirium as a common sign in the older population. Aspiration is the most common cause (Marik & Kaplan, 2003) and aspiration pneumonia is often underdiagnosed (Marrie, 2000). Treatment decisions for pneumonia should not be based on age alone. For community-acquired pneumonia the Infectious Diseases Society of America and the American Thoracic Society (Mandell et al., 2007) recommended that decisions about whether to give antibiotics should be based on the assessment of the severity of the illness and the goals of care. For older adults who have no other life-limiting illness, there is no reason to withhold antibiotics. But, for the patient with late-stage Alzheimer’s disease, treatment for pneumonia might only serve to prolong the dying process, and it is reasonable to discuss this with the family.
Nutrition and hydration support are issues that have emotional overtones for most families. Administration of intravenous solutions is relatively easy for the short term, but must be weighed against the physiological burden of increased fluid intake for older adults whose systems may not be able to adequately process extra fluids. Decisions of hydration should be based on thorough assessment of the patient and clarification of the goals of hydration therapy.
For patients who have difficulty swallowing or eating, whether because of disease or loss of consciousness, long-term nourishment requires assistance. For those with an intact gastrointestinal tract, tube feedings are the preferred method. The PEG tube is the most commonly used means of providing tube feedings. Because the mortality 1 year after PEG placement is approximately 60% (McMahon, Hurley, Kamath, & Mueller, 2005), there should be care consideration of the patient’s overall condition prior to placement. A systematic nutritional assessment in older patients admitted to the ICU and treated by mechanical ventilation is important (Dardaine, Dequin, Ripault, Constans, & Ginies, 2001). Please refer to chapter 14, “Nutrition and Hydration” for more information.
At times, decisions need to be made about surgery options. Surgeries that were once limited to younger patients, such as coronary artery bypass grafts, are now being performed on older patients with good outcomes; however, there are a number of considerations to keep in mind. The effects of aging in the patient should be assessed and taken into account in determining surgical risk. Older adult patients frequently are using their reserve to maintain equilibrium, and when additional stress is placed on the patient, the reserves are insufficient to survive. Although good data regarding optimal perioperative management of the older adult are presently lacking, awareness of the areas of potential vulnerability allows treatment to be designed with these limitations in mind.
Decision making about cardiac-assist devices includes whether to initiate, and whether or when to discontinue such devices. Cardiac assist devices (CADs) include pacemakers, implantable cardiac defibrillators (ICDs), left ventricular assist devices (LVADs), and artificial hearts. CADs are being offered to more and more older adults (Goldstein & Lynn, 2006) and Medicare is reimbursing more often for devices such as ICDs. When considering placement of cardiac-assist devices, especially devices such as ICDs, which may affect comfort, there needs to be frank discussion with the patient about the benefits and burdens of such treatment. Devices should only be implanted in appropriate older adults, as outlined by the 2006 ACC/AHA/ESC guidelines (i.e., they should not be implanted in patients with a prognosis of less than 1 year) (Zipes et al., 2006). Discussion about the possibility of discontinuing a CAD as the patient nears the end of life should be initiated as part of the preplacement counseling (Stevenson & Desai, 2006).
Research indicates that clinicians are rarely having discussions with patient or family about deactivating ICDs at the end-of-life (Goldstein, Lampert, Bradley, Lynn, & Krumholz, 2004). When devices are deactivated, the deactivation is usually in response to distressing actions of the ICD in the hours before the death (Goldstein et al., 2004). Conversations about deactivating ICDs may be awkward, especially because many patients and family members think of the ICD as a means of preventing death. With prognosis in cardiac patients being very difficult to determine, even when death is near, the opportunity to discuss deactivation of CADs may pass before the need to have the conversation becomes obvious (Goldstein et al.).
The question of deactivating a CAD should be raised when a diagnosis of an unrelated life-limiting condition such as a malignancy or sepsis occurs, or with progression of the cardiac disease. When the person is dying, the actions of some CADs, such as the ICD, may no longer be desirable. As the patient’s condition deteriorates and there is more cardiac instability, the ICD may fire more often, resulting in painful shocks and anxiety (Eckert & Jones, 2002; Glikson & Friedman, 2001). According to Goldstein and Lynn (2006, p. 15), “the dying that is in store with the device—a struggle to breathe or cognitive failure, for example—is much worse that the possible sudden death without it.”
Withdrawal of pacemakers and ICDs is neither painful nor difficult. In most cases they can be reprogrammed noninvasively (Mueller, Hook, & Hayes, 2003). People do not usually die shortly after deactivating these devices, and the cause of death may be unrelated to the cardiac condition treated by the device (Goldstein & Lynn, 2006).
With LVADs, artificial hearts, or other devices, there can be complications such as bleeding, infections, and cerebrovascular accidents that are undesirable (Goldstein & Lynn, 2006). With these devices, death may occur immediately after deactivation. In these cases it is important to reinforce for the patient and the family that deactivating the device allows for a natural death.
Another treatment decision relates to the use of ventilatory support. Older adults are at an increased risk of developing respiratory failure in response to an acute illness. Age-related changes to the chest wall and lungs, a reduced ventilatory response to hypercarbia, and a reduced response to hypoxemia during REM (rapid eye movement) sleep contribute to the increased risk of respiratory failure. About half of ICU admissions have a major respiratory component (El Solh & Ramadan, 2006). The current recommendation in older adults is that ventilatory support should be used for only potentially reversible situations.
Recent research indicates that for patients who have relatively good short-term prognoses, ventilator support and aggressive care are economically worthwhile, even for patients 75 years and older (Hamel et al., 2001). In one study, survival of older ICU patients (n = 116) after mechanical ventilation, at discharge, and at 6 months was studied in one ICU. Mortality in the ICU and 6 months after discharge was 31% and 52%, respectively. Most of the patients (91%) who survived for 6 months were able to return to their place of residence, and 89% had similar or improved functional status compared with their preadmission functional status (Dardaine et al., 2001).
The use of CPR in older adults is thought to be of little benefit (Edin, 2007) despite the fact that it is regularly offered. There are racial differences in preference for cardiopulmonary resuscitation with Black Americans more frequently choosing CPR (Borum, Lynn, & Zhong, 2000). A decision to forgo CPR, a Do Not Resuscitate (DNR) order, has implications beyond just the avoidance of CPR. How the issue of a DNR order is raised may lead families to believe that they are choosing certain death as compared to life.
Patients and families have concerns that care may be compromised when a DNR decision has been made. There is some evidence that might support this fear. One study of residents and attending physicians showed that, in hypothetical situations, physicians chose to initiate fewer interventions unrelated to CPR when a DNR order was present versus absent (Beach & Morrison, 2002). In another study, nurses had lower levels of agreement about whether they would use monitoring interventions (vital signs, weighing the patient) when the hypothetical patient had a DNR than when he/she did not (Sherman & Branum, 1995).
Recent guidelines from the Society of Critical Care Medicine stated that when comfort is the goal for a patient in critical care, each intervention, such as monitors or tests, should be evaluated in light of whether it increases comfort (Truog et al., 2008). Some treatments that are routine in the ICU such as routine laboratory tests, chest X-rays, daily weights, and endotracheal suctioning may not contribute to the patient’s comfort and should not be continued. Other treatments, such as intravenous administration of vasopressors or inotropes, may be consistent with a goal of comfort because they cause little pain, yet they increase perfusion of vital organs, which may increase the level of consciousness, hepatic and renal functions, and therefore decrease distressing symptoms (Truog et al.).
Discussions about care decision should take place with the patient and the family and should be held shortly after admission to the ICU (Lilly et al., 2000). This family meeting allows for an early discussion of the patients’ goals and expectations and the plan of care. Good communication early in the ICU stay can lead to the appropriate use of advanced life-support technology for patients with a high potential for survival and may allow the earlier withdrawal of life-support measures when they are ineffective (Lilly, Sonna, Haley, & Massaro, 2003).
Limiting Treatment
When the burdens of therapy substantially outweigh the benefits, decisions concerning continuation of life-sustaining treatments need to be reevaluated. Although ethically, withholding and withdrawing treatment are considered equivalent (Nuffield Council on Bioethics, 2006; Pawlik, 2006), the emotional impact of the two choices can be very different. Families may find it less distressing to “choose comfort” as opposed to “stopping care.” In either situation, families commonly express guilt and anger. It can be helpful to remind the family that neither they nor the health care team are responsible for the patient’s poor prognosis (Henig, Faul, & Raffin, 2001), but they all share the goal of determining the best course of treatment for the patient.
Because families may find it less distressing to withhold rather than withdraw life-sustaining treatments, it is critical to carefully consider decisions to initiate lifesustaining treatments such as antibiotics, tube feedings, vasopressors or inotropes, and ventilatory support. Often in the critical care setting decisions to initiate such treatments need to be made rapidly. With older adults rapid decision making may be difficult. Older adults are more likely to have older surrogate decision makers who may be used to deferring to the decision of the physician or who may desire a longer time to contemplate the consequences of decisions. For other older patients, decisions may be made by adult children or other relatives who want to seek the advice and agreement of additional family members before committing to a course of action. One way to reduce the time pressure in such situations is to “bring the family along” through the entire course of treatment. Letting family members, particularly family decision makers, know what the health care team is thinking in terms of what symptoms they are seeing and what those symptoms likely mean in terms of survival is important. One helpful technique is to identify clear milestones that indicate improvement or decline such as those shown in Table 8.1 (Henig et al., 2001). Families interviewed after a family member died in an ICU often expressed that they had no idea how serious things were until they were asked to make a decision regarding lifesustaining treatment (Kehl & Kirchhoff, 2004).
Withdrawal of Life-Sustaining Treatment
About Life Sustaining Treatments Signs of Improvement or Decline for Families to Watch for in Elderly Patients in Critical Care
About Life Sustaining Treatments Signs of Improvement or Decline for Families to Watch for in Elderly Patients in Critical Care

When decisions are being made about whether to withdraw life-sustaining treatment, the patient is seldom able to participate. Families make these decisions with support from the health professionals (Prendergast & Puntillo, 2002). Timely information about prognosis coupled with support is helpful to the families. Ideally, a family meeting should be held that will include family members, physicians, nurses, a social worker, a chaplain if desired by family, and other consultants such as palliative care team members or ethics consultants. This meeting should provide the family with information, as well as give them an opportunity to ask questions and express emotions (Curtis et al., 2002). To be most effective the clinicians should have timely information that can be conveyed in an honest and clear manner to the family. The clinicians also need to listen to what the family says and and respond to their needs (Norton, Tilden, Tolle, Nelson, & Eggman, 2003).
Life-support devices are intended to provide temporary support for patients with potentially reversible conditions and not intended to prevent death from occurring (Wood & Marik, 2004). When decisions need to be made about withholding or withdrawing life-sustaining measures, the presence of preexisting disease is a key consideration (Plaisier, Blostein, Hurt, & Malangoni, 2002). Older adults often have other conditions or comorbidities that make the use of life-sustaining therapies less advisable. In addition, elders are vulnerable to severe injury and have limited physiologic response to stress (Chang & Schecter, 2007). Withholding usual treatments in the presence of multiple comorbidities may be analogous to allowing natural death to occur.
Some recommendations have been suggested about how to best provide for a comfortable process when withdrawing ventilatory support. Campbell (2007) completed a systematic review of the research about how to withdraw mechanical ventilation. If the patient is brain dead, sedation or analgesia before or during withdrawal is not indicated. If premedication is needed, patient behaviors should guide the initiation and escalation of sedatives and opioids. Doses of opioids used to manage dyspnea are similar to those used during ventilator withdrawal.
Selection of a method for weaning the patient from the ventilator seems to be dependent on clinician preference rather than on patient characteristics (Campbell, 2007). Sometimes the endotracheal tube is removed immediately (extubation); sometimes the tube is removed after ceasing ventilatory support (terminal extubation). Terminal weaning is the process during which there is a gradual reduction of ventilatory support, with the tube removed at the end of the withdrawal. In general the endotracheal tube should be removed as early in the process as possible while taking into consideration the possibility of airway compromise. There are no specific recommendations for weaning from the ventilator based on age.
For older adults in the critical care setting, there are three possible outcomes: They may recover and be discharged from the ICU, aggressive therapy may be continued until it is no longer effective and the patient dies, or a shift in goals of care from cure to comfort may be chosen and the patient dies. When it is apparent that curative care is no longer possible, the options are narrowed to continuing aggressive therapy or focusing on comfort care. In either situation, patients and families will require open and honest communication, physical, psychosocial, and spiritual support, and preparation for the patient’s death.
Transition to Palliative Care
Once a decision has been made to withhold or withdraw treatment, the focus of care shifts from life-saving treatment to holistic, then palliative care (Mularski & Osborne, 2003). Good palliative care in the ICU includes symptom management, attending to the patient’s spiritual and psychosocial needs, supporting the family, and preparing the patient and family for the death. The shift to palliative care begins with assessing what the patient and family believes is a good death. The concept of a good death is highly individual and dependent on cultural, spiritual, family, and personal values (Kehl, 2006). To optimally care for an older adult dying in the critical care setting, we must first understand the patient’s wishes and then work with the patient and family to design care to meet the dying person’s needs.
Palliative care for older adults in the ICU requires aggressive care to promote comfort. Tests or procedures that do not promote comfort, laboratory work, and painful procedures should be discontinued. The nurse must be perceptive and vigilant regarding changes in condition. Good communication with both the family and the rest of the health care team can reduce family questions and concerns, as well as keep them informed and involved. It is important to avoid offering “nonoptions” as choices to family. If the physician or care team has decided that a treatment is medically futile, it should not be offered as an option to family, nor should the benefits be discussed. This provides consistent information from all care team members and reduces family confusion. Questions about medical futility should be discussed with the physician, the care team, and if there are persistent conflicts, with an ethics team or whatever institutional guidelines specify.
Transition to palliative care may also mean a transition to another location of care such as a palliative care unit, a hospice facility, or even the patient’s home. Although such transitions may be necessitated by hospital policies, economic factors, and the family’s ability to provide care, they should be kept to a minimum and they should be initiated as early as possible. One recent study found that one of the most common regrets of family members was that they did not transition to hospice sooner (Kehl & Kirchhoff, 2004). Other researchers have found that transition to a different setting during the last days of life can be confusing and traumatic (Duggleby & Berry, 2005). These transitions can be complicated in older adults, especially if they have memory impairment, because they disrupt relationships the patient has come to rely on, which can increase the risk of complications (Flannery, 2002).
Symptom Management
Pain
Symptom management is critical in palliative care, and pain is often considered the most distressing symptom. The SUPPORT study described up to 40% of older adults as having unrelieved or poorly relieved pain in their final days (SUPPORT Investigators, 1995). Managing pain in older adults requires excellent assessment of the pain and its sources. Often older adults experience both long-term or chronic pain from conditions such as arthritis or venous insufficiency, and an acute pain from coronary disease, cancer, or other acute illnesses. Both must be managed (Stein, 2006). This may require use of multiple agents such as NSAIDs (nonsteroidal antiinflammatory drugs) for arthritic pain, opioids for acute pain, and antidepressants for neuropathic pain.
There are a number of challenges to providing pain management to older adults in their final days. Usually the patient cannot communicate verbally and assessment of pain must use other mechanisms. There are a number of good nonverbal pain assessment tools. One example is the PASLAC (pain assessment checklist for seniors with limited ability to communicate), which was developed as an observational pain assessment tool for seniors (Fuchs-Lacelle & Hadjistavropoulos, 2004). PASLAC is considered one of the best nonverbal pain assessment instruments (Zwakhalen, Hamers, Abu-Saad, & Berger, 2006) because it discriminates between pain events and nonspecific distress (Fuchs-Lacelle & Hadjistavropoulos, 2004). A brief 24-item version is useful in clinical settings (Zwakhalen, Hamers, & Berger, 2007) and could be easily adapted to critical care.
Families often have many questions regarding pain management. These include whether pain medication is needed when the patient is unresponsive, if giving pain medication will hasten death, and if all pain can be controlled. Generally, if the patient is unresponsive, has no signs of pain, and is not on a medication that would mask signs of pain (such as pancuronium), a trial of reducing the medication can be attempted. Medications should be reduced by = 50% in 24 hours (Portenoy, 1994). If the patient shows any signs of pain, such as those on the PAINAD (pain assessment in advanced dementia (Warden, Hurley, & Volicer, 2003), the medication should be returned to the previous level. If there are no signs of pain, medications may be reduced by 25% every 8 hours with careful assessment.
There is no evidence that use of opioid pain medication hastens death in most situations. Aclassic study by Bruera, Macmillan, Pither, and MacDonald (1990) showed that patients who received morphine for end-stage symptom management had less dyspnea and less pain, but no significant changes in respiratory rate, pO2, Pco2
. Other research (Campbell, Bizek, & Thill, 1999; Chan et al., 2004) has reported no association between analgesia or sedation and time to death following withdrawing ventilatory support.
Evidence of how older adults respond to dosing of opioids is scarce or inconclusive (Masica et al., 2007). Aging results in an increased fat-to-lean body-mass ratio as well as decreased renal and hepatic functioning, which leads to slower metabolism of medications and increased risk for the build up of harmful metabolites (Stein, 2006). Older people may be more sensitive to analgesic effect of opioids with higher peak effect and longer duration of action (Kaiko, Wallenstein, Rogers, Grabinski, & Houde, 1982). Drugs such as meperidine and propoxyphene should be avoided, as should opioids with longer duration of action such as levorphanol, methadone, and sustainedrelease fentanyl. The latter should only be used in opioid-tolerant patients, and only if the patient does not tolerate other opioids such as morphine. Sedation and nausea are common in the first 24–72 hours with opioids, and the patient and family should be warned this is a common occurrence and not a sign of drug allergy.
Another risk in pain management with older adults is managing the accompanying constipation. Many older adults have underlying issues with constipation. When the constipating effects of opioids are added to this, the result can cause pain and complications such as bowel obstruction. Patients who are starting opioid therapy should be concurrently started on laxatives, in addition to whatever usual bowel management protocol they followed prior to initiating opioid therapy.
Pain in dying patients can be controlled, but sometimes the price is awareness. In cases where pain is unrelieved by usual management, palliative sedation may be an option. Palliative sedation is the use of sedative medications to make the patient unaware, usually caused by severe, intractable symptoms at the end of life (Cherny, 2006). If there is no treatment capable of providing adequate pain relief, a discussion of the risks and benefits of palliative sedation may be initiated by the health care provider. It is important that the patient and family understand that the purpose of palliative sedation is comfort, and that resuscitation is not desirable. Patients and family members should also understand that it is likely that the person will die without regaining consciousness, so good-byes must be said prior to initiating sedation. For further information concerning pain management in older adults, please refer to chapter 18, “Pain in the Critically Ill Older Adult.”
Dyspnea
Dyspnea is distressing to both the patient and the family; it occurs in up to 70% of dying persons, regardless of diagnosis (Reuben & Mor, 1986). For older adults who choose not to accept ventilatory support, or for those being withdrawn from ventilatory support, dyspnea may be a major concern. Dyspnea can be assessed and managed at the end of life, both with medications and nonpharmacological methods. The best way to assess dyspnea is self-report. For patients who cannot speak, behavioral clues such as restlessness, grunting at end-expiration, and a look of fear may indicate dyspnea (Campbell, 2004). After assessment, the first step in management is to remove or alleviate physical causes of dyspnea such as pulmonary edema, partial obstruction caused by mucus, or metabolic acidosis. Next, nonpharmacological methods can be employed. Positioning the patient upright maximizes the lung capacity and can decrease dyspnea. Positioning the patient in front of a fan provides cool airflow to the cheek (Schwartzstein, Lahive, Pope, Weinberger, & Weiss, 1987) and nose (Burgess & Whitelaw, 1988), which can reduce the perception of breathlessness.
The effectiveness of oxygen in relieving dyspnea in dying persons has not been clearly established (Bruera et al., 2003; Qaseem et al., 2008). Generally, if the patient is hypoxic and oxygen helps subjectively, it may be used. Some research has shown hypoxic cancer patients’ rating of dyspnea, respiratory rate, and respiratory efforts improve with oxygen use (Bruera, de Stoutz, Velasco-Leiva, Schoeller, & Hanson, 1993). Some patients may pull off the mask or cannula repeatedly and say the oxygen delivery system is suffocating them; for these patients, use of a fan is a better option.
Opioids, particularly morphine, have been used for more than a century to manage dyspnea. Low-dose morphine can decrease the sensation of breathlessness without decreasing the respiratory function of terminally ill patients (Bruera et al., 1990; Jennings, Davies, Higgins, & Broadley, 2001; Lorenz, Rosenfeld, & Wenger, 2007). Nebulized morphine has been tried, but there is no evidence that it is effective in treating dyspnea (Jennings et al.; Lorenz et al.). Sedatives have also been used in treating dyspnea, but the results have been mixed. Given the risk of paradoxical effects of longer acting sedatives, and the high risk that older adults will experience adverse effects (Senninger & Laxenaire, 1995), sedatives should only be used to reduce breathlessness if other alternatives have failed. Further research is needed before sedatives are widely used to reduce breathlessness at the end of life.
Audible secretions, often referred to as a “death rattle” are another troublesome symptom at the end of life. In patients on mechanical ventilation, audible secretions are often managed by increased suctioning and maintaining the endotracheal tube in place. For those who are not mechanically ventilated, or for patients who are extubated prior to death, this sound can be one of the most distressing issues for family, even though patients do not seem to be aware of it. Families need explanations that the noisy breathing does not indicate distress. The primary treatment is antisecretory agents, although there is currently no strong evidence of their effectiveness (Wee & Hillier, 2008). Hyoscyamine sulfate is the preferred medication (Back, Jenkins, Blower, & Beckhelling, 2001; Bennett et al., 2002; Wildiers & Menten, 2002). Hycoscyamine can be administered orally, buccally using oral drops, or parenterally. If audible secretions are a persistent issue, scopolamine transdermal patches can be used to dry secretions and reduce the noisy breathing. The patch lasts for approximately 3 days, but the onset is about 4 to 6 hours, so it is not a good choice if the patient’s condition is changing rapidly. Suctioning should be considered only if the secretions are distressing to the patient, proximal and accessible, and there is inadequate response to antisecretory agents because suctioning is irritating and can be traumatic to the patient.
Terminal Delirium/Restlessness
Terminal restlessness or delirium occurs in up to 88% of dying persons (Kehl, 2004). The condition has been described as “agitated delirium in a dying patient, frequently associated with impaired consciousness and multifocal myoclonus” (Burke, 1997, p. 39). Indicators of terminal delirium include frequent nonpurposeful motor activity (e.g., picking at bed sheets), the inability to concentrate or relax, disturbances in sleep–rest patterns, fluctuating levels of consciousness, cognitive failure and/or anxiety, and potential progression to agitation.
Management of terminal delirium begins with a thorough assessment. Issues such as pain, full bladder or bowels, and alcohol or tobacco withdrawal may cause restlessness and delirium. Other areas for assessment include existential distress and spiritual distress. Conditions that might contribute to restlessness should be resolved before pharmacologic measures are taken.
First-line pharmacological treatment for terminal restlessness or delirium is either haloperidol or chlorpromazine (see Table 8.2). Caution should be used in administering lorazepam because of the high risk of side effects in older adults. Further information on management of delirium can be found in chapter 26, “Delirium in Critical Illness.”
Nearing-death awareness (NDA) often appears similar to restlessness and delirium. NDA is an experience that occurs in many people who are close to death. It almost always includes seeing or speaking to deceased loved ones (Callanan & Kelley, 1992) and may include other visions. NDA differs from hallucinations in a number of ways. A person experiencing NDA does not find the experience disturbing. If the patient is frightened or disturbed by what he or she is hearing or seeing, medication such as haloperidol, which is appropriate for hallucinations, should be given. NDA generally does not respond to antipsychotic medications. In some cases, the dying person may have a message for the family that is conveyed as part of the nearingdeath experience. This message is often very symbolic. Much nearing-death communication is conveyed in travel terms or in ways that are appropriate to the lifestyle of the patient. Patients or family members may be unsure of how to respond to NDA. Calm reassurance that these are normal experiences for someone who is nearing death can greatly comfort both the patient and the family.
Skin Care
There are numerous aspects of the dying process that affect skin care in older adults. Older adults have lost much of the elasticity of their skin. Skin is often friable and needs protection from pressure and shear. Near the end of life, decreased nutritional and fluid intake are common. Decreased mobility and changes in the circulatory system also affect the skin, which makes decubiti a common occurrence.
The primary goals of wound care in older adults at the end of life are maintaining optimal skin integrity and minimizing discomfort (Schim & Cullen, 2005). These goals may not both be achievable in the final days and patient comfort should always take top priority. Good skin hygiene can assist in meeting both goals, as well as to reassure both the patient and the family that the patient is receiving good nursing care. Daily cleansing and use of appropriate moisturizers can increase comfort and prevent painful infection (De Conno, Ventafridda, & Saita, 1991). The advantages of turning the patient to decrease further pressure sores should be carefully weighed against any discomfort caused by turning and the patient’s wishes (Langemo, 2006). If the patient has significant skin breakdown and terminal skin failure, use of a low-pressure system is recommended to eliminate the need to turn the patient frequently and to decrease the pressure and pain on the skin surface.
The Kennedy terminal ulcer is a pressure sore that is unique to patients who are nearing death. It usually appears on the sacrum and is pear-shaped with irregular borders (Kennedy, 1989). Initially it often appears as if the skin is bruised or scraped. Tissue deterioration with the Kennedy terminal ulcer is rapid, and an ulcer that is barely noticeable may have progressed to stage III or IV within 8 hours. Because this type of ulcer usually occurs within 48 hours of the patient’s death, it is often left untreated other than covering the ulcer to prevent infection and managing any discomfort associated with it.

Recommended Pharmacological Treatment of Terminal Restlessness
Recommended Pharmacological Treatment of Terminal Restlessness

Preparation for Death
Although there is widespread agreement that patients and families need to be prepared for the death, there is little empiric evidence of how to do so. When the patient is an older adult, the family may include a spouse who is also older, siblings, adult children, and grandchildren. It is difficult to find materials to assist in preparing such a wide range of ages and education levels for death. The best starting point is to assess what they know, what they expect will happen, and what they want to know. Although many patients and families want detailed information about what the dying person will experience, others may prefer limited information.
In preparing patients, it is important to assess which issues would make their death “good” or “bad” (Kehl, 2006). Knowing the patient’s idea of a good death provides guidance both for future decision making and for planning care surrounding the death. For patients, offer to talk about the signs and symptoms that are likely  for them and that would indicate imminent dying. It is very important to dispel myths regarding dying. The patient should be reassured that they should NOT expect uncontrolled pain, dyspnea, or loss of control. Assess the patient’s wishes concerning palliative sedation for these symptoms if other management is not effective.
Preparing the family for death is as important as preparing the patient. Unprepared family members are more likely to have complicated grief or depression (Barry, Kasl, & Prigerson, 2002; Houts, Lipton, Harold, Simmons, & Barthlolomew, 1989). If the patient is having life-support measures withdrawn, preparation should begin at the family meeting when this decision is made. A pilot study testing a theoretically developed informational intervention showed some promise for helping families know what to expect at the time and possibly reducing later negative moods at 2 to 4 weeks (Kirchhoff, Palzkill, Kowalkowski, Mork, & Gretarsdottir, 2008). The intervention is based on Johnson’s self-regulation theory (Johnson, 1999) and includes information on what the family members are likely to experience; the reasons or causes for what they will experience; the time frame and usual sequence of events; the environment, equipment, and personnel present; and information on what they can do or say during the withdrawal experience.
Although many nurses give information on signs and symptoms the patient may experience (Kirchhoff, Beckstrand, &Anumandla, 2003), families also need information on what they will see and hear and the timeframe to expect. When families know what to expect they can make a more informed decision about whether to be present during the withdrawal process. Clinical experience indicates that although some families may want this preparatory information before the withdrawal begins, others may need the information to be given as the withdrawal unfolds. Families need to know that health care professionals will aggressively treat symptoms and manage the patient’s comfort.
Telling families common signs and symptoms helps them know what to expect. Common signs and symptoms in the final hours include dyspnea, pain, restlessness or delirium, fatigue, breathing-pattern changes, audible secretions, cold extremities, and mottling or cyanosis. Some of these signs or symptoms, like breathing-pattern change, do not require intervention if the goal of care is comfort, but families should be prepared for what they will see and be reassured that the breathing-pattern changes such as apnea are normal and not distressing to the patient. Families also need to be told what to do. They often need permission to touch or speak to the patient at this time. Although not age related, this appears to be a common issue. Current research by Kehl is exploring the best ways to prepare families for the final hours (Kehl, 2008).
Some patients will undergo an unsuccessful resuscitation attempt prior to death. Institutional policy will dictate whether family may be present during resuscitation efforts. If family is not allowed to be present, a family support person should offer to stay with the family. Another individual should be designated to bring information from the health care team to the family. Families are very frightened during such events and need both support and information. Families of older adults should be prepared for the possibility that resuscitation may not be successful, especially if the patient has multiple comorbidities or is frail. After the patient has died and the room has been quickly straightened, the family should be invited to spend time with the patient.
Care After Death
Organ Donation
Although there are no upper age limits on organ donation (Health Resources and Services Administration, 2008) there are often conditions that make the organs of older adults less desirable or ineligible for organ donation (Carter et al., 2000). Under the 2006 Universal Anatomical Gift Act (UAGA), organs of those who have indicated prior to death that they want to donate organs (such as on a driver’s license) must be considered for donation. If the individual signed a refusal for donation prior, then family members cannot override that refusal and should not be asked about donation. If there is no indication of the patient’s wishes, then it may be necessary to approach the family. If there is a discrepancy between the patient’s previous written wish to make an anatomical gift and an advance directive that requests withholding or withdrawing of life-sustaining measures, the health professionals may first determine the medical suitability of a patient’s organs. This way, if an older adult is not considered a suitable donor, questions about donation or a conflict between the desire to donate and desire for a natural death do not need to be raised (National Conference of Commissioners on Uniform State Laws, 2008).
Family Care After the Death
After the death, the patient’s family requires care and support. Family members who are present may need to hear it said that the patient has died. If the family is not present, they should be notified that the patient has died and offered the opportunity to come to the unit to see the patient.
Deceased patients who are organ donors may need to be taken to the operating room immediately after the heart stops to facilitate a controlled donation. This event can be especially hard on families, particularly if they are not present at the time of death. These family members may wish to visit the unit or view the room to assist them in accepting the patient’s death.
Families should be given as much control as possible after the death. They should be asked whether they would like a support person (social worker, chaplain, or personal spiritual leader) or would like any personal, family, cultural, or spiritual practices observed. If the family is present, they should be invited to assist in cleaning up the patient. If they prefer not to participate, they should be asked if they would like to be present for the final cares, or if they would prefer to wait in a private room while the body is prepared.
Patients should receive final cares according to hospital policy and their preferences. Usually, if there is a possibility that an autopsy may be required or requested, all tubes and lines must be left in place. Obvious soiling should be removed and the patient should be covered modestly. If an autopsy is not needed and institutional policy permits, the family should be asked if they would prefer having the endotracheal tube, intravenous lines, oxygen tubing, and other devices removed.
In their good-byes some families will wish for private time, and others want a professional present. Even if they wish to be alone with the patient, staff should be available if they change their minds. Although institutional policy may limit the time family can spend with the deceased, it is important to provide as much time as the family needs.
After-death bereavement programs are present in some critical care units and range from presence at the time of death, to bereavement calls and cards, or bereavement support groups (Williams, Harris, Randall, Nichols, &Brown, 2003). Bereavement support groups and grief counselors are often available through local hospices if the hospital does not offer these services. Families sometimes contact the critical care unit after the death to express thanks or to seek assistance with their grieving.
Summary
Caring for older adults who are facing end-of-life issues in the critical care setting can be challenging. Skill in communication, facilitation of decision making by the patient and family, excellent assessment skills, good symptom management, attention to the psychosocial–spiritual needs of the patient and family, and preparation for the time of death are all key elements to good end-of-life care. Assessment of the patient’s wishes for end-of-life care is a necessary starting point. For patients who have not provided information on their preferences through an advance directive or other communication, it is important to work with family to determine the goals of care. It is especially important to keep family members aware of the current condition of the patient as things may change rapidly in the critical care setting.
Withholding or withdrawing life-sustaining therapies are valid options for patients whose primary goal is quality of life. In these situations as the goals of care shift from a curative focus to a comfort focus, families should be reassured that the patient will continue to receive excellent care. Aggressive symptom management can provide physical comfort in the final days.
Whether death follows a failed resuscitation attempt, withdrawal of life-sustaining therapy, or occurs after a deliberate choice for comfort care, families need to be supported and given as much control as possible. Critical care units may choose to provide bereavement support to families after the death or they may refer individuals to local support groups or bereavement counselors. Throughout the process, from the first indications that death may be the outcome of the critical care stay, until after the death, open and honest communication and skill at facilitating patient and family discussions can make the end-of-life experience easier for older adults and their families.
References
Back, I. N., Jenkins, K., Blower, A., & Beckhelling, J. (2001). A study comparing hyoscine hydrobromide and glycopyrrolate in the treatment of death rattle. Palliative Medicine, 15(4), 329–336.
Barry, L. C., Kasl, S. V., & Prigerson, H. G. (2002). Psychiatric disorders among bereaved persons: The role of perceived circumstances of death and preparedness for death. American Journal of Geriatric Psychiatry, 10(4), 447–457.
Beach, M. C., & Morrison, R. S. (2002). The effect of do-not-resuscitate orders on physician decisionmaking. Journal of the American Geriatrics Society, 50(12), 2057–2061.
Bennett, M., Lucas, V., Brennan, M., Hughes, A., O’Donnell, V., & Wee, B. (2002). Using anti-muscarinic drugs in the management of death rattle: Evidence-based guidelines for palliative care. Palliative Medicine, 16(5), 369–374.
Borum, M. L., Lynn, J., & Zhong, Z. (2000). The effects of patient race on outcomes in seriously ill patients in SUPPORT: An overview of economic impact, medical intervention, and end-of-life decisions. Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments. Journal of the American Geriatrics Society, 48(5 Suppl.), S194–198.
Briggs, L. A., Kirchhoff, K. T., Hammes, B. J., Song, M. K., & Colvin, E. R. (2004). Patient-centered advance care planning in special patient populations: A pilot study. Journal of Professional Nursing, 20(1), 47–58.
Brostrom, L., Johansson, M., & Nielsen, M. K. (2007). “What the patient would have decided”: A fundamental problem with the substituted judgment standard. Medicine, Health Care and Philosophy, 10(3), 265–278.
Bruera, E., de Stoutz, N., Velasco-Leiva, A., Schoeller, T., & Hanson, J. (1993). Effects of oxygen on dyspnoea in hypoxaemic terminal-cancer patients. Lancet, 342(8862), 13–14.
Bruera, E., Macmillan, K., Pither, J., & MacDonald, R. N. (1990). Effects of morphine on the dyspnea of terminal cancer patients. Journal of Pain and Symptom Management, 5(6), 341–344.
Bruera, E., Sweeney, C., Willey, J., Palmer, J. L., Strasser, F., Morice, R. C., et al. (2003). A randomized controlled trial of supplemental oxygen versus air in cancer patients with dyspnea. Palliative Medicine, 17(8), 659–663.
Burgess, K. R., & Whitelaw, W. A. (1988). Effects of nasal cold receptors on pattern of breathing. Journal of Applied Physiology, 64(1), 371–376.
Burke, A. L. (1997). Palliative care: An update on “terminal restlessness.” Medical Journal of Australia, 166(1), 39–42.
Callanan, C., & Kelley, P. (1992). Final gifts. New York: Poseidon Press. Campbell, M. L. (2004). Terminal dyspnea and respiratory distress. Critical Care Clinics, 20(3), viii–ix, 403–417.
Campbell, M. L. (2007). How to withdraw mechanical ventilation: A systematic review of the literature. AACN Advanced Critical Care, 18(4), 397–403.
Campbell, M. L., Bizek, K. S., & Thill, M. (1999). Patient responses during rapid terminal weaning from mechanical ventilation: A prospective study. Critical Care Medicine, 27(1), 73–77.
Carrese, J. A., Mullaney, J. L., Faden, R. R., & Finucane, T. E. (2002). Planning for death but not serious future illness: Qualitative study of housebound elderly patients. BMJ, 325(7356), 125.
Carter, J. T., Lee, C. M., Weinstein, R. J., Lu, A. D., Dafoe, D. C., & Alfrey, E. J. (2000). Evaluation of the older cadaveric kidney donor: The impact of donor hypertension and creatinine clearance on graft performance and survival. Transplantation, 70(5), 765–771.
Chan, J. D., Treece, P. D., Engelberg, R. A., Crowley, L., Rubenfeld, G. D., Steinberg, K. P., et al. (2004). Narcotic and benzodiazepine use after withdrawal of life support: Association with time to death? Chest, 126(1), 286–293.
Chang, T. T., & Schecter, W. P. (2007). Injury in the elderly and end-of-life decisions. Surgical Clinics of North America, 87(1), viii, 229–245.
Cherny, N. I. (2006). Palliative sedation. In E. Breura, I. J. Higginson, C. Ripamonti, & C. von Gunten (Eds.), Textbook of palliative medicine (pp. 976–988). London: Hodder Arnold.
Curtis, J. R., Engelberg, R. A., Wenrich, M. D., Nielsen, E. L., Shannon, S. E., Treece, P. D., et al. (2002). Studying communication about end-of-life care during the ICU family conference: Development of a framework. Journal of Critical Care, 17(3), 147–160.
Dardaine, V., Dequin, P. F., Ripault, H., Constans, T., & Ginies, G. (2001). Outcome of older patients requiring ventilatory support in intensive care: Impact of nutritional status. Journal of the American Geriatrics Society, 49(5), 564–570.
De Conno, F., Ventafridda, V., & Saita, L. (1991). Skin problems in advanced and terminal cancer patients. Journal of Pain and Symptom Management, 6(4), 247–256.
Ditto, P. H., Jacobson, J. A., Smucker, W. D., Danks, J. H., & Fagerlin, A. (2006). Context changes choices: A prospective study of the effects of hospitalization on life-sustaining treatment preferences. Medical Decision Making, 26(4), 313–322.
Duggleby, W., & Berry, P. (2005). Transitions and shifting goals of care for palliative patients and their families. Clinical Journal of Oncology Nursing, 9(4), 425–428.
Eckert, M., & Jones, T. (2002). How does an implantable cardioverter defibrillator (ICD) affect the lives of patients and their families? International Journal of Nursing Practice, 8(3), 152–157.
Edin, M. G. (2007). Cardiopulmonary resuscitation in the frail elderly: Clinical, ethical and halakhic issues. Israeli Medical Association Journal, 9(3), 177–179.
El Solh, A. A., & Ramadan, F. H. (2006). Overview of respiratory failure in older adults. Journal of Intensive Care Medicine, 21(6), 345–351.
Flannery, R. B., Jr. (2002). Disrupted caring attachments: Implications for long-term care. American Journal of Alzheimer’s Disease and Other Dementias, 17(4), 227–231.
Foley, K. M., & Gelband, H. (2001). Executive summary. In K. M. Foley & H. Gelband (Eds.), Improving palliative care for cancer (pp. 1–8). Washington, DC: National Academy Press.
Formiga, F., Chivite, D., Ortega, C., Casas, S., Ramon, J. M., & Pujol, R. (2004). End-of-life preferences in elderly patients admitted for heart failure. QJM: An International Journal of Medicine, 97(12), 803–808.
Fuchs-Lacelle, S., & Hadjistavropoulos, T. (2004). Development and preliminary validation of the pain assessment checklist for seniors with limited ability to communicate (PACSLAC). Pain Management Nursing, 5(1), 37–49.
Glikson, M., & Friedman, P. A. (2001). The implantable cardioverter defibrillator. Lancet, 357(9262), 1107–1117.
Goldstein, N. E., Lampert, R., Bradley, E., Lynn, J., & Krumholz, H. M. (2004). Management of implantable cardioverter defibrillators in end-of-life care. Annals of Internal Medicine, 141(11), 835–838.
Goldstein, N. E., & Lynn, J. (2006). Trajectory of end-stage heart failure: The influence of technology and implications for policy change. Perspectives in Biology and Medicine, 49(1), 10–18.
Goodman, M. D., Tarnoff, M., & Slotman, G. J. (1998). Effect of advance directives on the management of elderly critically ill patients. Critical Care Medicine, 26(4), 701–704.
Gordon, M. (2002). Ethical challenges in end-of-life therapies in the elderly. Drugs & Aging, 19(5), 321–329.
Hamel, M. B., Davis, R. B., Teno, J. M., Knaus, W. A., Lynn, J., Harrell, F., Jr., et al. (1999). Older age, aggressiveness of care, and survival for seriously ill, hospitalized adults. Annals of Internal Medicine, 131(10), 721–728.
Hamel, M. B., Lynn, J., Teno, J. M., Covinsky, K. E., Wu, A. W., Galanos, A., et al. (2000). Agerelated differences in care preferences, treatment decisions, and clinical outcomes of seriously ill hospitalized adults: Lessons from SUPPORT. Journal of the American Geriatrics Society, 48(5 Suppl.), S176–182.
Hamel, M. B., Phillips, R. S., Davis, R. B., Teno, J., Desbiens, N., Lynn, J., et al. (2001). Are aggressive treatment strategies less cost-effective for older patients? The case of ventilator support and aggressive care for patients with acute respiratory failure. Journal of the American Geriatrics Society, 49(4), 382–390.
Health Resources and Services Administration. (2008). Organ donation & transplantation questions. Retrieved February 20, 2008, from http://answers.hrsa.gov
Henderson, M. L. (2004). Gerontological advance practice nurses: As end-of-life care facilitators. Geriatric Nursing, 25(4), 233–237.
Henig, N. R., Faul, J. L., & Raffin, T. A. (2001). Biomedical ethics and the withdrawal of advanced life support. Annual Review of Medicine, 52, 79–92.
Houts, P., Lipton, A., Harold, A., Simmons, M., & Barthlolomew, M. (1989). Predictors of grief among spouses of deceased cancer patients. Journal of Psychosocial Oncology, 7, 113–126.
Jennings, A., Davies, A., Higgins, J., & Broadley, K. (2001). Opioids for the palliation of breathlessness in terminal illness. Cochrane Database Systematic Review, 4.
Jezewski, M. A., & Meeker, M. A. (2005). Constituting advance directives from the perspective of people with chronic illnesses. Journal of Hospice and Palliative Nursing, 7(6), 319–327.
Johnson, J. E. (1999). Self-regulation theory and coping with physical illness. Research in Nursing and Health, 22(6), 435–448.
Kaiko, R. F., Wallenstein, S. L., Rogers, A. G., Grabinski, P. Y., & Houde, R. W. (1982). Narcotics in the elderly. Medical Clinics of North America, 66(5), 1079–1089.
Kaplan, V., Angus, D. C., Griffin, M. F., Clermont, G., Scott Watson, R., & Linde-Zwirble, W. T. (2002). Hospitalized community-acquired pneumonia in the elderly: Age- and sex-related patterns of care and outcome in the United States. American Journal of Respiratory Critical Care Medicine, 165(6), 766–772.
Kaufman, S. R. (1998). Intensive care, old age, and the problem of death in America. The Gerontologist, 38(6), 715–725.
Kehl, K. A. (2004). Treatment of terminal restlessness: A review of the evidence. Journal of Pain & Palliative Care Pharmacotherapy, 18(1), 5–30.
Kehl, K. A. (2006). Moving toward peace: An analysis of the concept of a good death. American Journal of Hospice and Palliative Care., 23(4), 277–286.
Kehl, K. A. (2008). Caring for the patient and the family in the last hours of life. Home Health Care Management Practice, 20(5), 408–413.
Kehl, K. A., & Kirchhoff, K. T. (2004, October). Family Perceptions of Qualtiy of Care at the End of Life in ICU and Hospice Facility. Paper presented at the 2004 National State of the Science Conference, Washington, DC.
Kennedy, K. L. (1989). The prevalence of pressure ulcers in an intermediate care facility. Decubitus, 2(2), 44–45.
Kirchhoff, K. T. (2005). Getting our ICU language straight. Critical Connections, 4, 1, 11. Kirchhoff, K. T., Beckstrand, R. L., & Anumandla, P. R. (2003). Analysis of end-of-life content in critical care nursing textbooks. Journal of Professional Nursing, 19(6), 372–381.
Kirchhoff, K. T., Palzkill, J., Kowalkowski, J., Mork, A., &Gretarsdottir, E. (2008). Preparing ICU families for withdrawal of life support: A pilot study. American Journal of Critical Care, 17(2), 113–121.
Kleinpell, R. M., & Ferrans, C. E. (1998). Factors influencing intensive care unit survival for critically ill elderly patients. Heart & Lung, 27(5), 337–343.
Knauft, E., Nielsen, E. L., Engelberg, R. A., Patrick, D. L., & Curtis, J. R. (2005). Barriers and facilitators to end-of-life care communication for patients with COPD. Chest, 127(6), 2188–2196.
Krohm, C., & Summers, S. K. (2002). The ethics of advance health care directives. In Advance health care directives: A handbook for professionals (pp. 23–32). Chicago: American Bar Association.
Langemo, D. K. (2006). When the goal is palliative care. Advances in Skin & Wound Care, 19(3), 148–154.
Lee, M. A., Brummel-Smith, K., Meyer, J., Drew, N., & London, M. R. (2000). Physician orders for life-sustaining treatment (POLST): Outcomes in a PACE program. Program of All-Inclusive Care for the Elderly. Journal of the American Geriatrics Society, 48(10), 1219–1225.
Lilly, C. M., De Meo, D. L., Sonna, L. A., Haley, K. J., Massaro, A. F., Wallace, R. F., et al. (2000). An intensive communication intervention for the critically ill. American Journal of Medicine, 109(6), 469–475.
Lilly, C. M., Sonna, L. A., Haley, K. J., & Massaro, A. F. (2003). Intensive communication: Four-year follow-up from a clinical practice study. Critical Care Medicine, 31(5 Suppl.), S394–S399.
Lloyd, C. B., Nietert, P. J., & Silvestri, G. A. (2004). Intensive care decision making in the seriously ill and elderly. Critical Care Medicine, 32(3), 649–654.
Lo, B., & Steinbrook, R. (2004). Resuscitating advance directives. Archives of Internal Medicine, 164(14), 1501–1506.
Loeb, M., McGeer, A., McArthur, M., Walter, S., & Simor, A. E. (1999). Risk factors for pneumonia and other lower respiratory tract infections in elderly residents of long-term care facilities. Archives of Internal Medicine, 159(17), 2058–2064.
Lorenz, K. A., Rosenfeld, K., & Wenger, N. (2007). Quality indicators for palliative and end-of-life care in vulnerable elders. Journal of the American Geriatrics Society, 55(Suppl. 2), S318–S326.
Mandell, L. A., Wunderink, R. G., Anzueto, A., Bartlett, J. G., Campbell, D., Dean, N. C., et al. (2007). IDSA/ATS Guidelines for CAP in Adults. Clinical Infectious Diseases, 44(Suppl. 2), 27–72.
Marik, P. E., & Kaplan, D. (2003). Aspiration pneumonia and dysphagia in the elderly. Chest, 124(1), 328–336.
Marrie, T. J. (2000). Community-acquired pneumonia in the elderly. Clinical Infectious Diseases, 31(4), 1066–1078.
Masica, A. L., Girard, T. D., Wilkinson, G. R., Thomason, J. W. W., Truman Pun, B., Nair, U. B., et al. (2007). Clinical sedation scores as indicators of sedative and analgesic drug exposure in intensive care unit patients. American Journal of Geriatric Pharmacotherapy, 5(3), 218–231.
McMahon, M. M., Hurley, D. L., Kamath, P. S., & Mueller, P. S. (2005). Medical and ethical aspects of long-term enteral tube feeding. Mayo Clinic Proceedings, 80(11), 1461–1476.
Mueller, P. S., Hook, C. C., & Hayes, D. L. (2003). Ethical analysis of withdrawal of pacemaker or implantable cardioverter-defibrillator support at the end of life. Mayo Clinic Proceedings, 78(8), 959–963.
Mularski, R. A., & Osborne, M. L. (2003). End-of-life care in the critically ill geriatric population. Critical Care Clinics, 19(4), viii, 789–810.
Nagasawa, Y. (2008). Proxy consent and counterfactuals. Bioethics, 22(1), 16–24. National Conference of Commissiorners on Uniform State Laws. (2008). UAGA summary. Retrieved August 15, 2008, from http://www.anatomicalgiftact.org/DesktopDefault.aspx?tabindex=1&t abid=67
Norton, S. A., Tilden, V. P., Tolle, S. W., Nelson, C. A., & Eggman, S. T. (2003). Life support withdrawal: Communication and conflict. American Journal of Critical Care, 12(6), 548–555.
Nuffield Council on Bioethics. (2006). Critical care decisions in fetal and neonatal medicine: Ethical issues. Retrieved August 15, 2008, from http://www.nuffieldbioethics.org/go/browseablepublications/ critic alCareDecisionFetalNeonatalMedicine/report_510.html
Pawlik, T. M. (2006). Withholding and withdrawing life-sustaining treatment: A surgeon’s perspective. Journal of the American College of Surgeons, 202(6), 990–994.
Plaisier, B. R., Blostein, P. A., Hurt, K. J., & Malangoni, M. A. (2002). Withholding/withdrawal of life support in trauma patients: Is there an age bias? American Surgeon, 68(2), 159–162.
Pollack, S. (2000). A new approach to Advance Directives. Critical Care Medicine, 28(9), 3146–3148.
Portenoy, R. K. (1994). Management of common opioid side effects during long-term therapy of cancer pain. Annals of the Academy of Medicine, Singapore, 23(2), 160–170.
Prendergast, T. J., & Puntillo, K. A. (2002). Withdrawal of life support: Intensive caring at the end of life. Journal of the American Medical Association, 288(21), 2732–2740.
Puchalski, C. M., Zhong, Z., Jacobs, M. M., Fox, E., Lynn, J., Harrold, J., et al. (2000). Patients who want their family and physician to make resuscitation decisions for them: Observations from SUPPORT and HELP. Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatment. Hospitalized Elderly Longitudinal Project. Journal of the American Geriatrics Society, 48(5 Suppl.), S84–S90.
Qaseem, A., Snow, V., Shekelle, P., Casey, D. E., Jr., Cross, J. T., Jr., Owens, D. K., et al. (2008). Evidence based interventions to improve the palliative care of pain, dyspnea, and depression at the end of life: A clinical practice guideline from the American College of Physicians. Annals of Internal Medicine, 148(2), 141–146.
Ramsey, G., & Mitty, E. (2003). Advance directives: Protecting patient’s rights. Retrieved November 7, 2007, from http://guideline.gov
Reuben, D. B., & Mor, V. (1986). Dyspnea in terminally ill cancer patients. Chest, 89(2), 234–236. Rocker, G., Cook, D., Sjokvist, P., Weaver, B., Finfer, S., McDonald, E., et al. (2004). Clinician predictions
of intensive care unit mortality. Critical Care Medicine, 32(5), 1149–1154. Schim, S. M., & Cullen, B. (2005). Wound care at end of life. Nursing Clinics of North America, 40(2),
281–294. Schneiderman, L. J., Pearlman, R. A., Kaplan, R. M., Anderson, J. P., & Rosenberg, E. M. (1992).
Relationship of general advance directive instructions to specific life-sustaining treatment preferences in patients with serious illness. Archives of Internal Medicine, 152(10), 2114–2122.
Schwartzstein, R. M., Lahive, K., Pope, A., Weinberger, S. E., &Weiss, J. W. (1987). Cold facial stimulation reduces breathlessness induced in normal subjects. American Review of Respiratory Disease, 136(1), 58–61.
Senninger, J. L., & Laxenaire, M. (1995). Violent paradoxal reactions secondary to the use of benzodiazepines. Annales Medico-Psychologiques (Paris), 153(4), 278–281; discussion 281–282.
Shapiro, S. P. (2007). When life imitates art: Surrogate decision making at the end of life. T opics in Stroke Rehabilitation, 14(4), 80–92.
Sherman, D. A., & Branum, K. (1995). Critical care nurses’ perceptions of appropriate care of the patient with orders not to resuscitate. Heart and Lung: Journal of Critical Care, 24(4), 321–329.
Somogyi-Zalud, E., Zhong, Z., Hamel, M. B., & Lynn, J. (2002). The use of life-sustaining treatments in hospitalized persons aged 80 and older. Journal of the American Geriatrics Society, 50(5), 930–934.
Somogyi-Zalud, E., Zhong, Z., Lynn, J., & Hamel, M. B. (2000). Elderly persons’ last six months of life: Findings from the Hospitalized Elderly Longitudinal Project. Journal of the American Geriatrics Society, 48(5 Suppl.), S131–S139.
Stein, W. M. (2006). Pain in older people. In E. Bruera, I. J. Higginson, C. Ripamonti, & C. von Gunten (Eds.), Textbook of palliative medicine (pp. 467–481). London: Hodder Arnold.
Stevenson, L. W., & Desai, A. S. (2006). Selecting patients for discussion of the ICD as primary prevention for sudden death in heart failure. Journal Cardiac Failure, 12(6), 407–412.
SUPPORT Investigators. (1995). 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). Journal of the American Medical Association, 274(20), 1591–1598.
Teno, J. M., Fisher, E., Hamel, M. B., Wu, A. W., Murphy, D. J., Wenger, N. S., et al. (2000). Decision making and outcomes of prolonged ICU stays in seriously ill patients. Journal of the American Geriatrics Society, 48(5 Suppl.), S70–S74.
Teno, J. M., Harrell, F. E., Jr., Knaus, W., Phillips, R. S., Wu, A. W., Connors, A., Jr., et al. (2000). Prediction of survival for older hospitalized patients: The HELP survival model. Hospitalized Elderly Longitudinal Project. Journal of the American Geriatrics Society, 48(5 Suppl.), S16–S24.
Tonelli, M. R. (2005). Waking the dying: Must we always attempt to involve critically ill patients in end-of-life decisions? Chest, 127(2), 637–642.
Truog, R. D., Campbell, M. L., Curtis, J. R., Haas, C. E., Luce, J. M., Rubenfeld, G. D., et al. (2008). Recommendations for end-of-life care in the intensive care unit: A consensus statement by the American Academy of Critical Care Medicine. Critical Care Medicine, 36(3), 953–963.
Van Den Noortgate, N., Vogelaers, D., Afschrift, M., & Colardyn, F. (1999). Intensive care for very elderly patients: Outcome and risk factors for in-hospital mortality. Age & Ageing, 28(3), 253-–256.
Warden, V., Hurley, A. C., & Volicer, L. (2003). Development and psychometric evaluation of the Pain Assessment in Advanced Dementia (PAINAD) scale. Journal of the American Medical Directors Association, 4(1), 9–15.
Wee, B., & Hillier, R. (2008). Interventions for noisy breathing in patients near to death. Cochrane Database System Review, (1), CD005177.
Wildiers, H., & Menten, J. (2002). Death rattle: Prevalence, prevention and treatment. Journal of Pain and Symptom Management, 23(4), 310–317.
Williams, R., Harris, S., Randall, L., Nichols, R., & Brown, S. (2003). A bereavement after-care service for intensive care relatives and staff: The story so far. Nursing in Critical Care, 8(3), 109–115.
Winzelberg, G. S., Patrick, D. L., Rhodes, L. A., & Deyo, R. A. (2005). Opportunities and challenges to improving end-of-life care for seriously ill elderly patients: A qualitative study of generalist physicians. Journal of Palliative Medicine, 8(2), 291–299.
Wood, K. A., & Marik, P. E. (2004). ICU care at the end of life. Chest, 126(5), 1403–1406. Wu, P., Lorenz, K. A., & Chodosh, J. (2008). Advance care planning among the oldest old. Journal of
Palliative Medicine, 11(2), 152–157. Zipes, D. P., Camm, A. J., Borggrefe, M., Buxton, A. E., Chaitman, B., Fromer, M., et al. (2006). ACC/
AHA/ESC 2006 Guidelines for management of patients with ventricular arrhythmias and the prevention of sudden cardiac death. Circulation, 114(10), e385–e484.
Zwakhalen, S. M., Hamers, J. P., Abu-Saad, H. H., & Berger, M. P. (2006). Pain in elderly people with severe dementia: A systematic review of behavioural pain assessment tools. BMC Geriatrics, 6,3.
Zwakhalen, S. M., Hamers, J. P., &Berger, M. P. (2007). Improving the clinical usefulness of a behavioural pain scale for older people with dementia. Journal of Advances in Nursing, 58(5), 493–502.


EmoticonEmoticon