Artificial Nutrition and Hydration at the End of Life

Abstract

Considerable controversy surrounds the issue of care at the end of life (EOL) for older adults. Technological advances and the legal, ethical, clinical, religious, cultural, personal, and fiscal considerations in the provision of artificial hydration and nutrition support to older adults near death are presented in this comprehensive review.

KEYWORDS:

• artificial hydration
• end of life
• nutrition support
• older adult

INTRODUCTION

The burgeoning older adult population will result in a dramatic increase in the oldest old in the United States within the next few decades. Many in this segment will be racial, ethnic, and religious minorities, whose end of life (EOL) wishes have not been legally documented and whose personal, cultural, and religious observances regarding terminal illness, EOL care, and dying may not be well understood by the health care community (1). Many will be underinsured or uninsured, will be admitted to the health care system through emergencies, and will not be able or competent to voice their desires regarding how they wish to be cared for. The costs associated with this lack of clarity regarding EOL treatment are astronomical from a personal, religious, social, ethical, and clinical perspective (2).

Numerous concerns have arisen regarding care of the older adult at the end of life. Due to advances in technology and the ability to provide artificial nutrition and hydration (ANH), research into the legal, ethical, clinical, religious, cultural, personal, and fiscal aspects have been conducted (3). Often, family or friends (surrogates) become decision makers and the burden falls to them, not to the dying older adult (4). This is difficult when the older adult is not competent to voice their wishes and no legal documentation exists delineating their preferences (5).

The complexity and emotion attached to EOL care makes the process difficult for the individual as well as family, friends, health care providers, and society. Costs associated with continued provision of ANH and life support and taxpayer burden with increased Medicare and Medicaid utilization results in further controversy. Medical paradigms dictate saving lives at all costs, technological advances make extension of life possible, and societal influences result in discomfort in allowing death to occur without medical and technological intervention (6).

End of life approaches are associated with concepts such as hospice care, palliative care, intensive medical care, and combinations thereof. An individual's wishes regarding their EOL care should be delineated in a legal document where artificial feeding, hydration, mechanical ventilation, or other support is described (7). This documentation includes items of advance medical directive such as a conventional “Last Will and Testament,” Durable Power of Attorney for Health Care (Allowable Health Care Proxy), Living Will, “Do Not Resuscitate” order, or other documentation (8). Legal issues regarding what constitutes “end of life” or “death” in a highly technological environment, where the body can be maintained using life support, must also be considered (9). Table 1 provides definitions of terms and concepts, along with a brief overview of the context and philosophy associated with them.

TABLE 1 Description of Care Concepts and Legal Terms Related to End of Life Services† ‡

Service concept†	Description
Hospice care	Hospice care is a multidisciplinary approach to EOL care when there is a definitive terminal course of illness, usually defined as a six-month period before death. A network of professionals and volunteers, often in the home, provides support services that address all the needs of the dying person and their family. The goal is exceptional quality of life in the remaining time surrounded by family and friends in a familiar environment. Palliative services such as aggressive pain control are employed, but treatment of the underlying pathology is not attempted. Curative measures are not undertaken but supportive measures are. These include, but are not limited to, psychotherapy, social services, religious support, familial support and counseling, grief/anger management, physical/occupational therapy, dietetics, and provision of medical equipment and ancillary services.
Palliative care	Palliative care may be provided at any time during a person's illness. The focus is on symptom management during assessment and treatment of the underlying pathophysiology. It is not the same as hospice care, although both seek to decrease pain and suffering, often by using aggressive pain treatments and a holistic approach. The intent is to provide the best possible quality of life. It is a branch of internal medicine that is administered either in the health care setting or in the home. It is usually done using a team approach, involving palliative care physicians; nurses; dietitians; social workers; clergy; and physical, occupational, and psycho-therapists. Many larger institutions have palliative care teams or units affiliated with their intensive medical care division.
Intensive medical care	Critical care medical units perform intensive medical care. It is a subspecialty that is the most costly, most resource intensive branch of medicine. Life supportive measures, constant monitoring, technologically advanced treatments, and equipment are used to prolong life in a patient with potentially reversible critical illness, organ-system failure, traumatic injury, or in unstable surgical patients. ANH is usually part of this care.
Home health	Home health care is provision of medical services to an ill person in his or her home by a licensed home health agency using skilled medical professionals and is billed to the individual's insurance. It is overseen by the patient's primary medical provider and includes services deemed appropriate by a medical care plan. Services may include but are not limited to the provision of medical equipment, home aides for activities of daily living, nursing, dietetics, speech, occupational and physical therapy, as well as social work and, in some instances, hospice. Curative treatments, life-prolonging measures and other interventions are used in the person's place of residence. It is usually time limited, and insurance coverage or the person's financial resources determine time for provision of service.
Advance medical directive †
Living will	A legal document that can be drawn up and signed, stating what medical and other life-sustaining procedures may or may not be done in the event that a person becomes mentally incompetent, comatose, otherwise incapacitated, or terminally ill. Such documentation should be readily available in the event of an emergent situation and on file with the primary care provider and immediate family members. Included in this documentation should be decisions regarding artificial nutrition and hydration as well as other treatment modalities.
Power of attorney	A Durable Power of Attorney for Health Care Decisions is documentation that identifies the person who can make and carry out an individual's wishes in the event that there is an emergent situation, the individual is incapacitated mentally or physically, or comatose. This responsibility is also known as Health Care Proxy. General Durable Power of Attorney is a legal document that designates another person to act on the individual's behalf concerning legal or business matters. They can write checks, access bank accounts, etc. while the individual is incapacitated.
Health care proxy	The right of a legally named person to execute the wishes for treatment or refusal of treatment when an individual is incapacitated.
Last Will and Testament	Also known as a “Will,” this document is a legal set of directions for what should be done after a person dies. There is an executor for the will, a person determined a priori who is responsible for dealing with the dead person's estate, assets, or other interests. This is not a replacement for a “Living Will,” which can be executed while the person is still alive, but incompetent.
Other considerations ‡
Resuscitation	To restore to life or consciousness, involving (usually) cardiopulmonary efforts, such as mouth-to-mouth or cardiac massage. Do Not Resuscitate or DNR orders are medically and legally binding.
Pain management	Management of pain and suffering requires an interdisciplinary approach that uses several branches of medicine and allied health. This may include pharmacotherapy, biofeedback, physical and occupational therapies, and psychotherapeutic approaches. Pain medicine employs several different kinds of analgesic medications, including opioids. Opioid medication in EOL is common in palliative care, but is controversial due to effects on multiple organ systems and fears regarding their addictive potential. Decrements in respiratory competence and swallowing reflexes as well as decreases in gastrointestinal motility and hydration status is common, even with low dose opioid administration.
Persistent vegetative state	A state of extreme unresponsiveness, lasting for more than one month, with no awareness or higher cerebral function. Midbrain and brainstem functions are preserved, with sleep-wake cycling and autonomic control over functions such as breathing and heart rate. There may be some reflexive movement. End-stage dementia often leads to permanent vegetative state.
Permanent vegetative state	After approximately one year of persistent vegetative state, the diagnosis becomes “permanent vegetative state.”
Coma	State of extreme unresponsiveness, with no voluntary movement or behavior; sometimes normal reflexes may be lost. If the state persists for more than a month it becomes a persistent vegetative state. There is maintenance of autonomic functions. There are “grades” assigned to comas, delineating severity and type.
Brain death	Initial requirements for determination include: Neuroimaging or clinical findings that suggest catastrophic cerebral events where there are no identifiable other conditions that may represent brain death, such as severe metabolic derangement, intoxication, poisoning, or other acute vascular anomaly. There is persistent unresponsiveness, there are no reflexive responses to pain or any brainstem responses, and there is apnea, with cessation of respiration and circulation.
Uniform Determination of Death Act (UDDA)	^§ Adopted by almost all of the states in the United States, the UDDA 7180, 7181, and 7182 states: 7180, 7181 “An individual who has sustained either (1) irreversible cessation of circulation and respiratory functions, or (2) irreversible cessation of all functions of the entire brain, including the brain stem, is dead. A determination of death must be made in accordance with accepted medical standards.” 7182 “When an individual is pronounced dead by determining that the individual has sustained an irreversible cessation of all functions of the entire brain, including the brain stem, there shall be independent confirmation by another physician.”

Terminology and definitions differ slightly across Westernized nations, but several European nations have paved the way for discussions regarding EOL care in the United States and have enacted laws that set precedents for U.S. courts and medical bodies (10). Religious authorities have also described the beliefs and procedures appropriate for EOL care in various position papers. Associations such as the American Dietetic Association, Association of Speech Pathologists, and the Association of Hospice and Palliative Care, among others, have issued statements that describe considerations and procedures for EOL care in the competent and incompetent older adult (11-13). Consensus of findings indicates that legally documented personal choice should determine what can and cannot be done in terms of ANH in an older person near death (14). Unfortunately, this may not transpire in an emergency that is fraught with emotion, confusion, and occurs in a litigious environment (15).

Legal Issues Regarding ANH in End of Life Care

Several important decisions have been made in the past half-century regarding the legalities of provision, withholding, and withdrawing artificial nutrition and hydration. In a position paper from the American Dietetic Association, the Cruzan and Shiavo cases are outlined and the ensuing decisions explained in terms of ANH withdrawal under specific circumstances. Table 2 describes several important rulings governing ANH as a result of those cases.

TABLE 2 Decisions Resulting from Court Cases Involving ANH in Incompetent and Competent Patients ^§

Court case	Decision
Plaza Health 1984 (US)†	Competent elderly have a right to refuse to eat and/or have a tube placed for feeding.
Elbaum 1989 (US) ^∏	Moneys may not be collected for continued ANH when the continuation is against the wishes of the patient or the surrogate.
Cruzan 1990 (US)†	1. ANH is a “medical treatment.”2. A “surrogate” can voice the wishes of a patient and request withdrawal of ANH.3. “Denial of personhood to preserve the sanctity of life” is not appropriate and removal of ANH is permissible if the patient wanted it done.
Bland 1993 (UK)‡	Discontinuation of ANH when recovery is not possible is in the best interest of the patient.
Glucksberg 1997 (US)^¶	Physician-assisted suicide is not a constitutional right of the patient. Choosing to forgo life-sustaining treatment is.
Vacco 1997 (US) ^Ω	Provision of pain medication (which would hasten death) and deep sedation with withdrawal of ANH may be authorized by states to prevent suffering in a dying person upon his or her request.
Gardner 2001 ^§ (Australia)	There is no imposition of obligation to provide ANH to dying persons and in the absence of advance directives or a surrogate, the physician may withdraw ANH on the basis that it is inappropriate in the current clinical situation for an incompetent older adult.
Oregon Revision Statute Death with Dignity Act 2004 (US)^∧	“Assisted Suicide” is permissible with “an incurable, irreversible disease that has been medically confirmed and will with reasonable medical confirmation produce death within six months.”
Shiavo 2005 (US)*	Decisions regarding withdrawal of ANH must reside with only one guardian or surrogate in those incompetent persons without advance directives.

The Patient Self Determination Act of 1991, requiring that all publicly subsidized health care entities must inform patients of their right to choose whether they wish to be treated via the legal documents of advance directive, does not guarantee that patients will have advance directives, nor does it guarantee that even if they do, their requests will be honored (16, 17). In the event that decisions regarding ANH for an incompetent individual must be made, the complexities of the process and the legal weight of surrogate decision-making varies from state to state and country to country (18). If a competent person chooses to forgo treatment when he or she is dying, even if it is not in his best interest, it is legally binding in all Westernized countries to accept patient wishes (19).

The right to choose not to undergo treatment comes under the doctrines of informed consent and informed choice. Health care providers have the right to investigate the underpinnings of choosing to forgo treatment, for example, incompetence or mental illness. Self-determination allows the individual the right to withdraw, withhold, or otherwise designate a surrogate to do this for him or her in accordance to the 14th amendment to the U.S. Constitution (20). This is fraught with legal implications. “Physician-assisted suicide,” “passive euthanasia,” “double effect euthanasia,” and “aid in dying” are terms that are associated with actions that may hasten death in an older adult who is at the end of life. Physician-assisted suicide is killing oneself by use of a medical means, which is provided but not administered by a health care provider, either taking prescribed lethal medications or removing of life-sustaining treatments, such as ANH, by patient request (21). Many legal experts choose to differentiate between active and passive assistance with death. Withholding and withdrawing ANH is usually seen as passive assistance, legally, but may not be seen as such ethically (22). Legally, physicians may be able to withdraw ANH when it is not clinically warranted, or not in the best interest of the patient, or the state, but most are unwilling to do so (23). Sahm et al. reviewed the German position on passive versus active euthanasia and the distinction between palliative care and aid in dying and found that there was confusion among most regarding the differences. A decision not to act is still an act, thus not providing ANH is no different than removing ANH, and this implies that all direct or indirect methods should be legally sanctioned (24).

Tucker et al. reported on the new “Medical Right” or the influence of religious fundamentalist beliefs in medical ethics, which limit the provision of choices and information to patients at the end of their lives. The influence of fundamentalist views into clinical decision-making and provision of information to patients extends to provider's failure to explain a patient's right to choose to forgo ANH or the process of Voluntary Stopping of Eating and Drinking (VSED). It also curtails the provider's ability to refuse to perform medical procedures or institute technological interventions, such as feeding tube insertion. The “Right to Know End of Life Options Act,” passed in California in 2008 met vehement opposition, despite its seemingly innocuous dictate of ensuring that dying patients will be told their diagnosis, their options, and their timeframe until death, when they want to know this information (25). Communication between provider and patient regarding EOL issues is tantamount and should not be obscured by legal misconceptions or fears of litigation (26).

Ethical Issues in End of Life Care

Despite the knowledge that starvation and dehydration do not contribute to patient suffering at the end of life, and in fact may contribute to a comfortable passage from life, the ethics of not providing ANH continue to be hotly debated (27). Many investigators have found that dying persons overwhelmingly deny thirst and hunger; and application of lubricants or provision of ice chips and oral hygiene alleviate many xerostomic complaints (28). Provision of oral food and fluids should be strongly encouraged, as it gives comfort, pleasure, a sense of autonomy, and dignity and is viewed as essential care giving (29).

There has been little evidence that ANH extends the quantity of life, and in fact it may decrease quality of life in elders close to death (30). Provision of ANH may lead to false hope for the patient, his or her family, and the practitioner, which can be viewed as unethical (31). In a survey of physicians and surrogate decision makers at 288 institutions, the majority erroneously believed that ANH would provide the benefits of prolonged life, less aspiration, ease of symptoms, ease of medication provision, and improved nutritional status (p < 0.05, p < 0.001, p < 0.05, p < 0.001, p < 0.05, respectively) (32). In a survey of practitioners in residential facilities, the majority reported that ANH should be provided to prevent starvation at the end of life (p < 0.001), and the majority reported feeling ill-informed and unsure about ANH sequelae in the dying older adult and felt unable to communicate about these issues to family and surrogates (p < 0.03) (33). Practitioners must be able to address ethical issues surrounding ANH at the end of life. A framework for open communication regarding ethics near death can be found in Table 3. Providing information to patients and their families on all the choices for treatment at the end of life is essential.

TABLE 3 Questions to Ask Regarding the Ethics of Providing ANH at the End of Life^{¶ †§‡}

Process	Question
Framing	1. Is the patient able to make autonomous decisions?2. Are the patient's choices in line with professional assessment of beneficence?3. Are there conflicts in an ethical or moral sense?4. What is the nature of the decision that needs to be made?
Data collection	1. What are the facts regarding diagnosis, prognosis, and treatment outcome for this patient at this point in time?2. What are the religious, cultural, social, spiritual, and personal issues for this particular patient?3. What is the degree of physical, psychological, and spiritual suffering that the patient is experiencing?4. Is the patient clinically depressed, and if so, is it influencing his or her decision-making abilities? Will treatment of the underlying depression result in a different outcome?5. Is the patient demented? If so, does the harm of providing ANH outweigh the benefit?
Decision-making	1. Is the patient or a surrogate making the decision?2. Is there adequate information on the values, preferences, and wishes of this patient?3. What clinical options have been outlined?4. Have the ethics of each course of action been weighed and their true intent delineated (e.g., fiscal consequences to the family determines removal of ANH)?
Determinism	1. Has efficacy, benefit, and informed choice been conveyed to all involved in the decision-making process?2. Has conflict between opinions of professionals, the patient, the surrogate, the family, and any other entity been articulated?3. What steps have been taken to resolve these conflicts?
Individuality	1. Has every patient been treated as a unique case?2. Has a blanket approach to provision, withholding, and withdrawing ANH been taken? Have institutional policies, procedures, and culture been adequately evaluated to prevent a blanket approach to care?3. Is the decision right for this particular patient at this particular time and in this particular place?4. Has the decision been re-evaluated on a daily or even hourly basis?5. Has patient autonomy been sacrificed for sparing professional and/or family distress?6. Have steps been taken to ensure that stopping ANH has not resulting in stopping care?7. Has open ongoing communication been central to the process?8. Has adequate support been provided to the patient, the family, and the staff to ensure successful outcome, regardless of what course of action is taken?

Several investigators have argued that provision of sedation without artificial hydration is tantamount to slow euthanasia and is unethical (34, 35). Others believe that dehydration at the end of life is natural, ethical, and imperative, with or without sedation (36). In a case series report, Quill et al. described strategies for bridging the gap between both camps by identifying an alternative; namely patient choice to forgo food and fluids with a request for terminal sedation (37). Withholding food, fluids, and ventilation in a patient who has requested and received high doses of sedatives does result in a hastened death, which begs ethical, legal, and moral questions (38, 39). This is seen as “the slippery slope” of ANH withdrawal or withholding (40, 41). Justification for such actions can be found in patient autonomy and practitioner beneficence as long as the patient who requests these practices does not inform his or her provider that he is intentionally trying to hasten his own death (42). Thus, physicians are not legally, morally, or ethically guilty of assisted suicide, as long as the clinical presentation denotes certain death with discomfort in the near future (43).

ANH IN DEMENTIA CARE

Most persons, when asked if they wished to be kept alive via ANH, if demented at the end of their lives say no, but in the time of crises, many will have a tube placed (44). Most persons wish practitioners would start these discussions earlier than the end of life to help frame the ethics and decisions of dementia care near death (45). Older patients with dementia pose a unique challenge for practitioners concerning ANH (46). In demented patients, one of the first activities to fail is the ability to independently maintain oral nutrition and hydration (47). The weight loss, malnutrition, dehydration, and dysphagia seen in these patients is widely recognized as the entrance into the end stages of the disease (48). It is approximated that more than half of the oldest old have dementia, many in the advanced stages of the disease, and ANH is increasingly used to provide sustenance. Many demented patients will be kept alive long enough to progress to a persistent vegetative state (49). More than 20% of all older demented patients in the United States who have either been hospitalized or admitted to skilled nursing facilities have had tubes placed, and nearly half of those patients died within a year. The greater the number of hospitalizations and transfers between facilities, the greater the odds of a tube being placed in a demented older person (OR = 2.5, 95% CI = 2.0–3.2, p < 0.01) (50).

Provision of ANH in a demented patient often requires restraints to prevent removal of tubes (51). Two-thirds of nursing home residents will become agitated and pull their tubes within the first two weeks (52). Some studies have shown that upwards of 70% of dementia patients are fitted with mittens or other forms of undignified restraint for protection from tube pulling (53). Restraints may also limit mobility, increase anxiety and combative behaviors, increase skin abrasion and breakdown, decrease socialization, and engender other functional losses (54). Palliative care provision is ethically bound to prevent such suffering (55). In an effort to reduce tube placement in demented patients an educational program was instituted and a Palliative Medical Ethics Consultation Service was established for a network of inner city hospitals. A dramatic reduction in feeding tube placement was seen (p < 0.002) (56). There has also been some discussion of demented patients pulling their tubes, as a “direct expression of their will,” which should be honored and ANH subsequently withdrawn (57).

Many practitioners feel that an impaired patient is at the mercy of his or caregivers and ANH should be provided, continued, and defended as a fundamental obligation of basic care (58). Many practitioners are misinformed about the benefits of tube feeding in advanced dementia, with no appreciation for the harm (59). Practitioners, even geriatricians, do not know and do not inform surrogates of the long-term hazards of tube feeding a demented patient, such as tube pulling, infection, aspiration, or other consequences (60). In a survey of physicians, a hierarchical regression evidenced that tube placement was recommended for demented patients at risk for aspiration pneumonia (OR = 0.99, 95% CI = 0.98–0.99, p < 0.05), diagnosed dysphagia (OR = 1.82, 95% CI = 1.25–2.65, p < 0.001), weight loss (OR = 1.89, 95% CI = 1.33–2.7, p < 0.001) to prevent uncomfortable death (OR = 1.70, 95% CI = 1.17–2.48, p < 0.01) and solely due to diagnosis of end stage dementia (OR = 1.87, 95% CI = 1.23–2.83, p < 0.01) (61). Unfortunately there is little acknowledgement of terminal dehydration and starvation being naturally anesthetic (62-63).

There is no data to support the use of ANH in advanced dementia (64). Families who are uninformed are opting for Percutaneous Endoscopic Gastrostomy (PEG) tube placement with increasing frequency, such that 30% of all PEG tubes are placed in patients with dementia and more than 10% of institutionalized demented patients are tube fed (65). Eggenberger proposed an ethical framework for practitioners to assist families and surrogates with decisions regarding ANH for dementia. The author suggested providing information on the hazards of ANH, clearly communicating the prognosis, describing the benefits of natural diminution of hydration and nutrition, and ethically describing and focusing the family's attention on “quality of life” (66). A Cochrane Review of ANH in advanced dementia patients found that there was a significant void in the scientific literature regarding quality of life indicators and hazards analysis in this population, particularly in the area of PEG and Nasogastric (NG) tube feeding (67). This makes it ethically difficult to convince families of the hazards and decrements in quality of life for demented patients with provision of ANH. Evidence needs to be readily available to practitioners for educational purposes (68).

Clinical Issues in End of Life Care

Artificial feeding and hydration can be done in a variety of ways. Some of the procedures are less invasive than others, and costs of feeding and hydrating vary dramatically from procedure to procedure (69). Risks also vary from method to method, with surgical procedures or intravenous methods being riskier in an older person. Feeding methods include Nasogastric or Nasoduodenal (NG, ND) tube feeding, Percutaneous Endoscopic Gastrostomy (PEG) tube feeding, surgical placement of the feeding tube to the duodenum or jejunum, and Peripheral or Total Parenteral Nutrition (70). Hydration methodologies include using hypodermoclysis, rectal fluid infusion, and intravenous (IV) hydration peripherally or through catheterization of the sub-clavian vein (71). The latter is known as placing a central line, which is often fitted with several ports, so that a variety of substances can be infused simultaneously (44).

Ratios of nursing staff to patients at the end of life are related to dehydration and weight loss (72). Societies for enteral and parenteral nutrition worldwide have issued statements regarding the unacceptability of initiation of aggressive measures such as tube or IV feeding in geriatrics, if it is done solely to facilitate care with short staffing, to save time or placate family/staff (73). Provision of oral nutrition and hydration should be performed with patience, care, and time, whenever physiologically possible.

NON-ORAL, NON-SURGICAL TUBE FEEDING

Non-oral feeding includes less invasive measures such as a nasogastric tube, where a flexible plastic tube is inserted through the nose, down the esophagus, and into the stomach. A longer tube can be driven down past the pylorus to the duodenum of the intestine in the event that the stomach is not functional (74). These tubes are easily displaced, pulled, or inefficient in a confused older adult. The risk of aspiration is also heightened, even more so than careful oral feeding in this population (75). In a prospective study of mortality following NG tube feeding initiation in frail elderly, 65% died within 6 months of the tube insertion. Significant electrolytic disturbances were seen following NG feeding from baseline (p < 0.005) (76).

NON-ORAL, SURGICAL TUBE FEEDING

A more invasive procedure includes Percutaneous Endoscopic Gastrostomy (PEG) tube placement. There are three methods for placing the tube: endoscopic, surgical, or radiological (77). The most common is endoscopic, where an opening through the abdominal wall is made, and using endoscopic surgical procedures, a tube is fitted into the opening or stoma. Open surgical procedures are sometimes required for tube placement. This tube is used to administer formula that is appropriate for specific conditions directly to the stomach (78). In the event that the stomach is not functional, a tube can be placed further down in the digestive tract, so the duodenum or the jejunum is accessible. Formulas are available with the appropriate stage of digested macronutrients for the section of the gastrointestinal tract that is viable (79).

In a state-wide referendum for long-term care facilities, patient preferences for tube insertion were recorded and immediately translated into a “Physician Order for Life Sustaining Treatment” (POLST). Patients at participating facilities were educated about feeding tubes, and their preferences for EOL tube insertion were documented and honored. The program was associated with significant increases in tube refusal and subsequent withholding of ANH by providers at the end of life (80). Meta analyses show PEG placement has a negative effect on survival times in terminal older adults. Pooled proportions for survival post tube insertion were: 1 month = 0.81 (95% CI = 0.74–0.88), 2 months = 0.70 (95% CI = 0.65–0.74), 6 months = 0.56 (95% CI = 0.20–0.92), 1 year = 0.38 (95% CI = 0.26–0.49), revealing a progressive decline (81).

In an attempt to determine which clinical indices are predictive for tube feeding benefit among older persons, a retrospective study of all admissions at several facilities was performed. In those 80 years or older, ANH was associated with increasingly negative outcomes that were accurately predicted (90% positive predictive value) by albumin, pre-albumin, and C-Reactive Protein levels upon admission (82). Several benefits are seen with starvation including:

1.	deprivation leads to endogenous opioid production,
2.	ketosis produces somnolence and euphoria,
3.	reduction in gastric distention decreases peptide activation of hypothalamic receptors,
4.	reduced urea cycle metabolites results in less sensation, and
5.	hypernatremic conditions decrease nociception (83, 84).

Figure 1 depicts a framework for use of tube feeding in geriatrics. Despite evidence to the contrary, many providers feel that ANH is still an “ordinary” expression of caring in vulnerable populations such as dying older adults (85, 86). Such feeding modalities may be seen as an obligatory medical treatment no different than the administration of pain medication to reduce suffering (87, 88).

FIGURE 1 Decision tree for tube feeding.^§†‡

NON-ORAL HYDRATION

Oral hydration at the end of life is difficult to accomplish, the staff time and costs required to provide oral fluids and ice chips and to ensure adequacy are prohibitive (89, 90). Older adults have difficulties maintaining hydration. Xerostomia, due to a variety of factors such as hyposalivation, body water deficits, medications, and pathophysiological conditions is seen (91). Arguments exist for and against provision of non-oral hydration at the end of life (92). Arguments against hydration include life prolongation without symptom amelioration, increased use of urethral catheterization and urinary tract infection, greater nausea, vomiting, secretions, edema, ascites, skin breakdown, and “death rattle.” Death rattle is the noise produced by secretions in the airways, generally seen in dying patients (93). Dehydration is considered a natural anesthetic, and parenteral hydration can limit patient mobility and cause unnecessary discomfort (94). Arguments for non-oral hydration include decreasing toxicity from drug metabolites (usually opioids), decreasing nitrogenous waste accumulation, and providing the appearance of “doing something” even though there may be little value (95). Vullo et al., in a prospective study of dehydration in terminal patients, found that serum electrolyte levels and ratings of comfort remained within normal limits despite marked dehydration and impending death with no provision of fluids (p < 0.05, p < 0.05) (96). Soden et al. did not find longer survival times in hydrated versus dehydrated patients at the end of life (97).

Hypodermoclysis, also known as subcutaneous fluid administration, is a method of artificial hydration that has been used extensively for the past two decades in older adults. Considered less dangerous than intravenous or central line administration, fluids are infused through a butterfly needle into the subcutaneous tissues. While there can be reactions at the site, localized edema, or infection, it is considered safe, effective, and less expensive than other methods (98). Electrolyte solutions are used to prevent third spacing, and infusion rates can range from 120–500 ml/h. In a survey of patients and families in EOL care, it was found that hypodermoclysis was viewed less effective than IV fluids, when that was not the case. Patients and families were under the impression that when IVs were used, the amount, efficacy, and hydration provided by IV fluids was superior to that seen with hypodermoclysis, even though the amount, efficacy, and hydration were the same. The use of a butterfly needle was thought to be less effective, perhaps due to its decreased visibility and invasiveness (99).

Rectal hydration (proctoclysis) can also be used when resources for subcutaneous routes are scarce. Tap water can be used at rates of 100–400 ml/h, and aside from leakage there are few adverse effects (100). Tubes, such as NG, ND, PEG, or PEJ provide an easy route for fluid administration if already in place. Peripheral or central catheterization expressly for intravenous hydration is rarely recommended at the end of life (101). If a line is already in place, then IV provision of fluids is standard. Intravenous infusion of fluid can range from 250–3000 ml, and often this route of administration does not counter the xerostomia exhibited by dying patients (102). In a national survey of Japanese providers, it was found that <50% were aware of the failure of IV fluids to overcome xerostomic discomfort. Educational programs increased the likelihood of improved oral care and proper techniques for amelioration of xerostomia (lubrication, ice chips, and sips of water) for 95% of the providers who took part (103).

TOTAL PARENTERAL NUTRITION AND HYDRATION

This method of feeding and hydration requires the older adult to undergo the insertion of a catheter and ports to the sub-clavian vein in a clinical setting. This procedure is considerably more expensive than other routes. Central line placement increases the risk for sepsis, air emboli, line pulling, and other complications (104). Parenteral nutrition and hydration should only be considered in patients who expressly desire this route and whose expected survival is more than 3 months (105). Nutrients and fluids are infused at tolerable rates along with medications, electrolytes, and micronutrient mixtures. Constipation, nausea, xerostomia, and other symptoms may worsen despite adequacy using this route (106). Several reviews of TPN provision at the end of life in older persons have shown no improvement in survival (107).

Religious Issues in End of Life Care

Spiritual beliefs influence choices regarding EOL care. In a broad sense, spirituality encompasses religious beliefs, existentiality, and mysticism. In the United States 90% of Americans believe in a deity or “higher power”; 66% pray regularly and/or belong to a religious entity; 60% acknowledge religion as being central to their lives; and 82% maintain that there is a need for spiritual advancement in themselves and others (108). In a study of older inpatients with terminal illness, 90% reported using religion to cope (109). Spiritual conviction may shape a person's desire to prolong life, to avoid punishment in the afterlife, or, conversely, to hasten death and enter the afterlife, which he or she believes, will relieve all pain and suffering (110).

Spiritual beliefs also influence EOL decisions made by health professionals for their patients. Studies have shown that the providers’ degree of spirituality results in differences in the endorsement of euthanasia, physician-assisted suicide, and use of technological advances to prolong life (111). There are guidelines from the American Medical Association, Joint Commission on Accreditation of Healthcare Organizations, and the World Health Organization regarding respect for the spirituality of patients and the influence of spirituality on quality of life and EOL care (112-114). Familiarity with these guidelines aids in the provision of information to the older adult and his or her family, when faced with death. Similarly, it informs providers on how to react when a patient (or his or her surrogate) uses religious beliefs to justify EOL care decisions in the face of futility (115). Such justifications may include “[my deity] will grant a miracle” or “We cannot interfere with [my deity's] plan for us.” It is beyond the scope of practice for most nutritionists to address issues of spirituality and decision making at the end of life (116). Referral to a more appropriate person, such as a hospital chaplain or other clergy member, must be made. It is, however, within the realm of medical care providers, including nutritionists, to have some knowledge of different belief systems, respect those systems, and know who to contact should the need arise (117).

In the following sections, we explore some of the basic belief systems of the most commonly encountered religions with regards to EOL. This is done with full acknowledgement that each worship community and, indeed, each individual may have a unique set of beliefs. Individuals within any given belief system may not adhere to all tenets of their religion, and personal beliefs and choices vary considerably. Personal choices and individual preferences should guide clinical decision making in all circumstances. Considerable effort should be made to ascertain what the particular individual's wishes are, regardless of his or her recorded religious affiliation.

PROTESTANTISM

Western Christianity has so many denominations and observances that one view on “EOL care” is impossible. The view on death in and of itself is more unified; with most believing that impending demise must be framed in the opportunity for repentance. If ANH is used to delay death and the opportunity for repentance is lengthened, or if it prolongs the time for penitence and increases the odds for amends (without sedation, which clouds judgment), it should be attempted. If the individual has already repented and prepared for salvation, then technological measures and ANH should not be employed (118).

The Protestant Church has many denominations, each of which has its own position on EOL care. The United Methodist Church, the largest branch of the Wesleyan denominations, has recently come under fire for “intentionally ambiguous” wording regarding EOL care, physician-assisted suicide, and euthanasia. The doctrine cites “death with dignity” as the position determining the extent to which treatments be applied to the dying older adult. Furthermore, the Book of Discipline states:

We assert the right of every person to die in dignity, with loving personal care and without efforts to prolong terminal illnesses merely because the technology is available to do so. (119, p. 91)

Additional resolutions made by the United Methodist church include: (a) references to the persistent vegetative state, (b) the ability of the health care team and family to withhold or withdraw technology, which prolongs life in the face of futility, and (c) in cases when a person's suffering and burdens of living outweigh the benefits in the face of inevitable death, the cessation of life may be considered (120). The church repeatedly advises respect for self-determination and overall treatment benefit. Quality of life and legally documented personal choice trumps technological advancement in EOL care.

More conservative or fundamentalist branches of the protestant church are more in line with Catholicism in terms of EOL care (121). The belief that life is a gift from God and that stewardship of that life is the ultimate responsibility of the individual results in a less favorable attitude toward removal, or withholding, of ANH. Some branches feel that pain and suffering at the end of life is in line with the suffering of Christ and assists in the preparation for the transition to the afterlife (122). Provision of fluid and nutrition without sedation or pain medication may be requested. Consultation with family and clergy is important, but the wishes of a mentally competent individual trump the concerns of the health care team, family, friends, and even clergy for most Protestant denominations. There are exceptions, such as fundamentalist groups that eschew all medical intervention, who may forbid all treatment at the end of life, despite a competent patient's wishes.

CATHOLICISM

The Catholic Church condones palliative care at the end of the older adult's life. This is to ensure the dignity of life and make suffering bearable (123). Any attempt at euthanasia or assisted suicide is prohibited (2). The church's position also states that “extraordinary” treatments are not necessary or required. These include excessively costly, painful, dangerous, or otherwise unattainable treatments. However, this categorization of “extraordinary” does not include artificial nutrition and hydration, even to someone in a persistent vegetative state (124).

In 2004, the Pope found that ANH was not “extraordinary” nor was it a medical act. The church maintained that provision of nourishment by any means is considered natural and should be administered to reduce suffering (125). The removal of ANH is considered euthanasia by omission, unless the case can be made for the artificial nourishment causing significant distress to the older person. This may be the case when there is a complication, such as an infection, or in the event that assimilation of nutrients is no longer possible due to organ system failure (126).

There is considerable controversy regarding the Vatican feeding tube doctrine. Catholic health care facilities have been forced to change policies and procedures regarding individual autonomy of decision making for EOL care (127). It should be noted that a case was made for the failure of ANH to prolong life and reduce suffering in older persons with advanced dementia, and there have been discussions among American diocese regarding ANH being extraordinary in these circumstances, but most continue to defer to Vatican feeding tube policies (128). Furthermore, advanced directives that are not in line with the Vatican cannot be honored at Catholic health care facilities. Providers must be cognizant of this when dealing with older patients and their families when death is near (129).

In statements put forth by the Administrative Committee of the National Conference of Catholic Bishops in 1993 titled “Nutrition and Hydration: Moral and Pastoral Reflections” there is considerable attention given to calculation of “benefit” versus “burden.” The calculations should not include the fiscal or other costs to the family, the patient, the hospital, or the government but may include the emotional costs of debilitation when no response or benefit is seen (130). Scholars in Roman Catholicism also acknowledge that the afterlife warrants allowing death to occur, and “doing everything possible” to extend life, including provision of ANH, is counterintuitive, as “joining” the Father, Son, and Holy Ghost is desirable (2). This position differs from other monotheistic religions, such as Judaism and Islam.

JUDAISM

The Jewish rabbinical doctrine differs slightly between Orthodox, Conservative, Reformed, and Reconstructionist branches of Judaism in terms of the beliefs and issues regarding EOL care and ANH as well as on issues of surrogacy and decision making for the older adult (131). While recognizing the inevitability of death, life is held sacred and preservation of life is of great importance to religious Jews. Jewish medical ethics are the basis for the four tenets of secular medical ethics (Autonomy, Beneficence, Non-malfeasance, and Justice) with only slight differences (132). These are outline in Table 4.

TABLE 4 The Four Tenets of Jewish and Secular Medical Ethics^§

1. “Autonomy in Judaism means that an individual's decision-making is his own, but should comply with Jewish beliefs and medical authority.”

2. “Beneficence in Jewish teaching implies that religious Jews should attempt treatments, which would extend life and impart benefit.”

3. “Non-malfeasance implies that it is the individuals’ responsibility to avoid harming themselves and their god given bodies.”

4. “Justice, which can be interpreted as fairness in limiting or denying treatment or care based on availability and resources.”

Orthodox Judaism espouses ANH as basic care as long as it is done in a manner that benefits the patient and does no harm. An individual may only refuse treatment if it does no good or causes suffering or complications (133). Treatment may only be withheld or withdrawn if it delays death while causing severe additional pain and suffering, but may not be withdrawn or withheld if it extends life and causes no additional major complications. Advance directives can be honored if a Rabbi has been included in their development (134).

Conservative Judaism interprets Jewish medical ethics to fit modern society and Reformed Judaism sees Jewish medical ethics as a nonbinding guide such that the individual's wishes trumps traditional views (135). Reconstructionist Judaism spans the spectrum in terms of keeping with Jewish medical ethics, but all denominations accept intentionally shortening one's life (suicide) as forbidden. Jewish medical ethics prohibit assisted suicide or euthanasia because deliberately speeding up a person's death is considered murder (136). Competent patients have the right to forgo treatment at the very end of life under specific circumstances in Judaism; however the physician is not under any obligation to tell the patient that he or she is near death in those circumstances, particularly if telling the patient the truth will result in additional pain and suffering (135). Withholding the timeline to death from older patients is permissible and is often encouraged.

Judaism texts explicitly state that providing relief from pain is tantamount for the dying person, and all methods of pain relief must be employed (137). Because dehydration may reduce pain and nutrition support may actually do more harm then good, many make the case for withholding ANH because it violates the doctrine regarding delaying death while causing increased pain and suffering (36). It should be noted, however, that terminal dehydration, hospice without provision of ANH, and withdrawing or withholding ANH is not considered aligned with Jewish teachings unless there is proof of “goses” (less than 72 hours until death) and futility of intervention under any denomination of Judaism. Once death is reasonably determined to be imminent, the 72 hours prior should be peaceful with no invasive or uncomfortable procedures. Futility of intervention when death is imminent should be determined by several qualified medical professionals (139).

Surrogate decision making by Rabbis, family members, friends, or the physician has some specific considerations. The Jewish older person who has no legal documentation, who is not competent, and who has no Rabbi, friends, or family members who are privy to his or her wants must be dealt with in a very specific manner. This is important for health care practitioners to understand and honor as it affects the older adult's ability to be buried properly in a Jewish cemetery (139).

ISLAM

Islamic beliefs regarding EOL care include tenets from Islamic law or “Shariah.” Because life on earth is preparation for the afterlife, death is not to be resisted or feared. Death is not to be hastened, either, and euthanasia is not permitted in the Islamic religion. Suffering at the end of life is also not required or condoned. Comfort in dying is important so that the person can pray and seek forgiveness from a lifetime of excess and from others who he or she has wronged. Practicing Muslims often require home care so they may die surrounded by visitors and appropriate religious articles (140). Many Muslim scholars advocate for withholding ANH when the patient will not recover, but the withdrawal of ANH is more complex and must be weighed against the concept of “doing no harm” and decreasing suffering in the family since the family must watch over the dying older adult. Muslim family and friends may experience distress with failure to provide nutrition and hydration, and this must be factored into the decision (141).

Muslims espouse the right of the patient to forgo ANH, deference to the advance directive or the family's wishes, when death is close and inevitable (142). Muslim law prohibits active euthanasia, suicide, death to avoid suffering, and choosing to forgo medical intervention when there is a definitive cure or chance of recovery (143). In addition, a Muslim cleric should be involved in any discussions regarding ANH and EOL care. There are many Islamic tenets regarding how to behave in the face of death, how the person and his or her family must act, how prayer and washing must be handled, and the hierarchy of who makes decisions and how interactions may occur. It is important for nutrition professional to be sensitive to these issues and respect the belief systems as not to bring shame to the family or the patient (134).

HINDUISM

The Hindu religion believes that the soul can be released in a “good death,” which occurs in old age, at the right astrological time, and either in the home (on the ground) or at the banks of the Ganges river (144). Hinduism is polytheistic and involves a transcendent god that lives in all beings. The process of rebirth is dependent on the person's acts during his or her lifetime. Freedom through death is “Pravrajya” and allows reuniting with the Supreme Being in a gradual fashion (145). EOL care is seen as a familial responsibility in Hinduism and is a time of spirituality and ritualized action. Provision of ANH may be seen as important to a good death when the end is clearly not near; pain and suffering without the consciousness being clouded by analgesics may also be seen as tantamount to achieving oneness with the Supreme Being. Withdrawal or withholding of futile treatments such as ANH when death is inevitable is permissible, and many Hindus will prepare themselves for death by fasting (146). It is of interest that the Hindu patient is usually shielded from his or her diagnoses and any medical decision making; he does not have the autonomy to request or forgo ANH. This decision is left to the family and the community as not to taint the patient or his or her chances for a good death and successful reincarnation (147).

BUDDHISM

Buddhism is commonly practiced in Asian countries and is becoming more prevalent in Westernized areas. There are several forms of Buddhism. The two main branches are Theravadin and Mahayana. Mahayana has a subdivision of Vajrayana or “Tibetan” Buddhism, from which the Dalai Lama is the most recognized representative. Buddhist beliefs regarding EOL care differ from most other religions in that tradition dictates that rebirth is tantamount, all who are born will die, cumulative actions during ones lifetime dictate death and rebirth, and that all beings suffer. It is considered obligatory to prepare for one's death and to start early in this preparation (148). The “consciousness” of the being must be allowed to exit the body properly. A “good death” and “enlightenment” is important, and inappropriate provision of ANH interferes with the attainment of both (149).

Buddhists may allow withholding and withdrawing ANH (150). Spiritual comfort at the end of life is important in Buddhist tradition, while technological extension of life is not (151). Spirituality and wishes regarding ANH in the context of EOL care for patients practicing Buddhism must be carefully considered to allow the consciousness to exit the body properly and provide for a “good death.” Practitioners should aim for consulting with persons familiar with Buddhist traditions and spiritual needs in the absence of patient or surrogate ability to provide information since it is an essential component of care in this population (152).

OTHER RELIGIOUS SECTS

There are several fundamentalist religious entities that are opposed to medical intervention, including ANH (153). The largest of these is the Christian Scientist movement. Some of the smaller sects, which shun most medical interventions, include The Body, Church of the First Born, End of Time Ministries, Faith Assembly, Faith Tabernacle, Christ Church, Grace Baptist, and other geographically isolated, fundamentalist congregations. These do not include Jehovah's Witness, who believe that blood cannot be transfused, but ANH is permissible; nor does it include Scientologists, the majority of whom do not have specific concerns regarding ANH, but rather eschew psychotropic medication for psychological conditions (154).

Racial, Ethnic, and Cultural Issues in End of Life Care

Belief systems regarding death and dying in general often differ by racial and ethnic group and cultural orientation. In some cultures, death is celebrated; in others friends and family are not allowed to speak the name of the dead. Because of the vast spectrum of differences in terms of death and dying, the health care provider should consult with the older adult or persons familiar with the dying older adult's racial, ethnic, and contextual considerations before determining appropriate EOL care (155). In the absence of knowledge regarding any particular person's cultural and ethnic background and without the individual's or surrogate informant's ability to provide some indication of what should be done, a conservative approach to dealing with artificial nutrition, hydration, and other medical life-sustaining treatments should be undertaken (156).

In terms of EOL decision making, Caucasians were most influenced by medical information and fiscal considerations, while African Americans considered spiritual input most important (157). Hopp et al. studied racial preferences and delineations for EOL treatment and found that Caucasians were more likely than African Americans to have a living will (p < 0.001), designate a Durable Power of Attorney for Health Care (p < 0.032), and discuss treatment preferences before death (p < 0.002). Caucasians were more likely to limit aggressive intervention at the end of life (p < 0.007) and withhold artificial feeding and hydration (p < 0.034). Treatment decisions for African Americans were more likely to incorporate all possible interventions in order to prolong life (p < 0.013) (158).

A qualitative study found that African American women wanted aggressive intervention, feeding tube placement, and medical practitioner decision making to prolong life. This was in contrast to the Caucasian women surveyed, who were opposed. Hispanic women were found to be in favor of aggressive intervention but did not want feeding tubes placed (159). Mexican American men were found less likely to want aggressive measures and were more concerned about functional abilities at the end of life as opposed to their female counterparts (160). Asian and Hispanic ethnicity was found to be a predictor for greater emphasis on surrogate decision making, and all non-Caucasians were found to be less informed about advance directives and aggressive treatments near death (161). When European Americans were compared with Korean Americans, African Americans, and Mexican Americans, the Caucasian group was found to be the least likely to accept aggressive EOL care (p < 0.001). Mexican Americans were in favor of life extension through ANH and aggressive intervention (p < 0.001). Korean Americans were in favor of aggressive care (RR = 6.7, p < 0.0001) but not for themselves (RR = 1.2, p < 0.45), and African Americans were most likely to want to be kept alive by artificial means (RR = 2.1, p < 0.002) (1). In a study of 4,920 institutionalized older adults, African Americans were far more likely to want and use feeding tubes. In a main effect logistic regression, dysphagia (OR = 5.4, 95% CI = 2.7–10.6, p < 0.0001) and stroke (OR = 2.8, 95% CI = 1.1–4.3, p < 0.03) were predictive of feeding tube request by competent, poor, urban Blacks (162).

In a study of Canadian- versus Middle Eastern-skilled nursing facilities, researchers found that patients with end-stage dementia (n = 376) were intubated and provided with ANH far more frequently in the Middle East. Statistically significant differences in the numbers of NG and PEG tube fed patients were seen (19% vs. 52.9% χ² p < 0.001) in comparable facilities (137).

In several studies conducted in Taiwan with predominantly Buddhist patients, it was found that most did not understand that ANH was providing excess nutrition and hydration at the end of life. Upon explanation, patients and families wished for a “good death” or withdrawal of ANH (OR = 0.53, 95% CI = 0.37–0.84, p < 0.005) (163). It should be noted that in oriental culture it is not respectful to disclose terminal illness and impeding death to an elder, making communication difficult. It is also culturally inappropriate to let an older adult “starve” (164).

Sexual orientation also altered views on EOL care and artificial nutrition and hydration. The gay community was found to favor physician-assisted suicide, euthanasia, palliative care, and was more likely to have advance directives and decline ANH or other aggressive measures than non-gay counterparts (165). Cultural competence is becoming increasingly important as populations diversify and requires practitioners to integrate culture, spirituality, religiosity, race, ethnicity, gender, and sexual preference into the process of EOL care (166).

Personal and Professional Issues in End of Life Care

Many persons at the end of life do not understand that ANH is a medical intervention and not just provision of food and water (167). In a study of currently healthy older adults (age μ = 79 years), several diagnoses and EOL scenarios were presented and 86% overwhelmingly rejected the use of ANH in treatment plans (p < 0.001) (168). Physician attitudes shape the preferences of patients who are less informed about potential interventions and survival, quality of life, and use of restraints should tube pulling become an issue (169). Bito et al. surveyed physicians to ascertain attitudes toward intervention at the end of life. Physicians themselves are unclear about the course of action in many instances regarding provision of, withholding of, or withdrawal of ANH and communicating the pros and cons to patients and families. Physicians who sought clinical ethics consultations were more likely to withdraw tube feeding and communicate outcomes more effectively (OR = 6.4, 95% CI = 2.5–16.3, p < 0.001) (170). Conflict avoidance in the instance of a family wishing to override an advance directive or a physician's opinion may result in shaping of patient treatments and choices by the Clinician (15). Physician attitudes, religiosity, and worldview shapes their opinion on intervention, withholding, and withdrawal of ANH; and this, in turn, subtly influences patient and family choices (171).

In a study of physicians, it was determined that physicians were no more likely to have advance directives or documentation limiting ANH for EOL care than the general population (172). Physicians were more apt to want ANH and aggressive treatment for themselves, which grew with increasing levels of physical and psychological debilitation experienced with aging and morbidity (AOR = 2.1, 95% CI = 1.2–3.9 and AOR = 5.3, 95% CI = 1.6–17.8) (173). Palliative care professionals, who were educated in death, dying, and comfort care, overwhelmingly support terminal dehydration and starvation in opposition to acute care personnel and the families of end-stage patients (174).

In a mixed method study of health professionals, family caregivers, and patients regarding tube feeding, quality of life, feeding withdrawal, and choice, it was found that patients’ perceptions were not similar to those of their caregivers or the personnel involved in their care. The majority of patients felt they had no choice regarding tube insertion, did not receive enough information, and would choose to stop feeding. This is in sharp contrast to dietitians who overwhelmingly responded that the patient was given a choice, was provided information, and feeding should be continued. Nurses and caregivers fell within those two extremes (175).

In a review of the literature, Bryon et al. referenced several studies that look at a health provider's roles in guiding the decision-making process to initiate, withhold, or withdraw ANH for older adults at the end of their lives (176). Several contextual issues regarding the practitioner's role as a guide in ANH decisions are identified in Table 5. Practitioners should always attempt to place patients’ personal choice into the forefront of the decision-making process.

TABLE 5 Contextual Issues in the Health Professional's Role as a Guide During the Decision-Making Process for ANH Provision at the End of Life ^{§ ‡}

Health professionals should be the patient's advocate so that decisions made by the physician or the family are in line with the patients wishes.

Health professionals should be the guide for the patient, family, and the physician so that the decision is in the patients’ best interest.

Health professionals should be the facilitator of open communication between family, patient, and physician at a time when issues are complex and fraught with emotion.

In an effort to understand the propensity for provision of ANH at the end of life, several reports have indicated that both the public and their health care providers have “lost an appreciation for death and dying, with comfort, dignity and meaning at the end of one's life” (177). Competent patient refusal of ANH has been categorized as “passive voluntary euthanasia,” and many view attempts at informing the patient and his or her family about the benefits of refusal as “assisted euthanasia” (178, 179). Health professional's view of death as a professional and personal “failure” in Western medicine is transferred to the patient and his or her family and results in fear and denial, which in turn leads to opting for aggressive intervention and poor choices for EOL care (180). In contrast to health care providers, and in the absence of imminent death, older adults were found to view quality of life, independence, dignity, and comfort to be important factors to consider prior to death, while life at all costs, quantity of life, and control were unimportant (181).

Despite the passage of the “Patient Self-Determination Act” in 1990 requiring that patients be asked about their advance directive, there has been little increase in discussions and documentation of older adults’ EOL care decisions (182). Health care providers rarely initiate discussions about EOL treatment options with competent older adults prior to emergencies. Discussions that ensue post-hospital admission or in close proximity to death significantly influence choices made for, or by, the patient. The choice regarding provision of ANH is influenced by whether a nurse, physician, or social worker initiated the discussion and whether the provider had been trained in EOL issues (183).

Israeli physicians, nurses, and social workers were surveyed regarding their involvement in ANH decision making at the end of life. Physicians desired the greatest use and social workers desired the least use of ANH. Social workers in Israel reported the highest involvement in discussions with patients and their families regarding the use of ANH and aggressive interventions (184). McFarland et al. found that use of a training curriculum to teach physicians the value of communication, philosophical difference, and acceptance regarding death and dying was ineffective. The Education for Physicians on End of Life Care Curriculum (EPEOLCC) improved knowledge regarding pain and symptom management, but the other variables remained unchanged (p < 0.001, p < 0.001, p < 0.10) (185). Pugh et al. found that among clinicians in the United Kingdom, physicians were more likely to initiate, continue, and oppose withdrawal of aggressive treatment and ANH while nurses were opposed to intitiation, continuation, and aggressive measures in EOL care. Nurses were more likely to consider patient suffering, belief systems, and the need for open communication and education regarding death and dying. Statistically significant differences between physicians and nurses were observed regarding ANH provision and withdrawal (χ² = 9.3, p < 0.01, χ² = 7.4, p < 0.03) (186). The European Union has recently begun to push for transparency of discussions with patients and their families regarding EOL treatment options, inclusive of euthanasia (187).

Cross culturally, Japanese health care professionals have been studied extensively regarding attitudes toward ANH in EOL care. The research found that patient and family wishes regarding ANH were not routinely considered; health professionals were distressed when ANH was withheld, withdrawn, or refused; and disagreements between health professionals regarding provision of ANH were prevalent (188, 189). Miyashita et al. found that the provision of education to Japanese older persons regarding palliative care in the home decreased preferences for aggressive EOL treatment and ANH (190). Morita et al. found ANH was considered a basic standard to be continued until death, and withdrawal or withholding ANH was not acceptable in a national sample of less than 6,000 Japanese older adults (191, 192). Japan only recently developed clinical guidelines for provision of ANH at the end of life (193).

Fiscal Issues in End of Life Care

The research on the costs of ANH provision and EOL care is conflicting and usually does not distinguish expenditures by age. It is believed that the costs of caring for older persons are higher than their younger counterparts, even near death (194). Researchers have sought to refute the assumptions that old age and dying is a primary contributor to health care expenditures in the United States, due to the burgeoning population of older adults (Table 6).

TABLE 6 Seven “Myths” of Aging, End of Life Care, and Fiscal Solvency†

1. “The growing number of older people has been driving the rise in U.S. health care costs.”

2. “As the population ages, health care costs will overwhelm and bankrupt the nation.”

3. “Limits on EOL care for the very old would save Medicare significant amounts of money.”

4. “Aggressive hospital care for the aged is futile and the money spent is wasted.”

5. “It is common for older people to receive heroic, high tech treatments at the end of life.”

6. “Medicare covers everything that older adults need in terms of their health care.”

7. “Advance directives would resolve the dilemma of how aggressively to provide care.”

Pan et al. provided a body of evidence refuting the myths around aging, EOL care, and decrements in fiscal solvency, as shown in Table 6. Statistical analyses of health care expenditure growth over time attributable to the aging population amounted to 7% of the increase, found to be insignificant in light of total expenditures. Forecasting revealed that healthier aging, less chronic disease and disability, and overall improvements in health expectancy would result in less expenditure for older adults until near the time of death. The expenditures for the year before death remained stable, and the high costs of ANH and other aggressive treatments were contributory. The authors maintained that the greater the age, the less likely the use of these measures because of health care rationing based on advancing age. Half the expenditures for Medicare were in aggressive treatment in the past 60 days prior to death, but this constituted only 3% of all expenditures (195).

Several other studies reported vastly different findings. McGarry et al. (196) found that the economic burden of EOL care bankrupts middle income and poor women, whose savings are depleted by aggressive medical expenditures for their spouses whom they survive. The costs are estimated to be upwards of 70% of their total income (196). Similarly, when EOL fiscal burden was examined for both widow and widowers, 63% of income share was found to be consumed in the EOL care of the spouse, leaving the remaining family member in poverty. Advance planning with fiscal conservancy was strongly encouraged (197). A Canadian investigation of inpatient costs at the end of life found that cost trajectories for treatment including ANH were substantial; comparable to the 27% of the U.S. Medicare and 12% total health care budget spent on older adults’ EOL care (198). The costs associated with PEG tube feeding in older adults in the United States were found to be upwards of $32,000 annually per patient in U.S. skilled nursing facilities in the late 1990s (22).

Due to the phenomenon of compression of morbidity and mortality on aging cohorts and survivorship, a time to death modeling schema was used to identify changes in expenditures related to EOL care among persons 65 years of age or older by decade to death. The models point to an increase in non-hospital expenditures, such as skilled nursing and prolonged lag time to death in the oldest old. While the authors acknowledged that advance directives, hospice, palliative care, and decreasing utilization of aggressive measures with advancing age will offset costs, there is still the prospect of inordinate fiscal burden due to the use of technology in EOL care, such as ANH (199). Conservative estimations from the mid 1990s suggest that 30 billion dollars would be saved annually by the institutionalization of hospice and advance directives limiting technologies such as ANH at the end of an older person's life (27).

Published studies have emerged regarding the use of ethics consultations with the dying and their families. Substantial cost savings occurred when the ethics consultation resulted in the removal of ANH, other life support treatments, discontinuation of diagnostics and monitoring procedures, and specialty consultations. The savings were in the hundreds of thousands of dollars per patient (200). Palliative care and hospice consultation provides substantial savings on treatment, staffing, and monitoring, including ANH (201). Integration of palliative care teams into critical care diminished costs associated with ANH, with discontinuation of ANH by 18% of patients who saw palliative care personnel versus those who did not (202).

Falls et al., in analyses of “quality of dying,” found fiscal solvency and cost containment to be important to Americans at the end of life. The U.S. “medical model,” with its “emphasis on the use of technology to cure disease and extend life at all costs” is falling out of favor with the Boomer generation, who want choice, information, and options for death. Palliative care models abound internationally, and the cost savings and options they provide have become increasingly attractive to the U.S. health care consumer (203).

SUMMARY OF ISSUES IN END OF LIFE CARE FOR THE OLDER ADULT

End of life care is controversial and difficult to discuss. Medical paradigms result in the use of technology to extend life for as long as possible through measures such as ANH (204). Quantity of life does not equate to quality of life, and older adults should be informed of their right to request or forgo life-sustaining treatments, such as ANH (205). Older adults should be advised to document their wishes and be educated on concepts of care such as palliation and hospice (206). They should be informed of the natural progression and anesthetic value of starvation and dehydration in the terminal stages as well as the lack of benefit of ANH in many circumstances (207). Older adults should be informed about the risks and burdens of life prolongation through artificial means long before emergencies arise and EOL care is instituted (208). Practitioners must be well informed regarding the legal, ethical, clinical, religious, cultural, personal, and fiscal aspects of ANH in the dying older adult (209). Practitioners must be cognizant of their own biases and make sure that every person is evaluated individually so that a “blanket” approach is not taken (210). Open communication and informed choice between providers, patients, families, surrogates, and third party payers should be the goal for EOL care in older adults (211).

TAKE AWAY POINTS

	Many older persons will not have advance medical directives explicitly documenting their wishes for EOL care, and even if they do, many will not have their wishes honored.
	Many older adults are unaware of the care concepts of palliation, hospice, and home health.
	Many older persons are not aware of the hazards, lack of benefit, or the burden of care with provision of ANH at the end of life, particularly with dementia and terminal compromise.
	Many practitioners as well as the general population are unaware of the complexities, both legal and ethical, in the withholding or withdrawal of ANH, particularly when there is a surrogate decision-maker and/or the person's wishes are not clearly and legally articulated.
	Many practitioners are not well versed in the religious and cultural influences on care choices at the end of life and have little understanding of the effect it has on patients and their families.
	Many practitioners are not cognizant of the fact that their own ethical, religious, spiritual, cultural, and environmental biases influence their provision of information and open communication about EOL care and choices available to their patients.
	Many providers as well as the general population have no knowledge of the costs associated with the provision of ANH and do not consider fiscal responsibility when instituting technological advances in terminal situations.
	Many providers as well as the general population have failed to discriminate between “quality” and “quantity” of life and have lost an appreciation of “death with dignity.”
	Many providers fail to use ethical decision-making strategies that are individualized for each person, with repeated re-evaluation over time. A blanket approach to provision of EOL care is often institutionalized, which is a disservice to patients, families, and society at large.

Additional information

Notes on contributors

Roschelle A. Heuberger

Note from the Editor: Please see the opinion piece that directly follows this review. It is based on a series of questions posed to three experts: Anthony N. Galanos, MA, MD, a nationally recognized physician expert in the palliative care field; Emma C. Neff, JD, an attorney with palliative care training; and Roschelle A. Heuberger, PhD, RD, author of this review.

Notes

^§Uniform Law Commissioners. Uniform Death Determination Act. Adapted from http://www.nccusl.org/nccusl/uniformact_factsheets/uniformacts-fs-udda.asp, accessed June 1, 2010.

†President and Fellows of Harvard College. A Guide to Living Wills and Health Care Proxies (Harvard Special Health Report). Harvard Annual Report. 2008; 1:27.

‡Faymonville ME, Pantke KH, Berre J, et al. Cerebral functions in brain damaged patients: what is meant by coma, vegetative state, minimally conscious state, locked in syndrome and brain death? Der Anaesthetist. 2004;53:1195–2003.

†American Dietetic Association. Position of the American Dietetic Association: ethical and legal issues in nutrition, hydration and feeding. J Am Dietet Assoc. 2008; 108:873–82.

^§Ashby MA, Mendelson D. Gardner; re BVW: Victorian Supreme Court makes landmark Australian ruling on tube feeding. Med J Austr. 2004; 181:442–5.

‡Ashby M, Stofell B. Artificial hydration and alimentation at the end of life: a reply to Craig. J Med Ethics. 1995; 21:135–40.

^¶Bernat JL. Ethical and legal issues in palliative care. Neurol Clinics. 2001; 19:969–87.

^ΩCantor NL. Glucksberg, the putative right to adequate pain relief, and death with dignity. J Health Law. 2001; 34:301–33.

^∏Goodhall L. Tube feeding dilemmas: can artificial nutrition and hydration be legally or ethically withheld or withdrawn? J Adv Nurs. 1997; 25:217–22.

*Mendelson D, Ashby M. The medical provision of hydration and nutrition: two very different outcomes in Victoria and Florida. J Law Med. 2004; 11:282–91.

^∧Orentlicher D, Callahan CM. Feeding tubes, slippery slopes, and physician-assisted suicide. J Legal Med. 2004; 25:389–409.

^¶Bennett-Jacobs B, Taylor C. Seeing artificial hydration and nutrition through an ethical lens. Home Healthcare Nurs. 2005; 23:739–43.

†Becker R. Providing hydration at the end of life: ethics and practice. Nurs Rev. 2008; 104(Suppl. 9):15.

^§Ohlenberg E. We withdrew nutrition—not care. RN. 1996; 59:36–40.

^†Mathes MM. Withholding and withdrawing nutrition and hydration by medical means: ethical perspectives. MEDSURG Nurs. 2001; 10:96–9.

^§Rosner F, Abramson N. Fluids and nutrition: perspectives from Jewish law (Halachah). Southern Med J. 2009; 102:248–50.

‡Adapted from Bryon E, Gastmans C, de Casterle BD. Decision making about artificial feeding in end of life care: literature review. J Adv Nurs. 2008; 63:2–14.

^§Fins JJ, Miller FG, Acres CA, Bacchetta MD, Huzzard LL, Rapkin BD. End-of-life decision-making in the hospital: current practice and future prospects. J Pain Symptom Manage. 1999; 17:6–15.

†Pan CX, Chai E, Farber J. Myths of the high medical costs of old age and dying. Int J Health Serv. 2008; 38:253–75.