Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 87
--> 4 The Health Care System and the Dying Patient Why are you afraid? I am the one who is dying!…But please believe me, if you care, you can't go wrong.…Death may get to be routine to you, but it is new to me. Anonymous, American Journal of Nursing, 1970 With help and support from family, friends, and others in the community, many people can live their lives, while dying, with some or considerable independence from the health care system. Many, if not most, however, draw heavily on that system for care—and caring—in the form of clinical services, counseling, and practical assistance with both medical and nonmedical needs. Thus, while the role of family and community resources should be acknowledged and strengthened, it is also essential to understand how care systems serve patients well and poorly and to identify the system characteristics that contribute to poor care. Such understandings, which will depend on better data and research than now exist, will help provide the basis for steps to remove the impediments to good care and to fortify the foundations for reliably excellent care. In general, care systems—both as discrete organizations and as unevenly connected arrays of community institutions and services—require people (supported by facilities and processes) who are prepared to determine what care is appropriate, to arrange its provision, and to monitor performance for consistency with organizational and external norms. Broadly, this means having the capacity to provide or arrange for symptom prevention and relief; attention to emotional and spiritual needs and goals; care for the patient and family as a unit; sensitive communication, goal setting, and advance planning; interdisciplinary care; and services appropriate to the various settings and ways in which people die.
OCR for page 88
--> These process of care elements are, in a sense, statements of expectations for the care system. Most of these elements were discussed in Chapter 3, which emphasized the importance of sympathetic but clear consideration of prognosis and goals and fitting care strategies to circumstances. This chapter considers the major settings of care in which people die and identifies questions about the ways care is structured, provided, and coordinated. It concludes by considering aspirations for an ideal care system and what this implies for the mix of organizations, programs, settings, personnel, procedures, and policies that make up care systems. Unlike new mothers or women undergoing mastectomies, who have recently been the subject of highly publicized criticisms of early discharge, dying patients are not themselves a potent lobbying group and their survivors are often exhausted, grieving, and expected to put their lives back together and move on. Thus, health care professionals, managers, and others have a particular responsibility to press for care systems that people can trust to serve them well as they die. Characterizing Care Systems Trying to present a coherent picture of health care systems as they serve—or fail to serve—those who are dying is not easy. First, the two million people who die each year have both variable and common characteristics and needs. Second, the organizations and personnel that may be involved in end-of-life care are likewise numerous and variable. Nationally, there are roughly 6,000 hospitals, 16,000 nursing homes, 11,000 to 15,000 home health care and hospice agencies, 650,000 generalist and specialist physicians, 2 million nurses, tens of thousands of social workers involved in health care,1 and numerous other categories of health personnel and facilities including several hundred health maintenance organizations (HMOs) and other managed care and health insurance arrangements. Third, data about care at the end of life are very limited. Even the term health care system has no fixed meaning. It can be used in at least four different ways—not because people are being careless in their language but because the term is intrinsically general and capable of applying to several situations. First, the term health care system may be used to describe and analyze a community's or region's array of health care 1 According the U.S. Bureau of Labor Statistics, there were 666,000 "degreed Human Services Workers" in 1995 (Gibelman, 1997), which is the category that includes social workers. The National Association of Social Workers has 160,000 members. The committee did not locate a specific count or estimate of those employed by health-related organizations or otherwise involved in health care rather than other human services.
OCR for page 89
--> organizations and services, whatever their relationships. This is consistent with one dictionary definition of a "system" as a set of objects grouped together for classification or analysis (The American Heritage Dictionary, 3d ed.). In this usage, the health care system in New York City could be characterized as an aggregation of loosely interacting (sometimes cooperating, sometimes competing, sometimes self-absorbed) components within a large, socially and economically complex geographic area. Fragmentation has been cited as the key characteristic of such systems in the United States (Shortell et al., 1996). Although they may be viewed as rather disorderly systems, they are—analytically—still systems rather than nonsystems.2 The deficiencies of community health care systems, in the past, prompted a variety of voluntary and regulatory efforts to plan and control the development of health resources (especially facilities and advanced technologies). In part as a result of their weaknesses and in part as a result of shifting political tides at the community and national level, these sorts of health-planning mechanisms have largely been abandoned in favor of more market-based strategies. In a second and broader sense, the term health care system may also encompass the norms, public policies, and social values that shape the delivery of health care in a community or a society. This is consistent with another dictionary definition of a "system" as the prevailing social order (The American Heritage Dictionary, 3rd ed.) Thus, a reference to the Canadian health care system or the American health care system may signify not just a collection of institutions and personnel but also a culture. A third and much narrower use of the term applies to a particular entity that integrates a comprehensive range of health care services and the facilities and personnel to provide those services (Coddington et al., 1996; Shortell et al., 1996). The integration is formal and institutionalized through explicit controls related to personnel, budgets, and other matters. Systems in this sense may be more or less geographically concentrated, as is the Henry Ford Health System in Michigan, or geographically dispersed, as is the U.S. military health care system, which is not only national but international. 3 Finally, an even narrower use of the term is as a rough synonym for 2 The term nonsystem is, however, useful in directing attention to serious faults in the mechanisms that structure routine or expected transactions among system components or that govern the distribution of resources relative to demand or need. 3 Integration is not an "either/or" characteristic but a continuum. That continuum is clearly evident in the nation's managed care organizations, which range from fairly strongly integrated systems such as Henry Ford to weakly linked entities based on limited and often unstable contractual relationships. A hospice organization is a specialized care system that emphasizes palliative care for terminally ill people and their families and that may—like managed care organizations—be more or less strongly integrated. Medicare hospice regulations favor at least moderate levels of integration.
OCR for page 90
--> a provider organization such as a hospital, nursing home, or hospice program. In this chapter, the discussion tends to focus on community health care systems, but it will be clear that such systems are embedded in a national health care system and include organizational systems as components. The committee uses the term care system to highlight the special role in care for terminally ill patients—and frail individuals more generally—of nonmedical services such as spiritual and bereavement counseling, respite care, and housekeeping assistance. In addition to formal or organized health care systems, informal care systems can also be distinguished; they include the family, religious communities, and folk culture (Kleinman, 1978; Kleinman et al., 1978). These systems play a central role for many if not most patients and families, but they were not the focus of this study. Illustrative Case Histories To illustrate the variability of care for those who are dying, this section presents several more cases. As in Chapter 3, these cases do not exactly describe a single patient or institution or represent statistically typical patients. The cases synthesize committee experiences, cases in the medical literature, research findings, and specific problems reported to the committee. Several describe situations that particularly strain care systems. "Ellen Artbur" This case, synthesized from the hospice literature and committee experiences, illustrates a patient and family for whom a limited range of hospice services work well and whose personal circumstances support living well while dying. Ellen Arthur, a 78-year-old retired teacher diagnosed with kidney cancer, underwent surgery and a trial of experimental chemoimmunotherapy. Tests then showed that the cancer had spread to her lungs and bone. She was incurably ill but not imminently dying and could reasonably be expected to live for another year, perhaps two. She felt fairly well except for mild fatigue. She and her husband were financially comfortable, well educated, and surrounded by supportive family and friends. They reviewed the durable power of attorney and related documents that they had prepared many years ago. After several months of fairly normal activity, pain and weakness began to require an increasing amount of medication. Ellen Arthur's physician concluded that she could very well die within the next six months—probably less—and certified this so that she qualified for the Medicare hospice benefit. Her physician coordinated care with the hospice medical
OCR for page 91
--> director who, in turn, worked with the patient, her family, and an interdisciplinary team to implement the care plan as initially designed and later adjusted as the illness progressed. A hospice nurse was the center of the care team, which focused on two primary problems—fluctuating pain and fatigue. Protocols allowed the nurse to adjust pain medications within defined boundaries, and she advised ways to prevent or soothe other symptoms. For example, she suggested ice chips and glycerin swabs to ease dry mouth and advised balancing rest and activity to reduce the burden of fatigue. The family needed little other direct service from the hospice. Physical therapy and other medical services were not indicated, and the family found emotional and spiritual comfort in their friends and their faith and in reviewing their life together. Friends also pitched in to provide occasional practical help with meals, errands, cleaning the house, and respite time. John Arthur was informed about how to recognize changes in his wife's condition, especially signs that death was imminent. He knew that if something happened that he could not handle—seizures, for example—he could call the hospice any time, day or night, and help would be sent. The Arthurs were informed that if they called 911 in such an emergency, the protocol for paramedics in their jurisdiction required attempts at resuscitation and other interventions and transport to the hospital. Ellen Arthur died at home with her husband at her side. "Solomon Katz" This example, adapted from a case analyzed in The New England Journal of Medicine (Morrison et al., 1996) illustrates aggressive medical culture, non-beneficial and even inhumane technical interventions, and disregard for patient and family concerns and preferences. It is marked by fragmented services and failure to provide appropriate referrals to hospital and community resources. No one had overall responsibility for this patient's care and well-being. Solomon Katz, a 75-year-old retired postal worker, experienced progressive weakness in his left leg for over a year, but he sought care in the emergency department only after he fell and could not get up by himself. After an array of tests, he was diagnosed with lung cancer metastatic to the brain. He accepted medication but refused a lung biopsy. Hospital staff described him as "in denial," but a psychiatric consultant assessed him as reacting appropriately and capable of making his own decisions. He revealed that his wife died of cancer two years earlier; he did not want to go through the treatments she had. After continued pressure from the oncology team, he agreed to a biopsy, which confirmed the previous diagnosis. They offered options for life-prolonging but invasive treatment; he declined. He was discharged home with home care and follow-up appoint-
OCR for page 92
--> ments at two clinics. The medical staff did not refer him to the hospital's social work staff, and no one sat down with Solomon Katz and his family to talk with them about the prognosis, palliative care options, or their concerns and preferences. He did not keep his follow-up appointments, but no one checked to see why. Three months later, Mr. Katz was again brought to the emergency department following grand mal seizures. He was lethargic and could not talk; a CT scan showed progression of the brain tumor, including partial brain stem herniation. He was started on intravenous medications and fluids, and then oxygen and nasogastric tube feedings. The neurology team wanted to resect the brain tumor. His son, a grocery store clerk, refused surgery and asked for a do-not-resuscitate order on the basis of his father's previously expressed wishes. During the next three weeks, Mr. Katz was minimally responsive but repeatedly removed the nasogastric tube despite restraints. After pressure from the hospital staff, the son agreed to insertion of a percutaneous gastrotomy tube. On the following day, Solomon Katz died of a cardiac arrest with no family present. The son was upset by the entire experience, particularly when he later learned that his father's dying could have been a less brutal experience and that his care could have been managed differently with an emphasis on understanding his goals as his life ended and providing physical and emotional comfort. The case attracted the attention of the hospital ethics committee, which concluded such clearly inappropriate care indicated serious system problems. It began to mobilize an institution-wide effort at self-examination, staff education, process changes, and quality measurement and improvement. "Dorothy Chang" The next case presents a middle-age woman who died in an intensive care unit following a cascade of complications related to a bone marrow transplant and other medical problems. When it was time for life support to be ended, a protocol was followed that emphasized patient comfort and attention to the family. Dorothy Chang was a 54-year-old woman with aplastic anemia who received a bone marrow transplant in August. She tolerated the procedure well, but her discharge from the hospital was delayed due to recurrent bacterial infections. Her infections resulted from a narrowed esophagus. An attempted corrective procedure was complicated by rupture of the esophagus and a life-threatening infection that required placement of a drainage tube in the chest and intravenous antibiotics in the ICU. She remained intubated with recurring pneumonia. The pneumonia slowly improved, but a trial with removal of the ventilator failed. Significant muscular weakness developed, exacerbated by
OCR for page 93
--> chronic steroid use. A tracheostomy was performed, and a feeding tube was placed with some difficulty. In early November, a head CT scan revealed a lesion consistent with a fungal brain abscess. Despite treatment, another CT revealed advancing disease, and a brain biopsy suggested Aspergillus infection, which occurs mainly in immune-compromised people and is very resistant to treatment. Ms. Chang's condition declined further. She became minimally responsive, and her husband agreed with her physicians that further life-supporting treatment would only prolong death. With support from the hospital's comfort care team, Ms. Chang was moved to a more private area where her family could be with her when the ventilator was removed by an experienced physician known to the family. After all monitors and other supports were removed except an intravenous morphine drip, she was given intravenous morphine and midazolam to ensure that she would not experience distress. After the ventilator was disconnected, her breathing became irregular and stopped, and then her heartbeats began to pause before ceasing. Her family remained quietly with her for a period after death was declared. Following this patient's death after more than four months in the ICU, the ICU director decided that this case needed a careful review to identify any flaws in the unit's performance and areas for improvement. The initial assessment was that both the intensive treatment and its eventual withdrawal were appropriate for the patient's particular combination of problems but that the decision for esophageal dilation should be reviewed with gastroenterologists. "Millie Morrisey" The problems of appropriate end-of-life care for nursing home patients are increasingly being recognized. This case illustrates both inadequate expertise in palliative care and faulty organizational procedures for assuring that patient and family wishes are respected. Millie Morrisey, an 86-year-old retired clerical worker, suffered from Alzheimer's disease and congestive heart failure. Before her illness became so severe, she and her niece had discussed her future, and she prepared a living will that expressed her wishes, including that resuscitation, mechanical ventilation, and other life-prolonging measures not be attempted in a medical crisis. She did not want to burden her niece, and she accepted that nursing home care would be the best option when she could no longer safely be left by herself, which happened about a year later. After two years of nursing home care, Millie Morrisey developed pneumonia and respiratory distress. As had been its custom, the nursing home staff called an ambulance. Documentation of preferences—which included foregoing antibiotics, mechanical ventilation, and tube feeding—were nei-
OCR for page 94
--> ther sought nor readily accessible in her chart. The hospital quickly admitted her to the ICU, initiated antibiotics and mechanical ventilation, and used physical restraints so that she would not dislodge the tubes and lines to which she was connected. She mostly did not seem to understand what was happening and was agitated and tearful. After three days during which her niece informed the ICU team of her aunt's wishes, Millie Morrisey was successfully withdrawn from the ventilator and died a few days later, receiving comfort care only. The niece was still upset, and the nursing home staff concluded they needed to avoid similar problems in the future. The nursing director learned that she could participate in a new statewide working group comprised of many health care, social service, and religious organizations that were attempting to devise clinical and administrative guidelines to understand and honor patient and family goals and to assure excellent end-of-life care. "Darrell Henson" This case demonstrates the involvement of nontraditional family. It also underscores the role that poor quality care can play in turning a patient's attention to suicide. Darrell Henson, a 67-year-old owner of a small business, was diagnosed with prostate cancer. He had surgery and remained in remission for two years. Around Thanksgiving, the cancer reappeared in bone metastases, and Mr. Henson began to suffer terribly from pain and nausea. The medications prescribed by his physician did little to help, but the physician feared being accused of overprescribing narcotics and was reluctant to do more. Mr. Henson begged his companion, Charles Jones, to help him die because life seemed unendurable with such pain. Mr. Jones, distressed by his companion's suicidal thoughts, called another physician, one who had cared for his mother. This doctor visited Mr. Henson at home, did a careful assessment of his pain and other symptoms, and developed a care plan that emphasized effective pain relief and supportive home services. Although not a conventionally religious man, Mr. Henson was relieved not to violate the prohibitions against suicide taught to him in childhood. He was able to remain home, celebrate Christmas with his companion and their friends, and—despite weakness—say his goodbyes before dying in January. "George Lincoln" This case illustrates the disruptions that may be caused by health care restructuring and consolidation. George Lincoln was an 84-year-old retired engineer whose health gradually deteriorated as multiple organ systems began to fail. Through his former
OCR for page 95
--> employer's health plan, he had long been enrolled in a large HMO. He had been pleased with the HMO, which arranged much appreciated hospice care when his wife died of breast cancer several years earlier. Then, however, the HMO was sold to a corporation with out-of-state headquarters, his personal physician was not on the new provider list, many of his numerous medications were switched to lower-cost drugs, and referral to a cardiologist experienced with very old patients proved difficult. Mostly confined to his home (and often to his bed) and experiencing considerable physical distress but still mentally alert, Mr. Lincoln concluded—with considerable equanimity—that "my time is coming." He raised the issue of hospice care with his new primary care physician. He discovered that the HMO's long standing relationship with a community hospice had been terminated in favor of referrals to a fairly new organization. Mr. Lincoln was not told that as a Medicare beneficiary he was not limited to this hospice, and he found the new hospice provided a much-reduced level of care—and caring. He was, in any case, too fatigued and depressed to complain. The cases above and in Chapter 3 highlight some of the positive and negative features of the care systems available to dying patients and those close to them. They vary in the degree to which care systems help the process of dying be, as described in Chapter 1, free from avoidable distress and suffering for patients, families, and caregivers; in general accord with patients' and families' wishes; and reasonably consistent with clinical, cultural, and ethical standards. Some examples were constructed to show patients who suffered reasonably avoidable distress; some involved disregard for patient and family wishes and perhaps violated clinical or ethical standards; several showed people and institutions attempting to cope with changing expectations or to correct problems. The sections below consider different settings of care from a more general analytical perspective. Settings For End-of-Life Care As described in Chapter 2, people die in many settings—mainly hospitals, nursing homes, and homes. Each setting is appropriate and feasible under some circumstances and not under others. Not only are home, hospital, and nursing home settings quite different from each other, each category encompasses great variability. This discussion starts with the hospital because it is still the most frequent site for end-of-life care, despite increased pressure from various sources to shift care to other settings. Later sections consider nursing homes, home care as part of a hospice program, home care without hospice, and coordination of care within and across settings. Transitions from one setting to another create many problems as the weak or even nonexistent links among health system components are exposed. Con-
OCR for page 96
--> sistent with other chapters, one problem in drafting this chapter was the relative lack of descriptive and evaluative data on the experience of dying patients and the effects of different settings and processes of care. Hospitals Although the situation is changing, more deaths occur in hospitals than in other settings, and most people receive some hospital care—brief or extended—after being diagnosed with serious progressive illness. Nonetheless, curing disease and prolonging life are the central missions of these institutions. Hospital culture often regards death as a failure, in part because modern medicine has been so successful in rescuing, stabilizing, or curing people with serious medical problems and in part because a significant minority of acutely ill and injured patients who die often do so before the end of a normal life span. A major challenge in hospital care is identifying better ways to maintain vigorous efforts to cure illness and prolong life while, at a minimum, avoiding needless physical and emotional harm to patients who die. Every hospital's goal should be to provide as good a death for the patient and family as a patient's illness and circumstances permit. Shortcomings in hospital care have been documented in a number of areas. A large study that attempted both to understand hospital care of patients with advanced illnesses and to improve that care identified opportunities for improvements in patient-physician communication, understanding and honoring of patient preferences, and symptom management (SUPPORT Principal Investigators, 1995). It also found considerable amounts of intensive interventions in the days before death, a finding that the investigators found troubling but that some critics argued might be appropriate and consistent with patient's wishes (Emanuel, 1995b; Lo, 1995). The study also found that the strategies designed to improve care did not appear to have modified behavior or outcomes Other studies, cited in Chapters 3 and 5, have found problems related to unrelieved symptoms in hospitalized patients, poor documentation of patient preferences for care, early discharge of dying patients, and underuse of home care. As discussed further in Chapter 6, the financial incentives for hospitals to discharge patients early warrant reexamination in the context of end-of-life care. The growth of hospice combined with pressures to minimize hospital use is shifting more care to other settings, often the home, sometimes a nursing home. This shift can be expected to continue as physicians, nurses, social workers and others working with hospitals, hospices, nursing homes, and professional and community education programs help dying patients
OCR for page 97
--> and their caregivers understand how to cope with the period of dying and the moment of death without reflexively turning to hospitalization. The committee identified a number of basic issues and questions that will be generally relevant for hospitals caring for gravely ill and dying patients. They are offered in part as a guide for hospitals concerned about understanding and improving care and in part as a guide to the formulation of a research agenda. For the most part, the issues and questions outlined below translate into matters of staffing and procedures involving personnel with appropriate attitudes, skills, and knowledge; in some cases, they involve physical design of facilities. How well are patients and families informed about who is responsible for care, what they can expect, and who they can look to for information and assistance? How specifically are the preferences and circumstances of patients and families determined, assessed, recorded, and accommodated? Are processes focused narrowly on written orders or more broadly on advance care planning? What provisions are made for children, non-English speakers, cultural minorities, and others who may not fit routine administrative and clinical procedures? What internal or external expertise is available to help clinicians, patients, and families with clinical evaluation, symptom prevention and management, advance care planning, and decisions about foregoing or withdrawing treatments, bereavement support, and other matters? How is this expertise organized and shared? Are other staff routinely educated about the availability of this expertise and expected to use it when appropriate? Have hospital personnel been trained in methods of communicating bad news, discerning patient and family wishes and concerns, and respecting dignity through their language and other behavior? What structures and processes—for example, ethics committees—are in place to prevent, moderate, or mediate conflicts among clinicians, families, payers, and others? What care options are available for dying patients within the hospital, for example, a designated area of palliative care beds governed by different rules regarding visiting hours and other matters for dying patients? Do explicit criteria and processes guide decisions about the use of these options, and are relevant staff throughout the hospital aware of them? Can hospital procedures and the physical environment be modified in other ways (e.g., in the ICU) to reduce stress and discomfort for patients and families? Are ICU staff, specialty services, and other personnel trained to recognize patients (and families) for whom the goals of curative or life-prolonging care should be reconsidered with particular attention to the goals of physical and emotional comfort and symptom relief? Are proce-
OCR for page 111
--> Some hospices depend heavily on volunteers or contract personnel instead of employed staff. Volunteers serve in many capacities and are generally screened to match their talents and interests to organizational and patient needs. Some provide the key clinical services while others provide emotional and practical support to patients and families or perform administrative and informational tasks for the hospice. A hospice patient may stay under the care of his or her personal physician, and many patients and physicians prefer such continuity. Alternatively, the hospice medical director may serve as the patient's attending physician. In the committee's experience, it is sometimes difficult for the personal physician to stay involved if house calls are required. Also, some otherwise excellent physicians may not be comfortable with an established hospice team, and others may not be fully familiar with the principles and techniques of palliative care, including appropriate use of opioids. Even if a patient is transferred to a hospice-based physician for good reasons, the personal physician can stay involved to provide support and information that may be useful for the hospice team. Many patients and families who have had a long-standing relationship with a physician undoubtedly would appreciate further communication, such as a call when death is near and a call or letter afterwards. Effective teamwork requires active and continuing effort by all team members and by those responsible for overseeing their performance. For palliative care teams, the challenges are many (Ajemian, 1993). Team members come from very different professional backgrounds, which creates the potential for role conflict. Sorting out and maintaining clear responsibilities for decisionmaking, information provision, and other tasks is important if people are to know what is expected of them. Even so, conflict management may still be necessary. Part-time staff and volunteers, however important, can be awkward to fit with an established team, and relationships with others who have cared or are caring for the patient can become strained. For example, if hospice personnel are involved with the care of a long-term nursing home patient, staff at the nursing home may have established a close relationship with that patient and may be upset if their care for the patient is ignored. Coordination is a pervasive and continuing challenge that requires a balance between the time required for coordinating activities and the time devoted to the activities themselves (e.g., caregiving, documentation). Inpatient Palliative Care Teams Clearly identifiable inpatient palliative care teams have not been routine for nursing homes or hospitals. Medicare-certified hospices must have an interdisciplinary palliative care team (HCFA, 1996a). The Joint Com-
OCR for page 112
--> mission on Accreditation of Healthcare Organization's standards for nursing homes call for an interdisciplinary team whose responsibilities include care planning and, by implication, palliative care for those needing pain management and similar care (JCAHO, 1996a). For hospitals, Joint Commission standards refer generally to teamwork and interdisciplinary collaboration but do not explicitly call for an interdisciplinary team in care for dying patients (JCAHO, 1996b). It is the committee's experience that many hospitals and nursing homes have no personnel with clearly designated palliative care expertise. Hospital ethics committees may help resolve conflicts among patients, families, and clinicians about care goals for gravely ill patients, but this is not equivalent to expertise in symptom prevention and relief or in other aspects of palliative care. Depending on an institution's objectives and characteristics, the options for developing palliative care expertise include interdisciplinary care and consulting teams or designated individuals who may be called upon to provide or assist with palliative care. The care team may be the option best suited for the varied and complex needs of seriously ill, hospitalized patients, whereas designated personnel and care protocols may be reasonable for nursing homes that can call on outside experts for consultation about more difficult situations including those that might otherwise appear to call for hospitalization. In some cases, members of a designated hospital-based palliative care team may function less as regular caregivers than as consultants, particularly if the hospital does not have an inpatient palliative care unit (Abrahm et al., 1996). Such a team would typically function in an environment in which a variety of curative, palliative, rehabilitative, and other services are provided and in which very sick patients with many different diagnoses and problems are treated. The members of the team would help physicians and nurses care for patients with refractory pain or other especially difficult problems in symptom prevention and management. In an academic medical center, this kind of palliative care team might consist of one or more physicians, specialist nurses, social workers, pharmacists, and chaplains. In some environments, other approaches may make sense. For example, one academic health center that served mainly as an emergency and trauma facility for a disadvantaged inner-city population established a supportive care team for hopelessly ill patients (Carlson et al., 1988; Campbell and Frank, 1997). In this case, the team consisted of a clinical nurse specialist and a small, rotating group of physicians. The availability of a palliative care team should not imply that those routinely caring for dying patients can do without the clinical and interpersonal skills and attitudes needed to prevent and relieve common causes of
OCR for page 113
--> distress for dying patients and those close to them. Nurses are central in this regard. Given the current medical culture, the organization of a palliative care team may reinforce the role of ethics committees and others in helping hospital staff recognize when aggressive diagnostic or life-prolonging interventions are inappropriate and how to communicate sensitively with patients and families to help them focus on what can be done. Members of such a team may also play an important part in undergraduate and graduate medical education and in research to build a stronger knowledge base for end-of-life care. They may provide consulting services to hospice care teams, nursing home staff, and others faced with particularly difficult clinical problems. Educational and research roles of palliative care specialists are discussed further in Chapter 8. Managed Care as a Coordinating Strategy As indicated earlier, this report uses the term managed care broadly to refer to organizations that direct patients to a limited panel of health care providers, pay providers in ways that encourage economical use of services, and/or employ administrative and educational mechanisms to control access to expensive services. Although not usually counted among such organizations, hospice is a form of managed care applied to certain terminally ill patients. Conventional HMOs and managed care arrangements have not yet enrolled a large proportion of the Medicare population overall.6 Nationwide, only 10 percent of Medicare beneficiaries are enrolled in HMOS. However, in California, Oregon, Arizona, and Hawaii, the figure is over 30 percent, and in some Oregon counties, it is over 40 percent (Zarabozo et al., 1996). Although Medicare enrollment in HMOs under represents those most likely to die (e.g., the oldest old and functionally impaired), an increasing number of older people can be expected to be enrolled in managed care when they die (Zarabozo et al., 1996). A portion of these individuals will, however, be cared for and die in nursing homes, and much of this care will be paid for by private funds or Medicaid. Although state Medicaid programs may be considering managed care for poor, functionally impaired recipients, its feasibility and benefits for most nursing home patients is not clear. The pros and cons of using conventional managed care to cover people with serious chronic illness have been debated, although little attention has 6 The situation for younger adults and children who are eligible for Medicaid is quite different. States are increasingly shifting these beneficiaries to some form of managed care.
OCR for page 114
--> focused on those considered terminally ill (Schlesinger and Mechanic, 1993; Miles et al., 1995; Jones, 1996a,b; Miles, 1996; Paris, 1996; Post, 1996; Spielman, 1996; Virnig et al., 1996). The federal government has funded a number of social health maintenance organization (SHMO) and other demonstration projects specially designed to improve coordination of care and quality of life for frail elderly people (Vladeck et al., 1993). Research suggests that such plans are more likely to use case managers to monitor patient needs than conventional managed care plans, and they tend to provide somewhat different patterns of care for patients compared with fee-for-service settings (Harrington and Newcomer, 1991). Regardless of the pros and cons of SHMOs, they are not now available to most Medicare beneficiaries with serious chronic illnesses, including terminal illnesses. Several HMOs, for example, several Kaiser plans, have long operated their own hospices, but most are thought to contract with hospices, as is likely the case for other organizations such as risk-bearing medical groups or integrated health systems. Some managed care organizations have established arrangements for managing the care of seriously ill individuals that closely parallel hospice programs in some respects but that may be more flexible (e.g., not explicitly limited to patients expected to live six months or less) (Blackman, 1995b; Baines, Gendron, et al., 1996; Della Penna, 1996; Emma, 1996). These arrangements may be viewed by organizations as sufficiently valuable on the bases of cost, quality, and patient satisfaction that they are included in their standard benefit package rather than priced separately as an additional benefit. In comparison to traditional fee-for-service medicine, several positive features have been attributed to managed care (most often cohesive, integrated health plans such as well-established staff- or group-model HMOs). These features—which, if implemented effectively, parallel the model of palliative care—include7 more attention to preventing medical problems, greater continuity of care, reduced levels of inappropriate treatment, better use of health care teams coordinated by primary care clinicians, and coverage of services not included in traditional health insurance. The usual criticisms of managed care—particularly for people with chronic, progressive, or advanced illnesses—are that 7 For further discussion of potential advantages and disadvantages of managed care, see, for example, Brown, 1983; IOM, 1993a, 1996b, 1997; Miles et al., 1995; Jones, 1996b.
OCR for page 115
--> managed care plans have been designed primarily to serve relatively healthy and younger populations; they tend to attract or keep healthier individuals, while those who are less healthy stay with or move back to programs such as traditional Medicare; the composition of physician panels and use of other practitioners reflect the composition of the enrolled population and the plan design and thus may not match the needs of chronically ill patients; current capitated payment methodologies and a competitive marketplace make it very much in the financial interest of health plans to avoid enrolling patients with serious chronic illness; and these same financial incentives encourage underservice. The information available to the committee primarily involves HMOs and not other forms of managed care, and very little is known specifically about care for those who are dying. Evidence of underservice or poorer outcomes is mostly speculative or anecdotal. A few studies do, however, reinforce concerns about the effect of some managed care arrangements on people with serious chronic illness (see, e.g., Shaughnessy et al., 1994; Ware et al., 1996a, b; but also Brown et al., 1993). A recent study of a sample of frail elderly Medicare patients in a single integrated health system with both HMO and fee-for-service (FFS) patients concluded that the "HMO studied here served frail enrollees in ways that increased rather than reduced total and preventable hospital readmissions" compared to FFS beneficiaries (Experton, Li, et al., 1996, p. 1). The same researchers also found, in a related analysis, that about the same proportion of HMO and FFS patients used some home health services but that the HMO members who used home health care services received significantly fewer services than FFS members (Experton et al., 1997). The researchers suggested that the HMO should investigate premature hospital discharges and access to post-acute care" (Experton, Ozminkowski, et al., 1996). The study found no difference in total hospital days, emergency room use, or use of skilled nursing or rehabilitation services. Again, these findings are consistent with other research, which suggests possible problems in the treatment of people with serious chronic illness who need further attention from managed care plans, policymakers, and researchers. In Minnesota, Miles and colleagues indicate that health plans ration the use of hospice care, visiting nurse care, respite care, and spiritual and psychological counseling for end-of-life care (Miles et al., 1995). He notes that such practices are "not unique to end-of-life care" but that their application to those approaching death "promises to be a hot spot" (Miles et al., 1995, p. 304). More generally, the committee heard some concerns from hospice or-
OCR for page 116
--> ganizations about the practices (actual or feared) of managed care organizations. One concern expressed to the committee involved the "micromanagement" by some HMOs of care for non-Medicare patients enrolled in hospice (Murphy, 1996; Tehan, 1996). Practices mentioned included denial of certification for appropriately intensive home care or for appropriate hospitalization for palliative care. Such practices have quality as well as financial implications. They may reflect that newer managed care organizations serving mainly younger patients are unfamiliar with hospice. Thus, just as those in fee-for-service medicine may need education about palliative care and hospice programs, so may those in managed care. Revisiting the Care System at the Community and National Levels The concerns raised earlier in this chapter and the promise of initiatives such as those described in this report led the committee to consider characteristics of community care systems that would more effectively and reliably serve dying patients and their families. Such systems would offer comprehensive community health care and other resources that would permit greater integration of care across settings and more flexibility for patients, families, and clinicians. As depicted in Box 4.1, "whole-community" approaches to end-of-life care would draw on the medical and nonmedical approaches to care described in Chapter 3. They would also accommodate patients with unusual conditions and circumstances as well as those with more typical situations. The aspirations of whole-community approaches to care at the end of life include: making palliative care available wherever and whenever the dying patient is cared for; promoting timely referral to hospice for patients whose medical problems and personal circumstances make hospice care feasible and desirable; allowing more flexible arrangements for patients whose prognosis is less predictable but for whom palliative services would be valuable in preventing and relieving symptoms and discouraging unwanted or inappropriate life-prolonging interventions; encouraging, when possible, the continued involvement of personal physicians after patients' referral to hospice; promoting fuller and more sensitive discussion of diagnosis, prognosis, and care objectives; and, by taking these steps, helping dying patients and their families to plan ahead and prepare for dying and death.
OCR for page 117
--> BOX 4.1 A Whole-Community Model for Care at the End of Life Programs and settings of care suited to the needs and circumstances of different kinds of dying patients • Home hospice programs • Other palliative care arrangements for patients who do not fit the home care model —Day programs in hospitals and nursing homes, similar to those developed by geriatricians —"Step-down" arrangements including nursing homes that permit a less intensive and less expensive level of inpatient care when appropriate —Specialized inpatient palliative care beds for those with severe symptoms that cannot be well managed elsewhere —Respite programs to relieve families of patients with a long dying trajectory (e.g., those with Alzheimer's disease) that imposes major physical and emotional burdens on families Personnel, protocols, and other mechanisms that support high-quality, efficient, timely, and coordinated care • Practical and valid assessment instruments and practice guidelines for patient evaluation and management that can be applied at both the individual and organizational level • Protocols for evaluating patient's need for referral or transfer to other individual or organizational caregivers • Procedures for implementing patient transitions in ways that encourage continuity of care, respect patient and family preferences and comfort, and assure the transfer of necessary patient information • Consulting and crisis teams that extend and intensify efforts to allow patients to remain home despite difficult medical problems or crises • Ongoing professional education programs fitted to the varying needs of all clinicians who care for dying patients • Performance monitoring and improvement programs intended to identify and correct problems and to improve the average quality of care Public and private policies, practices, and attitudes that help organizations and individuals • Provider payment, coverage, and oversight policies that, at a minimum, do not restrict access to appropriate, timely palliative care and, as a goal, promote it • Support systems provided through workplaces, religious congregations, and other institutions to ease the emotional, financial, and practical burdens experienced by dying patients and their families • Public education programs that aim to improve general awareness, to encourage advance care planning, and to provide specific information at the time of need about resources for physical, emotional, spiritual, and practical caring at the end of life
OCR for page 118
--> The component resources of an ideal system to serve these objectives would include a mix of organizations, programs, settings, personnel, procedures, and policies. A system with these components would reflect the understanding that there is not just one way to care for dying patients. The people, tools, and other resources would need to be available in many settings and be flexible enough to respond to patients who do not comfortably fit the routines and standards that serve most patients well. Clearly, the elements of such a whole community system represent an aspiration. The difficulties faced by those seeking to improve care at the end of life will vary from community to community, and some elements may be easier to achieve than others. Overall, however, this model implies effort on multiple levels—within individual health care and other organizations and government agencies and through cooperative community, regional, and national initiatives. Conclusion Desirable and obtainable care for those approaching death is determined by both individual and system characteristics and their interactions. The individual's disease, clinical status, emotional state, and preferences determine much about what clinical care is appropriate. Personal values, financial circumstances, family structure, and other patient or family characteristics set limits on what is possible or, at least, what is more or less difficult to accomplish. Over these individual conditions, clinicians and health care organizations typically superimpose a template or protocol that guides but need not rigidly dictate care. The nature of this template or protocol may, in turn, be partly determined by such factors as medical culture, locale (e.g., urban or rural), supply of health care resources (e.g., hospital beds, physicians, organized hospices), delivery system integration, organizational mission and leadership, community norms, statutory requirements, research linking processes of care with outcomes, and internal and external mechanisms for monitoring and improving the quality and outcomes of care. The committee believes that many problems with care at the end of life reflect disincentives for good care and that strategies to improve care will need to focus on change at the system rather than the individual level. As noted in Chapter 3, in reaction against medical and technological overreaching, efforts to improve care systems for the dying share certain emphases with more general efforts to make health care more patient-centered and more concerned about patients' quality of life, not just their physiology. Although it is very much concerned with systems of care, the patient-centered care perspective tends to emphasize what is desirable and appropriate for the individual patient. Quality improvement strategies tend
OCR for page 119
--> to be more population-oriented in their emphasis on institutional goals, priority setting, and statistical analyses. Both have emerged in an environment even more strikingly characterized by efforts to cut the cost of health care under the rubric of managed care. Overall, patient-centered, quality-improving, and cost-driven strategies for care of those with advanced or terminal illnesses may complement each other; they also may conflict. Chapter 5 considers ways of identifying how well care systems serve patients and those close to them. Chapters 6 and 7 consider some of the economic and legal factors that may support or impede good outcomes. Addendum Physicians' Orders for Life-Sustaining Treatment (Used with permission of the Center for Ethics in Health Care, Oregon Health Sciences University) Note: The form as actually used is bright pink for visibility and tamper resistance.
OCR for page 120
OCR for page 121
Representative terms from entire chapter: