In 1997, the Robert Wood Johnson Foundation funded the Last Acts project designed to educate the public, policymakers, and healthcare professionals on issues surrounding end-of-life care. The following precepts of palliative care, developed by the Last Acts Palliative Care Task Force, affirm a vision of better care:

Respecting patient goals, preferences and choices
Comprehensive caring
Utilizing the strengths of interdisciplinary resources
Acknowledging and addressing caregiver concerns
Building systems and mechanisms of support.

PAST TRENDS

Why, at the beginning of the twenty-first century, do we need a grassroots movement to help us learn something as basically human as how to die? We need to relearn death because people today die in a different way from our forebears: we tend to die older, from different causes and in different environments. In 1900 people often died at home, surrounded by family. Physicians routinely comforted the dying and their families. In the past century, medical and public health advances have almost doubled the average life expectancy, from less than 50 years to near 80¹. People who die in old age tend to experience a long period of functional decline before death, and thus require intensive caregiving and well-coordinated medical care. As medical advances allowed us to delay death, we moved death out of the home and into institutions. Today, although most people say they would prefer to die at home, 56% die in hospital and 19% in nursing homes².

Remembering death

In the USA, we became so caught up in our ability to cure disease that our healthcare system forgot that death is inevitable. When Medicare, the federal health insurance programme for older Americans, was enacted in 1965, it was largely intended to reduce the financial burden of episodic, acute, hospital stays on families. In an attempt to remember death and reduce costly inpatient hospital stays at the end of life, Medicare established a hospice benefit in 1983. Hospice was seen as an alternative for use when life-prolonging options had been exhausted. The Medicare hospice benefit is available only to patients whose doctors are willing to certify they have a life expectancy of six months or less, who agree to receive only palliative care and who have a full-time primary caregiver. Despite its good intentions, the hospice benefit does not help everyone; it remembers only certain types of death.

Hospice in the USA works well for people with reasonably predictable diseases, especially end-stage cancer. At least 60% of hospice enrollees have a cancer diagnosis³^,⁴. It is less helpful for people with less predictable diseases, such as heart disease, or diseases with a long period of decline, such as Alzheimer's. Hospice works well for late-stage AIDS patients, but prognostication in AIDS is harder than in cancer.

The recent history of the palliative care movement in the USA has been driven largely by Medicare funding. When Congress approved the Medicare hospice benefit, it dramatically altered how and where hospice care was provided. The number of hospice admissions immediately increased; by 1985, 40% of people dying of cancer were enrolled in hospice. Because the Medicare hospice benefit emphasizes home care, hospices that had been built around an inpatient model changed the way they operated, shifting their emphasis to home care in order to qualify for Medicare reimbursement. As Lukashok writes, ‘By providing a hospice benefit under Medicare, the government has, in effect, defined what a hospice is’³.

Medicare did more than define hospice when it limited hospice eligibility to a certain type of dying patient—a patient with a home, a caregiver and a terminal diagnosis with an easily predicted disease course. The government also determined who would probably receive end-of-life palliative care and who would not. Medicare's hospice eligibility rules have curtailed the development of a broader palliative care movement in the USA, and until very recently few palliative care programmes emphasized treatment of pain and other symptoms in conjunction with attempts to cure disease.

Medicare's hospice eligibility rules have led to a decreased length of hospice stay; many patients do not enter hospice until their final days, when opportunities to reap maximum benefit have been lost. Because of the six-month rule, hospices may hesitate to accept patients until death is clearly imminent to avoid being charged with Medicare fraud or abuse. With many diseases it is impossible for physicians to feel confident predicting death within six months. Similarly, the six-month prognosis and the requirement to stop life-prolonging efforts have the implication of giving up—something that neither doctors nor families are comfortable doing.

The Medicare hospice benefit was a good step towards a public acknowledgment that people die, and that care for the dying is a valuable part of healthcare. We now need to expand hospice to enable more people to receive excellent end-of-life care, and we need to apply the precepts of palliative care to all care, not just care at the end of life. Various clinicians around the USA have begun to accomplish these goals, and some have documented their experiences inspiringly in the monograph Pioneer Programs in Palliative Care: Nine Case Studies⁵.

Public policy, pain relief and physician-assisted suicide

Unfortunately, too few patients receive adequate end-of-life palliative care. The Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments (SUPPORT) documented the type of care dying patients in teaching hospitals preferred, and how closely their wishes were followed. The investigators found that patients' wishes were frequently not followed, and that pain was common. Half the patients able to communicate in the last three days of life said they were in severe pain. These findings demonstrated the need for hospitals and health systems to pursue a higher standard for end-of-life care and motivated many doctors to improve the end-of-life care they provided⁶. The next challenge for the medical profession is to pursue a higher standard for palliative care throughout the life-span.

In the 1990s, the most public image of the ‘death with dignity’ movement was Jack Kevorkian and the debate he ignited over physician-assisted suicide. As a result of this debate, Oregon became the first State to legalize assisted suicide in certain situations. The fact that Oregonians voted twice to legalize physician-assisted suicide indicates that many Americans are not happy with how they perceive their end-of-life options. Yet, to judge by the small number of people who have committed suicide under Oregon's law, public demand for suicide is low⁷^,⁸. The real message for medicine from the physician-assisted-suicide movement is that the public is afraid to die in our hospitals. There is good reason for this⁶.

The moral debate surrounding the passage of Oregon's law has had several positive effects on palliative care in the USA. First, it served as a wake-up call for doctors who had previously ignored the fact that their patients need good care when they are dying, not just when they are recovering. Also, since Oregon's law requires patients and their doctors to take all other steps to relieve suffering before assisted suicide becomes a legal option, the law seems to have forced healthcare professionals to improve the quality of palliative care⁹.

TRENDS IN PALLIATIVE CARE

Most Americans say they would prefer to die at home. But, as The Dartmouth Atlas of Health Care 1999 reports, depending on where they live, anything from 20% to more than 50% of Americans die in hospital. People are more likely to die in hospital in regions well supplied with hospital beds (Figure 1)¹⁰^,¹¹. The Dartmouth Atlas findings show that¹⁰:

Percentage of Medicare enrollee deaths in hospitals (source, Ref. 10)

Among Medicare enrollees, 15% to more than 50% will experience at least one stay in an intensive-care unit during the last six months of life
On average, 11% of Medicare enrollees will spend seven or more days in intensive care during the last six months of life
As many as 30% of Medicare enrollees will be admitted to intensive care during their terminal hospital stay.

We do not know the right number of hospital deaths, and this number may vary depending on the numbers of old people living alone, patterns of illness and financial context¹⁰^,¹¹. However, since we do know that most American deaths occur in institutions, clinicians must provide good palliative care wherever the patient is. The following case reports illustrate why we need positive, realistic, end-of-life options outside of the home.

Case 1

Mrs P is an 88-year-old widow dying of colon cancer. Her grown children live in other States and cannot visit for extended periods of time. For Mrs P a hospital or subacute unit might be a more supportive environment than her home. Even if she could afford a full-time paid caregiver at home, she is unlikely to have as many different types of social and caregiving contacts at home as she would in a structured caregiving setting such as a nursing home or an impatient hospice programme. Unfortunately most nursing homes are not equipped to provide the intensive sophisticated palliative care services that a terminal cancer patient requires; and inpatient palliative care services or hospices can only serve patients with short life expectancy (a few days to a week). In the USA there is no good option for Mrs P under the current system.
Case 2

Ms A is a 29-year-old single mother dying of AIDS. She has valiantly fought her illness and is not ready to die. Yet when she comes into the hospital with her final infection, the intravenous antibiotics cannot save her. She dies in the hospital because she has been receiving curative care that would have been unavailable at home, but she is also able to benefit from palliative services aimed at making sure she is comfortable and helping her family come to terms with her imminent death. Not everyone who dies in a hospital has accepted death; many are still hoping for cure.
Case 3

Sam, age 18, is brought into the intensive-care unit with serious head injuries sustained in a motorcycle accident. It is clear that he will die within a few days. To move Sam from the hospital back to his home, a studio apartment where he lives alone, would be senseless. Sam will die in the hospital because there is no other reasonable option.
Case 4

Mrs M is a woman of 65 with late-stage ovarian cancer. She is eligible for hospice, and this sort of care appeals to her, but her 70-year-old husband cannot cope with her physical needs and she feels uncomfortable having her children see her bloated body and deal with her draining wounds. Instead, she opts to enter a hospital-based palliative care programme so that professionals can care for her physical needs and her family have the time and energy to provide emotional support. The problem with this option is that, if life expectancy is greater than one to two weeks, the hospital is not an appropriate or sustainable option.

People with serious illnesses and their families have diverse needs. To accommodate these needs, clinicians are creating new palliative care delivery models in the USA. Palliative care, including hospice, can be provided in a variety of settings including the hospital, the nursing home and the patient's own home. To learn more about the extent to which these models have been instituted throughout the USA, the Robert Wood Johnson Foundation funded the Center to Advance Palliative Care, based at the Mount Sinai School of Medicine in New York City, to survey more than 2000 hospitals nationwide, asking them if they had palliative care or pain programmes and, if so, what type¹². The percentages for each model described below represent the breakdown of types of programmes within hospitals that had palliative care or pain programmes. They do not add up to 100 because some hospitals have more than one type of programme:

Consultation service (43%)—a team composed of doctors and nurses, typically with a social worker and/or bereavement counsellor, sees patients with palliative care needs anywhere in the hospital
Dedicated inpatient unit (23%)—palliative care beds are clustered in one area of the hospital, concentrating together patients with similar needs. A dedicated unit, which may be combined with the consultative service, provides visibility and may enhance acceptance of palliative care by hospital staff
Combined hospice—palliative care unit (36%)^*—inpatient unit serves both hospice patients and hospital patients with palliative care needs
Community hospice—hospital contract (36%)^*—through a contractual arrangement with a community hospice programme, a hospital provides palliative care in an inpatient unit. This model expands Medicare payment options available to hospitals for these palliative care services, since they are reimbursed (albeit inadequately) through the Medicare hospice benefit
Hospital outpatient palliative care clinic (32%)—clinic operates in conjunction with consultative service or inpatient unit to provide continuity of care after discharge from the hospital. This model promotes care continuity and is often linked to home visiting programmes.

THE FUTURE OF PALLIATIVE CARE: CHALLENGES AND STRATEGIES

Care for the dying in the USA is largely publicly funded: almost three-quarters of those who die each year are over-65 and on Medicare, and many others are insured by a combination of Medicare and Medicaid. Therefore, government policies are of major importance in determining what kind of care people at the end of life receive¹².

However, if our goal is to integrate palliative care into all areas of medicine, not just end-of-life care, reform must extend beyond the way the government pays for healthcare. Current barries to providing adequate palliative care in the USA include: (1) financing mechanisms, (2) inadequate supply of trained physicians, (3) no continuity of care, (4) cultural issues—seeing death as a failure; a discomfort in talking about death—and (5) physician attitudes.

The barriers to palliative care can be surmounted as follows:

Create recognition for a strong board-certified specialty of palliative medicine—this would provide credibility to the field while attracting medical leaders and encouraging the expansion and improved understanding of the existing knowledge base¹³
Improve the way Medicare pays providers and institutions for palliative care—Medicare inadequately pays physicians who offer time-consuming palliative-care consultations. Also, the ‘diagnostic related group’ system for Medicare inpatient hospital payments does not specifically recognize the relief of suffering as a legitimate goal of hospital admission. More research is needed to determine the best way to provide palliative care to all who need it while fairly compensating those institutions that provide it
Relax Medicare's hospice eligibility rules—to make hospice accessible to people dying of any disease, including diseases less predictable than cancer
Educate medical providers—palliative care should be taught at the medical undergraduate and graduate levels. Good palliative care requires high-tech knowledge of pharmacology and symptom management as well as medical ethics, and cannot be learned in one or two lectures
Educate consumers—consumers should expect, demand, and hold health providers accountable for providing high-quality palliative care.

Footnotes

The study grouped these two categories together. 36% of respondents offered one or both of these models of care (combined hospice-palliative care unit or community hospice-hospital contract).

References

1.Rowe J, Kahn RL. Successful Aging. New York: Pantheon Books, 1998
2.Data from the National Center for Health Statistics National Mortality Followback Survey [http://www.cdc.gov/nchs/releases/98facts/93nmfs]
3.Bosanquet N. Patterns of use of service. In: Bosanquet N, Salisbury C, ed. Providing a Palliative Care Service: Towards an Evidence Base. Oxford: Oxford University Press, 1999
4.National Hospice Organization Fact Sheet, 1999 [http://www.nho.org]
5.The Robert Wood Johnson Foundation and the Milbank Memorial Fund. Pioneer Programs in Palliative Care: Nine Case Studies. New York: Milbank Memorial Fund, 2000
6.The SUPPORT Principal Investigators. A controlled trial to improve care for seriously ill hospitalized patients: the Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments (SUPPORT). JAMA 1995;274: 1591-8 [PubMed] [Google Scholar]
7.Chin AE, Hedberg K, Higginson GK, Fleming DW. Legalized physician-assisted suicide in Oregon—the first year's experience. N Engl J Med 1999;340: 577-83 [DOI] [PubMed] [Google Scholar]
8.Sullivan AD, Hedberg K, Fleming DW. Legalized physician-assisted suicide in Oregon—the second year. N Engl J Med 2000;342: 598-604 [DOI] [PubMed] [Google Scholar]
9.Ganzini L, Nelson HD, Schmidt TA, Kraemer DF, Delorit MA, Lee MA. Physicians' experiences with the Oregon Death with Dignity Act. N Engl J Med 2000;342: 557-63 [DOI] [PubMed] [Google Scholar]
10.Wennberg JE, ed. The Dartmouth Atlas of Health Care 1999. Washington, DC: AHA Press, 1999
11.Wennberg J. Which rate is right? N Engl J Med 1986;314: 310-11 [DOI] [PubMed] [Google Scholar]
12.Pan CX, Morrison RS, Meier DE, et al. How prevalent are hospital-based palliative care programs? Status report and future directions. J Pall Med 2001. (in press) [DOI] [PubMed]
13.Field MJ, Cassel CK, eds. Approaching Death: Improving Care at the End of Life. Washington, DC: National Academy Press, 1997 [PubMed]