Correspondence

Hospital Care for Dying Patients

N Engl J Med 1996; 335:1765-1767December 5, 1996

Article

To the Editor:

The proposal by Miller and Fins (June 27 issue)1 that hospital care be restructured, with more explicit attention to dying patients, highlights an important and unmet need within our health care system. The quality of hospice care has been excellent for patients who qualify for the Medicare hospice benefit, require extensive informal support, and have a fairly predictable prognosis, but the majority of people in the United States still die in hospitals or nursing homes. Although I agree that we need to restructure hospital care for dying patients, I see no particular need to link that care to the intensive care unit (ICU), as the authors suggest. In fact, many patients who should be eligible for high-quality palliative care in the hospital are patients who are not candidates for the ICU. In addition, since palliative services are needed for patients throughout the hospital, a separate unit must collaborate closely with physicians and other clinical staff from all the other clinical departments.

It is important that this discussion receive the highest level of attention from both hospital administrators and health care professionals. One recent development that will aid this process enormously is a new code in the International Classification of Diseases, 9th Revision, Clinical Modification, which was announced by the Health Care Financing Administration (HCFA) in the May 31 Federal Register and went into effect on October 1, 1996.2 Designated the V code, it is for patients who receive terminal or palliative care related to their primary diagnoses. Hospital-chart abstracters must link the V code to a principal diagnosis indicated on the hospital coding sheet. The HCFA intends to study the use of this code for one year, then formulate an appropriate diagnosis-related group (DRG) for payment related to terminal care, in conjunction with existing diagnostic codes. This strategy will allow variability in DRGs according to diagnosis but will add important supplemental information. The code can identify a set of resources expended under palliative care regardless of the diagnosis, making it possible to monitor the quality of care and the extent of its use better.

Advances in clinical research and medical practice have made it possible for physicians to record orders associated with appropriate palliative care clearly enough to be recognized by coders who abstract medical records for billing and quality-assurance purposes. In addition, the existence of such a code and the possibility that hospital payment can be linked to the documentation of palliative care will encourage physicians to be more explicit in documenting plans for the care of patients approaching death. This, in turn, may lead to changes that will help overcome the reluctance of physicians to acknowledge the likelihood of a patient's death and to discuss these issues more fully with family members. These changes may also make it possible for institutions to provide more appropriate care. For the HCFA to have the necessary data to generate an appropriate DRG, it will be necessary for physicians and medical-record abstracters to be aware of the importance of this code and learn to use it appropriately.

Christine K. Cassel, M.D.
Mount Sinai School of Medicine, New York, NY 10029-6574

2 References

1
Miller FG, Fins JJ. A proposal to restructure hospital care for dying patients. N Engl J Med 1996;334:1740-1742
Full Text | Web of Science | Medline

2
Cassel CK, Vladeck BC. ICD-9 code for palliative or terminal care. N Engl J Med 1996;335:1232-1234
Full Text | Web of Science | Medline

To the Editor:

Unfortunately, establishing a new unit does not address the core cultural issues that stand in the way of improved care for the dying. The overreliance on technology; a “never say die” attitude by patients, families, and physicians; and the fear of personally confronting the issues of death — these are the core social and cultural issues standing in the way of better care for the dying. Although a noble idea, designating a special unit within the acute care hospital will in no way eliminate these fundamental barriers.

David E. Weissman, M.D.
Medical College of Wisconsin, Milwaukee, WI 53226-3596

To the Editor:

At Northwestern Memorial Hospital, the principal teaching hospital of the Northwestern University Medical School, we have had a 12-bed acute hospice and palliative care unit for 10 years.1 We work closely with patients and their attending and consulting physicians on medical, surgical, and specialty ICUs, the recovery room, and the emergency department on a daily basis. Patients are seen by a consultation team and may be transferred to our hospice–palliative care unit for closer attention to symptom control and psychosocial and spiritual care. A recent case illustrates the use of this unit.

An 81-year-old woman was sent to the emergency department from a nursing home because of acute respiratory distress and altered mental status. In the absence of advance directives, she was intubated, and a clinical diagnosis of acute stroke was made and confirmed on the basis of a computed tomographic scan of her head. The emergency room physician recognized the possibility that admission to the ICU might not best serve the interests of this patient. Her family members concurred, the hospice–palliative care unit was consulted from the emergency department, and the patient was transferred directly to the hospice–palliative care unit, where she died comfortably, surrounded by family members in a homelike environment.

We are currently building a replacement academic hospital, to be completed in 1999, where the acute hospice–palliative care unit will be located directly above the ICUs. We are gratified to see that Drs. Miller and Fins endorse our approach.

Charles F. von Gunten, M.D., Ph.D.
Kathy J. Neely, M.D.
Northwestern University Medical School, Chicago, IL 60611-2906

1 References

1
Kellar N, Martinez J, Finis N, Bolger A, von Gunten CF. Characterization of an acute inpatient hospice palliative care unit in a U.S. teaching hospital. J Nurs Adm 1996;26:16-20
CrossRef | Web of Science | Medline

To the Editor:

. . . Although the principles of palliative care could be incorporated into the practice of an alternative-care unit adjacent to the ICU, moving patients into a separate unit carries the risk of confirming the denial of death by modern society and tidying up the failures of curative medicine when there is “nothing more to be done.” An alternative model, widely used in hospitals in the United Kingdom, is a multidisciplinary palliative care team that sees patients in an advisory capacity. In many situations, simple advice on how to control pain and other symptoms is sufficient, and the advisory role provides opportunities for discussion and learning at each consultation, thus enabling a greater number of patients and staff to be exposed firsthand to the principles of palliative care. It could be argued that palliative care is an approach or a philosophy rather than a unit. The cost, effectiveness, and educational value of both models require evaluation.

Karen Forbes, M.R.C.P.
University of Bristol, Bristol BS2 8ED, United Kingdom

To the Editor:

The proposal by Miller and Fins to restructure hospital care for dying patients echoes our 1976 article announcing the establishment of the palliative care service at McGill's Royal Victoria Hospital.1 This service offered a continuum of hospice-style services: a 12-bed palliative care unit (now a 16-bed unit), consultation to the active-treatment units, a home care program, and bereavement support.

Has the palliative care service at Royal Victoria Hospital resulted in the benefits Miller and Fins envision? As they predicted, effective reform has required “changes in practice, clinical education and physical space”2 and has resulted in benefits analogous to those of “natural childbirth.” Death is recast as a natural event in a setting that fosters a sense of safety, permitting personal reflection, growth, and reconciliation through the input of a care-giving team that includes both patient and family. The setting has also resulted in a rich cross-fertilization between the palliative care service and the hospital system. The patient on the palliative care service has immediate access to a full spectrum of consultation and therapeutics, and competence in palliative care is available to the larger institution — for example, with the music therapist assisting at the cardiac catheterization of an anxious patient, and the palliative care physicians offering a dynamic option to therapeutic nihilism.

Balfour M. Mount, M.D.
McGill University, Montreal, QC H3G 1Y6, Canada

2 References

1
Mount BM. The problem of caring for the dying in a general hospital: the palliative care unit as a possible solution. Can Med Assoc J 1976;115:119-121
Web of Science | Medline

2
Ajemian I, Mount BM, eds. The RVH manual on palliative/hospice care. Salem, N.H.: Ayers, 1982.

To the Editor:

I fail to understand why Miller and Fins deem it necessary, or even appropriate, to affiliate their proposed alternative-care unit with the ICU. They talk specifically of dying patients transferred from the ICU. I suspect that a rather small percentage of patients hospitalized with what proves to be terminal illness come from the ICU. The others, who come from the general medical and surgical floors, need the same kind of care as those from the ICU. Should they be excluded from this new unit? Surely not.

The ICU and the palliative care unit are as disparate as two units in a hospital can be. To try to affiliate them makes little if any sense. They should be separate, independent parts of the overall hospital structure.

Larry Hill, M.D.
Manila, Philippines, APO AP 96440

To the Editor:

There have been excellent palliative care units in existence for more than a decade that serve all the functions Miller and Fins describe in their article: patient care, teaching, and research on outcomes. As in many other areas of health care reform, our Canadian colleagues have led the way. In particular, we can point to the work of Eduardo Bruera at the University of Edmonton and Neil MacDonald and Balfour Mount at McGill University as examples of first-rate palliative medicine that serves the needs of dying patients, educates future physicians and other health care workers, and deepens our understanding of the needs of the dying through both basic and outcomes research.1-4 In this country, the work of Kathleen Foley and her colleagues at Memorial Sloan-Kettering Cancer Center, as well as the palliative care units established by Charles von Gunten at Northwestern University and Declan Walsh at the Cleveland Clinic, have enabled us to build and expand on the original pioneering work.5

The article by Miller and Fins focuses too narrowly on patients in the ICU and overlooks the fact that hospice and palliative care units also need to meet the needs of larger groups of patients — namely, those who come from home (primarily patients with disseminated cancer) not necessarily to die but to seek care that will stabilize their symptoms so that they may return to the community. The proposal seems to overlook entirely the role of home hospice and community-based care, with the inpatient unit often serving primarily as a transition between the hospital and the home.

Finally, we were astonished that the authors cited an article by one of us, which describes in detail a hospice and palliative care unit that already meets all the goals the authors outline in their article. The unit has been in existence for more than a year and is easily available for observation by Dr. Miller, since it is located at the University of Virginia Hospital (almost adjacent to our medical ICU), the same institution where he serves as an instructor in bioethics.

Carlos F. Gomez, M.D., Ph.D.
John Herrmann, M.A.
University of Virginia School of Medicine, Charlottesville, VA 22908

5 References

1
Bruera E, Miller MJ, Kuehn N, MacEachern T, Hanson J. Estimate of survival of patients admitted to a palliative care unit: a prospective study. J Pain Symptom Manage 1992;7:82-86
CrossRef | Web of Science | Medline

2
Buckingham RW III, Lack SA, Mount BM, MacLean LD, Collins JT. Living with the dying: use of the technique of participant observation. Can Med Assoc J 1976;115:1211-1215
Web of Science | Medline

3
Fainsinger R, Miller MJ, Bruera E, Hanson J, Maceachern T. Symptom control during the last week of life on a palliative care unit. J Palliat Care 1991;7:5-11
Medline

4
Mount BM. The problem of caring for the dying in a general hospital: the palliative care unit as a possible solution. Can Med Assoc J 1976;115:119-121
Web of Science | Medline

5
Walsh D, Gombeski WR Jr, Goldstein P, Hayes D, Armour M. Managing a palliative oncology program: the role of a business plan. J Pain Symptom Manage 1994;9:109-118
CrossRef | Web of Science | Medline

Author/Editor Response

The authors reply:

To the Editor: In response to the findings of the Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatment (SUPPORT), we recommended a conceptual framework for restructuring the prevailing mode of care for hospitalized dying patients, modeled on the transformation of childbirth practices. Motivated by SUPPORT's examination of critically ill patients, many of whom spend considerable time in the ICU, we focused on the importance of an inpatient unit as an alternative to the ICU for patients who are dying. Nonetheless, we agree with Cassel and Hill that hospice-like alternative-care units should serve suitable patients from all parts of the hospital. However, the complete separation of the palliative care unit from the ICU, as Hill recommends, would reinforce the antithetical dichotomy between the two modes of care that we sought to challenge.

We are more sanguine than Weissman about the efficacy of such alternative-care units, combined with educational reform, to overcome cultural barriers that interfere with helping hospitalized patients achieve a peaceful and dignified death. By legitimizing and encouraging palliative care strategies within the acute care hospital, this reform can promote improved care of dying patients, as the case presented by von Gunten and Neely illustrates. We endorse Forbes's model of a multidisciplinary palliative care team but see this as complementary to an inpatient palliative care unit rather than an alternative. The successful work of Mount, von Gunten and Neely, Gomez and Herrmann, and others indicates the salience of restructuring the care of hospitalized dying patients. Contrary to the assertion of Gomez and Herrmann, we did not overlook the role of home hospice, since we explicitly mentioned that some patients served by the alternative-care unit might be discharged home with the support of hospice programs.

Any similarity between the ideas presented in our article and the palliative care program at the University of Virginia is coincidental, since we were not familiar with its planning or operation. We cited an article by Gomez to document the importance of linking an inpatient palliative care unit with improved education of residents and medical students in the care of dying patients. That article appeared in print after our manuscript had been accepted for publication.

We believe the time is ripe for reconfiguring the care of dying patients in a range of hospital settings and hope that the development of a new DRG for terminal care, noted by Cassel, advances the implementation and evaluation of hospital-based palliative care initiatives.

Franklin G. Miller, Ph.D.
University of Virginia, Charlottesville, VA 22908

Joseph J. Fins, M.D.
Cornell University Medical College, New York, NY 10021

Citing Articles (1)