Journal of Clinical Oncology

Volume 20(11) 1 June 2002 pp 2752-2755

Overcoming Obstacles to Hospice Care: An Ethical Examination of Inertia and Inaction

[THE ART OF ONCOLOGY: WHEN THE TUMOR IS NOT THE TARGET ]

Daugherty, Christopher K.; Steensma, David P.

From the Department of Medicine, MacLean Center for Clinical Medical Ethics, the Cancer Research Center at the University of Chicago, and Vitas Hospice, Chicago, IL, and Division of Medical Oncology, Department of Oncology and Division of Hematology, Department of Medicine, Mayo Clinic, Rochester, MN.

C.K.D. is a former recipient of a Career Development Award from the American Society of Clinical Oncology. His work is currently supported by a Faculty Scholar Award from the Soros Foundation Open Society Institute’s Project on Death in America and by National Institutes of Health grant no. R01 CA087605-01A1.

Address reprint requests to Christopher K. Daugherty, MD, University of Chicago, 5841 South Maryland Ave, MC 2115, Chicago, IL 60637-1470; email: [log in to unmask]

Outline

· VIEWING OBSTACLES TO HOSPICE REFERRAL AS ETHICAL DILEMMAS

· COMMUNICATING TERMINAL PROGNOSIS WHILE MAINTAINING HOPE

· CONSENT FOR HOSPICE AND WILLINGNESS TO ACCEPT A TERMINAL PROGNOSIS

· PATIENT ABANDONMENT, CONFLICTS OF INTEREST, AND LOSS OF CONTROL

· HOSPICE SERVICES AND PER DIEM REIMBURSEMENT: SUFFICIENT RESOURCES FOR ALL DYING CANCER PATIENTS?

· ACKNOWLEDGMENT

· REFERENCES

THE STATISTICS are sobering: more than half a million Americans die of cancer each year, but only 20% to 50% of eligible patients receive any formal hospice care. 1,2 When cancer patients do enroll in a hospice program, their median survival is just 2 to 3 weeks—barely enough time for many patients and families to understand the hospice program’s system, let alone receive the full range of benefits offered. 3-5 Some patients are even referred for hospice care so late in their disease course that they die before all the paperwork can be completed.

Hospice-based palliative care is widely acknowledged as an invaluable intervention for patients approaching the end of life. Dying of cancer without the help of a hospice program has been compared to undergoing surgery without anesthesia. Why, then, does the promise of hospice care so frequently go unfulfilled?

Some have argued that the fundamental problem is that cancer physicians are simply inaccurate in predicting actual time to death. 5 Others claim that physicians just don’t know when to quit—that oncologists keep offering aggressive anticancer therapy, including experimental interventions, even when an imminent fatal outcome should be obvious to everyone. Critics argue that because of systematic overestimation of patient survival time and an inflated view of the benefits of anticancer therapy, oncologists tend to refer their patients for hospice care far too late for them to receive meaningful end-of-life care—if they refer their patients at all.

Physicians do often overestimate their patients’ survival time, but this can only be part of the story. Oncologists can usually identify cancer patients with an expected survival of less than 6 months, which is the Medicare criterion for hospice referral in the United States. 6 But clearing this hurdle of knowledge does not mean all obstacles to action have been overcome. In fact, we believe that physicians’ prognostication skills are not as important in erecting barriers to hospice program referrals as are attitudes about how and when a grim prognosis should be communicated and fears about how hospice enrollment might change patient care.

VIEWING OBSTACLES TO HOSPICE REFERRAL AS ETHICAL DILEMMAS

If oncologists are hesitant or unwilling to refer patients with advanced cancer to hospice care, yet hospice programs provide useful and effective palliation, then the reasons for oncologists’ aversion to referral might best be viewed as ethical dilemmas. True ethical dilemmas, such as these, are complex and admit no easy answers. Are oncologists concerned that communicating a terminal prognosis to cancer patients and their families will take away hope? Is hospice care still appropriate for patients and families who are unwilling to accept a fatal prognosis, and just how might a suitable “informed consent” for hospice enrollment be obtained from such patients? Are physicians apprehensive that some patients receiving home hospice care will feel abandoned by their doctor? How do physicians deal with their own emotions about separation from their long-term patients? Do oncologists fear loss of control over their hospice-enrolled patients’ medical care, perhaps worrying that the hospice program will categorically deny expensive interventions such as transfusions, palliative radiotherapy, or antibiotics that may still offer some benefit? Each of these questions has challenging ethical and practical dimensions.

COMMUNICATING TERMINAL PROGNOSIS WHILE MAINTAINING HOPE

Some clinicians believe that it is not the physician’s place to hang black crepe in the sickroom; these physicians worry that honestly disclosing a terminal prognosis to patients and families risks destroying hope. 7-9 While most patients’ and family members’ hopes can be redirected toward more realistic outcomes (eg, death at home, death with dignity, and death without excessive pain), some people will remain convinced that accepting a terminal prognosis means giving up hope for the only outcome they consider truly meaningful: continued survival, or even a cure. For a few of these patients, talk about redirecting hope away from unrealistic expectations only irritates them or arouses suspicion. The clinician who is perceived as persistently insisting on changing care goals exclusively toward palliation may motivate the occasional patient to aggressively seek out someone who will paint a brighter picture—even if that picture is a mirage and even if it means resorting to dangerous or expensive alternatives. 10,11 These patients may search far and wide for someone who will act in concert with their unwavering goal for a cure: “If you can’t or won’t do anything more for me, doctor, then I’ll find someone who will!”

In this setting, it may be inappropriate for physicians to bludgeon a patient with a litany of reasons why it is time to “switch” to palliative therapy. It is easy to understand why some busy clinicians thrust into this situation ultimately accede to patients’ or families’ demands for further anticancer therapy, even while knowing full well that such therapy is unlikely to offer any real benefit. Some patients simply refuse to believe that enrolling in a hospice program is not the same as giving up.

Given these genuine concerns about destroying hope, it is important to keep in mind the precise documentation requirements for obtaining hospice benefits. In the United States, a physician must certify that a patient enrolling in a hospice program has an anticipated survival time of 6 months or less if the illness follows the usual course. But there is no requirement that the patient must actually believe that in his or her case, the disease will follow the usual course. And what if a patient lives longer than 6 months? There are no penalties for pessimism, and there are no consequences for an inaccurate hospice referral made in good faith. If a hospice-enrolled patient lives more than 6 months, the patient can be recertified for another 6-month benefit period if the prognosis remains guarded, which is usually the case for patients with advanced cancer. Thus only a reasonable possibility of dying in the next 6 months is required for hospice eligibility, not an absolute certainty of death.

Some clinicians may find it more comfortable to discuss the potential benefits of hospice program enrollment with patients if they are careful to speak in terms of possibility and likelihood rather than certainty. Such “possibility thinking” promotes good, comprehensive palliative care while not crushing those patients who need to cling to the fragile, flickering hope that their case will be the exception or the miracle.

CONSENT FOR HOSPICE AND WILLINGNESS TO ACCEPT A TERMINAL PROGNOSIS

In some settings, when oncology patients enroll in hospice care, they sign an informed consent document stating they understand that they have a terminal illness with a limited prognosis and do not wish to receive any further anticancer therapy. In addition, patients are often asked to sign an advance directive form specifically stating that when death arrives, they do not wish to attempt to delay it with dramatic interventions such as cardiopulmonary resuscitation or mechanical ventilation.

Although such requirements give legal protection and ensure complete disclosure of prognosis and goals, this process may be seen as a burden. Clinicians may worry that patients will perceive such formal, forceful documents as a demand to “sign their life away.” Physicians who consider this consent process an obstacle may shy away from a discussion of hospice enrollment, delaying the conversation for days and weeks while illness progresses and the opportunity for meaningful hospice care slips away.

In addition, even when physicians are completely forthright, some patients will simply deny their prognosis. Even the most compassionate oncologists and those most skilled in communicating with patients occasionally run across people who adamantly refuse to accept their impending death.

Physicians should realize that many hospice care providers are experienced in the art of negotiating expectations and prognosis and can help bear such burdens. Most palliative care professionals are aware that patient and family acceptance of a terminal prognosis is a dynamic process. The overall goal of hospice care is to relieve suffering, and many hospice workers understand that belaboring the point of prognosis disclosure and insisting on full and explicit documented understanding on the part of the patient can increase suffering. In such cases, hospice programs may be willing to make exceptions to their customary consent process.

PATIENT ABANDONMENT, CONFLICTS OF INTEREST, AND LOSS OF CONTROL

When patients have been receiving care for cancer for months or years before the final phase of their illness, they may view hospice program enrollment as a form of abandonment by a familiar, trusted physician. Patients who have become accustomed to frequent laboratory tests, chemotherapy cycles, hospitalizations, and visits to distant clinics may suddenly find themselves receiving most of their care at home, without easy access to the familiar clinic or hospital. Oncologists may be hesitant or unwilling to refer a patient for hospice care if they sense separation anxiety in the patient or family.

Fear of separation can be a two-way street. Medicine is, above all, a relationship, and the intense, prolonged clinical contact typical of cancer care can forge a strong emotional bond between a physician and a patient. Every oncologist tends and admires a few giant sequoias among their forest of patients. One study demonstrated that the longer a physician knows a patient, the less accurate the physician is in predicting the time to the patient’s death. 12 Unwillingness to accept an impending death becomes much more complicated when the patient and the doctor share it.

The good news is that hospice benefits allow oncologists to continue to play an important role in their patients’ care even after a patient has enrolled in a hospice program. 1 Patients receiving hospice care can continue to visit physicians in the outpatient setting as frequently as they wish, and physicians can bill as they normally would for outpatient services. Although many patients currently enrolled in hospice programs are simply not well enough to travel to an outpatient facility, those referred for hospice care at an earlier point in their clinical course can interact with their physicians in the clinic until the final stages of illness.

Should hospice-enrolled patients require readmission to an inpatient setting for respite care or for a complication that cannot be treated at home, it is often possible to admit the patient to the original, familiar care institution, even if the hospice program has no affiliation with that institution. This usually requires removing the patient from hospice care during the hospitalization, but there are currently no prohibitions or punishments for doing this, and patients can return to the care of the hospice program at the time of discharge.

Hospital admissions may be inconvenient for hospice programs, and some programs are justifiably concerned about the potential for being accused of fraud as a result of several recent high-profile hospice audits. 1 However, the abuses documented in these audits have typically involved hospice program enrollment of patients with chronic neurological diseases, patients with dementia, and residents of long-term health care facilities, not patients with advanced cancer who are being cared for at home.

It may seem cynical even to mention it, but it is no secret that there is at least one potential obstacle to hospice referral that represents a serious conflict of interest for oncologists: the current clinical payment structure in the United States favors administration of expensive chemotherapy and growth factors but is much less generous when it comes to thinking and giving advice—and pays nothing extra for caring. The snake of selfishness is, we hope, a rare species. But who has not wondered, at least once, whether a colleague’s choice of certain therapeutic agents was partially influenced by monetary considerations? Everyone agrees that making clinical decisions based on potential physician revenue is completely unacceptable, but the reality is that money speaks with a loud voice, and it may be hard not to listen. It is human nature to act in ways that are rewarded, and as Ralph Waldo Emerson once pointed out, often the only reward for virtue is virtue.

Financial considerations aside, most oncologists are extremely busy people, and it almost always takes less time to explain the side effects and schedule of a new treatment regimen than it does to discuss death and dying. It is also easier on emotions to talk about possible therapies than it is to peer into the murkiness of mortality, and it is no secret that medical education is better at preparing physicians for opening the medicine cabinet than for lifting the lid on a philosophical Pandora’s box. More training in communication skills and more practice at discussing sensitive topics certainly can help, but even eloquent, compassionate physicians are still bound by the 24-hour day. Yet somehow, time for such intimate discussions must be scavenged; these weighty matters are simply too important for procrastination, and delay is a disservice to patients.

HOSPICE SERVICES AND PER DIEM REIMBURSEMENT: SUFFICIENT RESOURCES FOR ALL DYING CANCER PATIENTS?

Another barrier to providing excellent end-of-life care to cancer patients is the relatively meager Medicare, Medicaid, and insurance benefits that provide the bulk of hospice program revenue in the United States. Hospice care is currently paid for through a per diem benefit, with the Medicare per diem ranging from $100 to $120, varying regionally. 1

This reimbursement is often adequate for administrative costs and for hospice benefits such as in-home equipment and medications for symptom control, especially in hospice programs where volunteers provide many services. But what about more expensive and aggressive palliative interventions such as antibiotics, patient-controlled anesthesia pumps, radiation therapy, and blood transfusions? The standard per diem does not begin to cover this level of care. As a result, physicians who feel strongly about the continued benefit of such interventions for patients at the end of life may not refer such patients to hospice programs if they believe that the program is unable or unwilling to provide these costly interventions. Such ethical concerns can snowball: physicians may not refer other patients for hospice care, even those who are not receiving aggressive and expensive interventions, because they fear that hospice programs may be “holding back” certain types of appropriate care because of a limited budget.

Oncologists should keep in mind that even the most resourceful hospice programs operate on a thin fiscal margin and that the money for expensive services must come from somewhere. Small numbers of costly patients may drive hospice programs to the brink of insolvency, where care for all enrolled patients will suffer. Delays in referral can exacerbate the problem: if expensive services are used for patients who spend only a short time under hospice care, the hospice program will have fewer days of per diem benefit over which to spread the expenses.

Some patients who continue to receive costly or aggressive interventions are not yet appropriate candidates for hospice care. But it is by no means obvious where to draw the line. Expensive therapies sometimes do offer a legitimate palliative benefit for patients with terminal cancer who have a high likelihood of death within 6 months. Hospice programs have an ethical obligation to care for these patients. Some creative solutions have been explored, such as subcontracting specialized services like radiation therapy and transfusion support to other health care agencies. 13 A demonstration project funded by the Robert Wood Johnson Foundation showed the feasibility of supporting limited administration of palliative chemotherapy to hospice-enrolled patients, including the use of selected experimental anticancer agents (I. Byock, B. Volk-Craft, personal communications, June 2001).

In reality, many hospice programs simply do not have the resources to provide this level of care. Yet some programs may still accept patients who need expensive services. These programs are willing to accept patients whose care will result in a net financial loss in order to develop more meaningful relationships with referring clinicians. The expectation is that these solid relationships will lead to the future referrals of other hospice-appropriate patients who have less intense clinical needs. Therefore, it is best not to make assumptions, but rather to communicate directly with hospice programs on a case-by-case basis.

In conclusion, it has been argued that some of the barriers in providing end-of-life cancer care in the United States might be eliminated by increasing or restructuring the Medicare hospice benefit. 1 Money is certainly important, but it is not enough. For hospice care to become more widely and appropriately used, efforts must be directed toward improving the referral patterns of the physicians who actually take care of cancer patients. Specific attention must be given to the complex ethical and practical obstacles to hospice program referral discussed above.

Continuing medical education programs that focus on the details of Medicare and insurance hospice benefits may improve physicians’ understanding of end-of-life care. Research efforts are also needed, aimed at understanding current communication and decision-making practices among cancer physicians and their patients to clarify how behaviors, beliefs, habits, and practices may create obstacles to adequate hospice-based, palliative care. Until education improves knowledge and research provides clarification, the ethical dilemmas faced by cancer physicians will continue to promote misunderstandings and will prevent patients with advanced cancer from receiving the best possible palliative care—and inertia and inaction will hold the day.

ACKNOWLEDGMENT

The authors thank Kathy Foley, MD, for her helpful comments during the writing of the manuscript and for her help and support in the process of examining the issues described in this article.

REFERENCES

1. Foley KM, Gelbard H (eds): Institute of Medicine Report: Improving Palliative Care for Cancer. Washington, DC, National Academy Press, 2001 [Context Link]

2. Virnig BA, Kind S, McBean M, et al: Geographic variation in hospice use prior to death. J Am Geriatr Soc 48: 1117-1125, 2000 [Medline Link] [CINAHL Link] [BIOSIS Previews Link] [Context Link]

3. Christakis NA, Escarce JJ: Survival of Medicare patients after enrollment in hospice. N Engl J Med 335: 72-78, 1996 [Context Link]

4. Reuben DB, Mor V, Hiris J: Clinical symptoms and length of survival in patients with terminal cancer. Arch Intern Med 148: 1586-1591, 1988 [Medline Link] [Context Link]

5. Forster LE, Lynn J: Predicting life span for applicants to inpatient hospice. Arch Intern Med 148: 2540-2543, 1988 [Medline Link] [Context Link]

6. Steensma DP, Loprinzi CL: The art and science of prognosis in patients with advanced cancer. Eur J Cancer 36: 2025-2027, 2000 [Medline Link] [Context Link]

7. Christakis NA: Death Foretold: Prophecy and Prognosis in Medical Care. Chicago, IL, University of Chicago Press, 2000 [Context Link]

8. Miyaji N: The power of compassion: Truth-telling among American doctors in the care of dying patients. Soc Sci Med 36: 249-264, 1993 [Medline Link] [CINAHL Link] [BIOSIS Previews Link] [Context Link]

9. Kodish E, Post SG: Oncology and hope. J Clin Oncol 13: 1817-1822, 1995 [Medline Link] [Context Link]

10. Eidinger RN, Schapira DV: Cancer patients’ insight into their treatment, prognosis, and unconventional therapies. Cancer 53: 2736-2740, 1984 [Medline Link] [Context Link]

11. Gordon EJ, Daugherty CK: Referral and decision making among advanced cancer patients participating in phase I trials at a single institution. J Clin Med Ethics 12: 31-38, 2001 [Context Link]

12. Christakis NA, Lamont EB: Extent and determinants of error in doctors’ prognoses in terminally ill patients: A prospective cohort study. BMJ 320: 469-472, 2000 [Fulltext Link] [Medline Link] [CINAHL Link] [BIOSIS Previews Link] [Context Link]

13. Leland JY, Schonwetter RS: Advances in hospice care. Clin Geriatr Med 13: 381-401, 1997 [Medline Link] [CINAHL Link] [Context Link]

Accession Number: 00005083-200206010-00023

Edward E. Rylander, M.D.

Diplomat American Board of Family Practice.

Diplomat American Board of Palliative Medicine.