Confronting Myths: The Native American Experience in an Academic Inpatient Palliative Care Consultation Program

Abstract

Background:

Recent end-of-life (EOL) care literature in non-Native American (NA) populations has demonstrated the benefits of EOL discussions. EOL discussions are associated with less aggressive care at EOL, better patient self-assessed quality of life, and less caregiver depression after the patient is deceased. There is no literature assessing these issues in NA populations. However, common myths that may affect care include: 1) NA patients will not discuss death and dying, 2) severely ill NA patients and families will not choose do not resuscitate (DNR) status, and 3) NA patients and families will not utilize hospice services if offered.

Methods:

Our study explored these issues utilizing a consultation database from the Palliative Care Consultation Service at University of New Mexico Hospital (UNMH). Statistical analyses were conducted using nonparametric Wilcoxon tests for continuous variables and Fisher's exact test for categorical variables.

Results and conclusion:

Study results demonstrate that health care providers can hold EOL care discussions with NA patients and NA patients' care preferences are affected by these discussions. The result do not support our hypothesis that there would be a lower rate of post-consult DNR status in NA patients (compared with non-NA). NA and non-NA patients and families participated in family meetings and their code status was affected to a similar degree. Furthermore, NA patients and their families choose hospice services at rates similar to non-NA patients seen by the palliative care consultation service.

Introduction

Epidemiology of American Indians/Alaska Natives (AI/AN) differs from that of the general U.S. population

In 2007, approximately 4 million U.S. citizens identified themselves as American Indians or Alaska Natives (AI/AN), also referred to as Native Americans (NAs). Many NAs belong to one of 560 federally recognized tribes in 35 states.¹ Demographic statistics illuminate population differences that impact palliative care services for NAs. Despite a younger population and a 15% increase in life span over the last 30 years, life expectancy for NAs is still approximately 7 years less than for U.S. whites.² Among the NA population, the leading causes of death are heart disease, malignancies, unintentional injuries, diabetes, liver disease, and suicide. Death rates are significantly higher for NA than for the general U.S. population (all ethnicities). The death rate due to alcoholism is 638% higher, due to tuberculosis is 400% higher, due to diabetes mellitus is 291% higher, due to unintentional injury is 215% higher, due to suicide is 91 % higher, and due to homicide is 81% higher. Many of the deaths occur in NAs age 15 to 50 years.^2-4

As death rates due to most causes have decreased since 1970, NAs are living now with more chronic illness. As a result, the dying process occurs more slowly than in the past and is often highly technological. It frequently takes place in facilities far from family and away from tribal homelands. It has been demonstrated that palliative care and hospice services improve quality of life for patients suffering from chronic and terminal illness,^5-12 yet NAs' access to these services lags far behind the rest of the nation. There are large portions of reservation lands that are not served by any Centers for Medicare and Medicaid Services (CMS)-accredited hospice program. Only a handful of hospice/palliative care programs exist throughout the federal Indian Health Service (IHS) and tribally operated health care programs, which serve approximately 1.8 million of these 4 million citizens.^4,13,14 In 2011, the federal Agency for Healthcare Research and Quality (AHRQ) released a National Health Care Disparities Report that further documents disparities in supportive and palliative care for NAs.¹⁵

Current health care system and myth creation

NA communities generally lack awareness that the current Western system of end-of-life (EOL) health care persists with life-prolonging efforts unless specific alternate plans are made. The IHS, most private health care systems, and emergency medical systems automatically initiate cardiopulmonary resuscitation (CPR) therapy, regardless of a patient's disease state, unless a documented do not resuscitate (DNR) order is available. Additionally, NA families and caregivers lack awareness of the parameters and processes for specifying preferences for EOL care. For some NAs and their clinical providers, the current legalistic advance directive requirements are experienced as intrusive, insensitive, and overwhelming, as they do not take into account cultural norms and practices of Native communities.^16,17 The reluctance of health care providers and families to discuss advance directives also may be influenced by cultural sensitivity to the belief held by some NAs that speaking of a future event will lead to its occurrence. For example, discussing EOL care options may be seen as bringing on the patient's death.¹⁸

Even experienced providers have difficulty knowing when and how to approach families to discuss care beyond cure, DNR decisions, pain management, organ donation, and autopsy. There are variations in beliefs among members of the same tribe and between different NA tribes. However, many AI cultures rely on the passage of time for decisions of importance to become clear. For these NAs, these issues are considered within the privacy of extended family meetings, “when the time comes.” Official guidance from Native communities on these matters is lacking as sovereign nations and tribal governments have not legislated yet on this matter. Unfortunately, many NA families feel that they have experienced discrimination in their access to appropriate life-sustaining tertiary care, which further complicates the context of communication and clinical decision making. Historical mistrust based on racial prejudice and related social disadvantage may skew NA EOL decision making toward a hesitancy to decline or withdraw aggressive yet “futile” interventions.¹⁹

All of the above contribute to the myth-making that AIs/NAs will not:

• Discuss dying and death

• Discuss EOL wishes

• Participate in goal-setting discussions

• Accept DNR code status

Palliative care is not at odds with Native American beliefs (or palliative care is consistent with many NA beliefs)

Since 2001, there have been numerous efforts to integrate palliative care services into AI and AN health care systems.^20-25 In the Southwestern United States, we have found that fewer cultural differences exist than one might expect regarding the desired care for a dying person. In general, NA families want a system of care that will honor the dying person's individual and family wishes. Common values include celebrating life, respect for traditional ways, hope for a peaceful death, and access to comfort-focused care while surrounded by family on or near tribal land. Even in Southwestern tribes where “death” is taboo for public discussion, caring for someone during this transition is not taboo and palliative care concepts are congruent with traditional beliefs. Native communities are open to palliative care services that will respect and support local traditions.²⁶ Families moving through the transition of death do not need to choose between traditional, complementary, alternative, or Western allopathic therapies.

The University of New Mexico Hospital experience

The University of New Mexico Hospital (UNMH) has demonstrated a commitment to serving equally New Mexico's diverse ethnic groups. New Mexico has the second-highest percentage of NA residents in the country (9.5% according to 2000 census data). UNMH is a state-funded “safety net” hospital for the state of New Mexico and is the only academic medical center and Level 1 trauma center in the state. Additionally, UNMH is the major tertiary referral center for the IHS hospitals in eastern Arizona and all of New Mexico. UNMH sees the highest percentage of NA patients among total admissions of any hospital in the country. Of all patients admitted to UNMH, 10.7% percent are NA, drawing from 19 regional Pueblo communities, Apache and Navajo Tribes, and urban NAs. Many NA patients admitted to UNMH reside on reservation lands with an average distance from UNMH of 250 miles. This article details the UNMH palliative care program's experience providing care for these Southwestern NA patients.

The myth that NAs are not willing to discuss EOL care may contribute to the low 2.8% national NA participation rate in the Medicare hospice benefit and low completion rate of advance directives.¹³ Our palliative care program's approach has led to utilization of the UNMH Inpatient Palliative Care Consultation Service among differing patient ethnic groups, We have not noted evidence of negative connotation in the NA population. Our program has improved the EOL care at UNMH for NAs and their families. We believe that our use of technology and multiple methods of education, including tele-health statewide community provider outreach, have affected our organization's cultural belief that all patients, regardless of ethnicity, should be provided equal access to palliative care.

It should be noted and stressed that there are vast differences among the hundreds of tribes and distinct NA cultures in the North American continent, and this article deals with our team's experience with patients from Southwestern U.S. tribes admitted to the UNMH only. Furthermore, there may also be great differences not only among tribes, but within tribes, regarding cultural and spiritual practices and desires regarding EOL care.

Methods

A new adult Palliative Care Consultation Service was initiated at UNMH in January 2009. The palliative service consists of an interdisciplinary team of physicians, a nurse practitioner, a NA health educator, social services, chaplaincy, and arts in medicine providers (musicians, poets, artists). The service is housed within the Division of Geriatrics in the Department of Internal Medicine, but consultation is not restricted to geriatric patients. During 2009, referrals to the consultation service originated from all clinical departments in the hospital, not only internal medicine. Whereas other providers (nurses, physical therapists, chaplains) may bring a need for referral to the attention of the primary team, referrals are made by physician only. The primary attending physician must approve of the referral.

After obtaining education from NA health care workers and providers regarding effective techniques for communication and a review of the literature regarding NA preferred patterns of communication with Southwestern native communities, the salliative care service embraced several of the following culturally appropriate techniques as listed in Table 1. These interview techniques evolved as the consult service evolved. During interdisciplinary team meetings, members shared their self-reflections regarding their communication skills and where they could improve them with NA families. These techniques have not been tested in a controlled manner for impact on outcomes or acceptability to patients and families.

Table 1.

Techniques that Facilitate Native American End-of-Life Discussions

• All patients/families are asked how they wish to receive information, how they make decisions, and who needs to be present.

• All patients/families queried if they would prefer a translator.

• Care provider often speaks in third person.

• Reassure that no harm is intended from the discussion.

• State that palliative care team and other providers “hope for the best.”

• Attention to appropriate eye/hand contact.

• Allow patient/family to control tempo of discussions.

• Encourage comfort with silence.

• Schedule family meetings at convenience of family.

• Support consultation with traditional healers.

• Provide sufficient time for decision making.

All patients and/or families (regardless of ethnicity) receiving a consult are asked how they wish to receive information, how they make decisions, and who needs to be present during discussions. No assumptions are made. Prior to discussions, all patients and/or families are queried if they would prefer to use a translator; which many do, even if they are bilingual. UNMH has a Navajo translator on staff, which has proven to be essential in communication with family members who speak only Navajo. Translators for other NA languages are available in person, by video tele-conference, or by telephone.

In discussions specifically with NA patients and families, the Palliative Care Consultation Service often speaks in the third person, as speaking directly about the patient's illness and prognosis may be seen as “wishing or causing the poor outcome” and eroding hope. For example: “When we have cared for other patients with illness similar to yours, we have experienced …” Additionally, at the beginning of family meetings with NA families, reassurance is given to the patient and family that no harm is intended from the discussion, and that the palliative and other health care providers “hope for the best” for the patient. Attention to appropriate eye/hand contact is also made, because firm handshakes and directly looking into the eyes of some NA people may appear to be confrontational and disrespectful. Comfort with silence and allowing the patient and family to determine the tempo of the conversation have proved beneficial, as well as scheduling family meetings at the convenience of the families because many live a distance from the hospital. Consultation with traditional healers is supported and encouraged, and traditional healers are welcomed into the UNMH setting. Families are reassured that decision making is a process, and efforts are made to ensure that the family has sufficient time to make decisions.

A consultation database was initiated at the start of the Palliative Care Consultation Service. Data such as ethnicity, age, sex, referring service, reason for consultation, and disposition were recorded for each patient. In addition, code status and the presence of a family meeting before and after palliative consultation were noted. A “family meeting” is defined as a documented discussion between the primary team and the patient and/or family about the patient's current condition, and/or prognosis, and/or treatment options.

The University HealthSystems Consortium (UHC) database was used to identify all patients admitted to UNMH in 2009. The UHC database is a collection of patient-related data submitted by members of mainly academic health centers for the purpose of performance improvement. There are 100 items on the matrix including 20 UHC Product lines for cost and mortality, 20 AHRQ Patient Safety Indicators, 10 AHRQ Inpatient Quality Indicators, and Readmission Measures.

For all patients who were not seen by the palliative service, primary diagnosis and ethnicity were determined from the UHC database. Patients' self-described ethnicity was recorded in the UNMH electronic medical record at hospital admission. Patients can identify as belonging to Hispanic/Latino, Non-Hispanic White, African American, Asian, NA, other, and unknown. These categories are used in the UHC database, and do not reflect the beliefs or assumptions of the authors. The ethnicity listed in the medical record was clarified by questioning the patient or family on initial palliative consultation.

Our pre-specified primary outcomes included the occurrence of pre- and postconsult family meetings and DNR code status among patients of differing ethnicities. Due to the myths mentioned above, our hypothesis was that fewer NA patients and families would decide upon a postconsultation DNR code status, compared with non-AI/AN patients. In terms of family meetings, we hypothesized that there would be no difference in rates based on patient ethnicity. Secondary outcomes included patient age, sex, primary diagnosis, reason for consultation, time to consult, and disposition.

In our statistical analyses, we used nonparametric Wilcoxon tests for continuous variables and Fisher's exact test for categorical variables.

Results

During 2009, 413 new patients were seen by the Palliative Care Consultation Service at UNMH, creating an average daily census of 10 patients. Thirteen percent of all patients seen by the Palliative Care Consultation Service in 2009 were NA. Patient information including age, sex, reason for consultation, and average time to consultation is listed in Table 2 for NA versus non-NA patients. The average age of the NA patients seen by the consult service was lower at 53.8 years, compared with 60.9 years for non-NA patients (p=0.01). Among NA patients, the most common reason for consultation was “clarifying the goals of care” (70%), followed by pain and symptom management (23%), and family support (4%). There was no statistically significant difference in reason for consultation or time to consultation between Native and non-Native patients. Table 3 lists the most common diagnoses for Native and non-Native patients who were seen by the palliative service.

Table 2.

Baseline Characteristics of Native American versus Non-Native American Patients

	Native American patients	Non-Native American patients	P value
Number of patients	53	360
Age in years	53.8±19.2 (22–87)	60.9±17.2 (18–102)	0.01
Time to consult in days from admission date	10.3±11.0 (1–52)	10.3±12.1 (1,100)	0.82
Sex			0.01
Male	19 (36%)	201 (56%)
Female	34 (64%)	159 (44%)
Reason for consultation			0.47 (overall)
Goals of care	37 (70%)	258 (72%)	0.75
Pain and symptommanagement	12 (23%)	81 (23%)	1.00
Advance careplanning	1 (2%)	4 (1%)	0.50
Family support	2 (4%)	4 (1%)	0.18
Comfort care	1 (2%)	6 (2%)	1.00
Other	0 (0%)	7 (2%)	0.60

Summaries are mean±standard deviation and range in parentheses for continuous variables and number and percentages in parentheses for categorical variables. P values computed using nonparametric Wilcoxon tests for continuous variables and Fisher's exact test for categorical variables.

Table 3.

Most Common Diagnoses for Native American Patients versus Non-Native American Patients (Exclusive of OB-GYN Patients)

All non-maternity adults UNMH	All PCS patients	PCS Native American patients
Pneumonia, organism NOS	Septicemia NOS	Alcoholic cirrhosis liver
Coronary atherosclerotic native vessel	Pneumonia, organism NOS	Food/vomit pneumonitis
Septicemia NOS	Alcohol cirrhosis liver	Septicemia NOS
Antineoplastic chemo enc	Neoplasm related pain	Pneumonia, organism NOS
Acute pancreatitis	Food/vomit pneumonitis	Streptococcal septicemia

NOS, not otherwise specified; PCS, palliative care service.

For 28% of NA patients seen by the palliative care team, a family meeting was held prior to the consultation (Table 4). This increased to 75% after a palliative care consultation (p<0.001). Twenty-three percent of NA patients had a DNR order prior to palliative care consultation compared with 62% after consultation (p<0.001). The corresponding percentages for non-NA patients seen by the consultation service are listed as comparisons. In Table 4, analysis demonstrates that the change in the percentage of patients that participate in a family meeting or agree with a DNR code status are statistically significant for both NA and non-NA patients (p<0.001). The only statistically significant difference between these two groups is in the rate of pre-consultation DNR order (23% for Native patients and 38% for non-Native patients, p=0.03).

Table 4.

Family Meeting, Code Status, and Disposition for Native American Patients versus Non-Native American Patients

	Native American patients	Non-Native American patients	P value
Family meeting
Preconsult	15 (28%)	110 (31%)	0.87
Postconsult	40 (75%)	242 (67%)	0.27
DNR
Preconsult	12 (23%)	136 (38%)	0.03
Postconsult	33 (62%)	244 (64%)	0.44
Disposition			0.21 (overall)
Death in hospital	24 (45%)	104 (29%)	0.02
Skilled nursing facility	8 (15%)	49 (14%)	0.70
Residential hospice	6 (11%)	55 (15%)	0.13
Home hospice	4 (8%)	40 (11%)	0.63
Home (with or without home care services, excludinghospice)	6 (11%)	78 (22%)	0.10
Other	5 (8%)	34 (9%)	1.00

Summaries are number and percentages in parentheses. P values computed using Fisher's exact test.

The change in % that agree to DNR or family meeting from preconsult to postconsult is statistically significant at p<0.001 level for both Native American and non-Native Amercians—that is, there is a statistically significant increase in % that agree to a family meeting and to a DRN after the consultation.

DNR, do not resuscitate.

Disposition of patients seen by the palliative care service is also displayed in Table 4. More NA patients die in the hospital than non-Native patients (p=0.02). Disposition to hospice (home or residential) is not statistically different between the two groups.

Discussion

Recent EOL care literature in non-NA populations has demonstrated the benefits of EOL discussions. Prognostic information and knowledge of the low likelihood of success of CPR affects medical decision making. Receipt of this information decreases the proportion of patients who choose this aggressive measures at EOL (CPR attempts, intensive care unit [ICU] admission).^26,27 Additionally, patients who are unrealistic about their prognosis are more likely to opt for more aggressive care at EOL, which does not change their survival time.²⁸ EOL discussions are not associated with an increased risk of patient depression, but are associated with less aggressive care at EOL, better patient self-assessed quality of life, and less caregiver depression after the patient is deceased.^29,30 Unfortunately, many patients who wish to have discussions about EOL issues with their physicians never do so, even if they make a specific request.^26-28,³¹

There is no literature assessing these issues in NA populations. However, common myths that may affect care include: 1) NA patients will not discuss death and dying, 2) severely ill NA patients and families will not choose DNR status, and 3) NA patients and families will not utilize hospice services if offered. These myths often lead health care providers to fear and avoid EOL discussions with NA patients and their families. This is the first study to demonstrate that health care providers can hold EOL care discussions with NA patients and NA patients' care preferences are affected by these discussions.

The results of this study do not support our hypothesis that there would be a lower rate of postconsult DNR status in NAs (compared WITH non-NAs). NA and non-NA patients and families participated in family meetings and their code status was affected to a similar degree. Furthermore, NA patients and their families choose hospice services at rates similar to non-Native patients seen by the Palliative Care Consultation Service.

One interpretation of the documented increase in DNR code status and acceptance of hospice for NA patients after palliative care consultation is NA patients and families responded to prognostic information and discussions that were held in a culturally appropriate manner. Lower rates of family meetings prior to consultation in NA patients may reflect providers' concerns about having these discussions, and fear of causing harm. Higher death rates in the hospital setting for NA patients may reflect the geographic diversity of NA patients in the state of New Mexico, and UNMH's status as the only Level 1 trauma center and “catchment” hospital for the state.

Limitations

One potential bias to our results might include the possibility that either referred patients or their physicians were more open to palliative care and EOL discussions than other, nonreferred patient/physician care teams. In addition, for NA patient referrals, many of the palliative care providers believe that a frequent trigger of consultation is concern on the part of professional caregivers of “causing harm” by making “cultural mistakes.”

This study was conducted with the population of NAs living in the Southwestern United States. It is unknown if this research would be applicable to other Native populations outside of the American Southwest. Lastly, patient and family satisfaction were not formally assessed, and will be the focus of future research.

Conclusions

Our work with Southwestern NA patients and their families has demonstrated that conversations about EOL issues can be held with NA patients, if done in a respectful, culturally appropriate manner. To be most effective, EOL discussions must be individually tailored and without assumptions, as we have found that the differences within a culture are often as great as between cultures. Health care providers are encouraged to learn as much as possible about the values and beliefs of the individuals and populations for which they provide care and abandon the myths that may obstruct EOL conversations with NAs.

Footnotes

Acknowledgments

The authors wish to thank Betty Skipper, Professor, Department of Community and Family Medicine, Edward Bedrick, Professor, Department of Internal Medicine, and Jim Little, RN, Manager, Outcomes Research at the University of New Mexico Hospital for their assistance with data analysis and statistical support.

Author Disclosure Statement

No competing financial interests exist.

References

Indian Health Services Fact Sheet, US DHHSUpdatedJanuary 26, 2006. www.ihs.gov/PublicInfo/PublicAffairs/Welcome_Info/ThisFacts.asp. 2012 January 9.

2000-2001 Trends in Indian Health. U.S. DHHS IHS U.S. Printing Office 2004UpdatedApril 30, 2007. www.ihs.gov/NonMedicalPrograms/IHS_Stats/Trends00.asp. 2012 January 9.

National Center for Health StatisticsUpdatedOctober 6, 2006. www.cdc.gov/nchs/fastats/lcod.htm. 2012 January 9.

New Mexico Department of Health, Racial and Ethnic Health Disparities Report Card, August 2010. www.health.state.nm.us/dpp/2010ReportCard.pdf. 2012 January 9.

Lorenz

, Lynn

, Dy

, Shugarman

, Wilkinson

, Mularski

, Shugarman

, Wilkinson

, Mularski

, Morton

, Hughes

, Hilton

, Maglione

, Rhodes

, Rolon

, Sun

, Shekelle

. Evidence for improving palliative care at the end of life: A systematic review. Ann Intern Med, 2008; 148:147–159.

Temel

, Greer

, Muzikansky

, Gallagher

, Admane

, Jackson

, Dahlin

, Blinderman

, Jacobsen

, Pirl

, Billings

, Lynch

. Early palliative care for patients with metastatic non-small-cell lung cancer. New Engl J Med, 201019363:733–742.

Leung

, Tsai

, Cheng

, Liu

, Chiu

, Wu

, Chen

. Can a good death and quality of life be achieved for patients with terminal cancer in a palliative care unit? J Palliat Med, 2010; 13:1433–1438.

Teno

, Clarridge

, Casey

, Welch

, Wetle

, Shield

, Mor

. Family perspectives on end-of-life care at the last place of care. JAMA, 2004; 291:88–93.

Casarett

, Pickard

, Bailey

, Ritchie

, Furman

, Rosenfeld

, Shreve

, Chen

, Shea

. Do palliative care consultations improve patient outcomes? J Am Geriatr Soc, 2008; 56:593–599.

10.

Higginson

, Finlay

I.G.

, Goodwin

D.M.

et al.

Is there evidence that palliative care teams alter end-of-life experiences of patients and their caregivers?

J Pain Symptom Manage, 25. 2003; 150–168.

11.

Manfredi

, Morrison

, Morris

, Goldhirsch

, Carter

, Meier

. Palliative care consultations: How do they impact the care of hospitalized patients? J Pain Symptom Manage, 2000; 20:166–173.

12.

Domer

, Kaur

. Palliative practice in Indian health. S D Med, 2008Spec No.36–40.

13.

Baldridge

. Moving Beyond Paradigm Paralysis: American Indian End-of-Life Care. CDC Health Aging Program, 2011; 1–12. www.chronicdisease.org. 2012 January 9.

14.

AHRQ: 2010 National Healthcare Disparities ReportUpdatedFebruary 2011. AHRQ Pub No. 11-0005. www.ahrq.gov/qual/grdr10.htlm. 2012 January 9.

15.

Gorospe

. Establishing palliative care for American Indians as a public health agenda. Internet J Pain Symptom Control Palliat Care, 2006; 4. www.ispub.com/journal/the-internet-journal-of-parn-symptom-control-and-palliative-care/volume-4-number-2/establishing-palliative-care-for-american-indians-as-a-public-health-agenda.html. 2012 January 9.

16.

Carrese

, Rhodes

. Western bioethics on the Navajo reservation. Benefit or harm? JAMA, 1995; 274:826–829.

17.

Carrese

. Rhodes LAA Bridging cultural differences in medical practice. The case of discussing negative information with Navajo patients. J Gen Intern Med, 2000; 15:92–96.

18.

Kitzes

. Guidelines for discussing palliative care and end-of-life planning in American Indian/Alaska Native communities. IHS Primary Care Provider, 2002; 27:89–92.

19.

Struthers

, Lowe

. Nursing in the Native American culture and historical trauma. Issues Ment Health Nurs, 2003; 24:257–272.

20.

Kitzes

, Domer

. Palliative care: An emerging issue for American Indians and Alaskan Natives. J Pain Palliat Care Pharmacother, 2003; 17:201–210.

21.

DeCourtney

, Jones

, Merriman

, Heavener

, Branch

. Establishing a culturally sensitive palliative care program in rural Alaska Native American communities. J Palliat Med, 2003; 6:501–510.

22.

Kitzes

, Berger

. End-of-life issues for American Indians/Alaska Natives: Insights from one Indian Health Service area. J Palliat Med, 2004; 7:830–838.

23.

Guidelines for Palliative Care Services in the Indian Health SystemUpdatedDecember 2006. www.ihs.gov/nonmedicalprograms/NC4/Documents/FINALGUIDELINES.pdf. 2012 January 9.

24.

Kitzes

. Palliative medicine, intractable pain, and end of life care. IHS Primary Care Provider, 1998; 23:143–144.

25.

Finke

, Bowannie

, Kitzes

. Palliative care in the Pueblo of Zuni. J Palliat Med, 2004; 7:135–143.

26.

Murphy

, Burrows

, Santilli

, Kemp

, Tenner

, Kreling

, Teno

. The Influence of the probability of survival on patients' preferences regarding cardiopulmonary resuscitation. New Engl J Med, 1994; 330:545–549.

27.

Weeks

, Kofoed

, Wallace

, Welch

. Advance directives and the cost of terminal hospitalization. Arch Intern Med, 1994; 154:2077–2083.

28.

Matsuyama

, Reddy

, Smith

. Why do patients choose chemotherapy near the end of life? A review of the perspective of those facing death from cancer. J Clin Oncol, 2006; 24:3490–3496.

29.

Hagerty

, Butow

, Ellis

, Lobb

, Pendlebury

, Leighl

, MacLeod

, Tattersall

MHN

. Communicating with realism and hope: Incurable cancer patients views on the disclosure of prognosis. J Clin Oncol, 2005; 23:1278–1288.

30.

Mitchell

. Reluctance to disclose difficult diagnoses: A narrative review comparing communication by psychiatrists and oncologists. Supp Care Cancer, 2007; 15:819–828.

31.

Wright

, Zhang

, Ray

, Mack

, Trice

, Balboni

, Mitchell

, Jackson

, Block

, Maciejewski

, Prigerson

. Associations between end-of-life discussions, patient mental health, medical care near death, and caregiver bereavement adjustment. JAMA, 2008; 300:1665–1673.