Can a Good Death and Quality of Life Be Achieved for Patients with Terminal Cancer in a Palliative Care Unit?

Abstract

Background:

Lack of evidence supporting the claim that palliative care can improve quality of life and promote good death in patients with terminal cancer.

Objectives:

This study was designed to evaluate the change of quality of life and quality of death over time and between patients of long and short survival in a palliative care unit.

Methods:

Patient demography, cancer sites, Eastern Cooperative Oncology Group (ECOG) status were collected at admission. Quality of life, including physical and psychological symptoms, social support, and spirituality was assessed daily after admission. Quality of death was assessed by a Good Death Scale (GDS) at admission and retrospectively for 2 days before death.

Results:

A total of 281 patients (52% women) were admitted and died in the study period. One hundred forty-five patients (51.6%) died within 3 weeks. Although those with short survival (<3 weeks) had more physical symptoms during the first week, there was no difference in quality of life dimensions at admission, at 1 week, and at 2 days before death between survival groups. Physical conditions deteriorated with time but other dimensions continued to improve until death. GDS and subdimensions continued to improve until death. Although those with long survival (≥3 weeks) have better scores for awareness, acceptance, timeliness, comfort, and GDS at admission, there was no difference between the two groups at 2 days before death.

Conclusion:

Under comprehensive palliative care, patients with terminal cancer can have good quality of life and experience a good death even with short survival.

Introduction

The World Health Organization (WHO) defines palliative care as “an interdisciplinary team approach that improves the quality of life of individuals and their families facing the problems associated with life-threatening illness, through the prevention and relief of suffering by means of early identification, assessment and treatment of pain and other problems, physical, psychosocial and spiritual.”¹ Thus, the aims of palliative and hospice care extend beyond improving quality of life, encompassing issues such as existential and spiritual aspects that are not included in the traditional quality of life agenda. Moreover, the quality of dying, or a “good death” has emerged as another important personal experience at the end-of-life condition.

There was no clear and consensual definition of good death in the literature. Some researchers have defined a good death as a personal evaluation of the dying experience as a whole, including expectations and value.² The Institute of Medicine (IOM) Committee on Care at the End of Life suggests that a good death is a condition that is “free from avoidable distress and suffering for patient, family and caregivers, in general accord with patient's and family's wishes, and reasonably consistent with clinical, cultural, and ethical standards.”³ The IOM concept extends the experience of good death from an individual to people closely related to the decedent and judges good death in terms of patient, family, and sociocultural preferences. Good death issues such as spiritual and existential issues, autonomy and control, hope, dignity, care tenor, burden to others, and aftermath concerns were not part of the concerns in medicine before the hospice movement thirty years ago.⁴

On the other hand, there is another line of thought for the definition of good death. McNamara⁵ challenged the notion of a shared standard of good death in the hospice model. There were increased concerns that the original hospice philosophy may be compromised by the increasing institutionalization, medicalization, and bureaucratization of hospice care when hospice care emerges into the mainstream medicine.^5,6 In the fieldwork in Australia, McNamara⁷ revealed ambivalence among palliative care practitioners in meeting the needs of the organization rather than the ideal of hospice philosophy and patients' choice. The emphasis of individual choice and control creates tension to the social and collective understanding and order of good death. “Good enough death” is one that supports the choice of the patient and perceived certainty of medical practice.⁵ In a literature review, Kehl⁸ concluded that the concept of good death was individualized, changeable over time, and related to personal perspective and experience.

Despite the rapid development and availability of hospice programs around the world, the benefit of hospice and palliative care still lacks high quality evidential proof.⁹ Moreover, there has been a paucity of information about the quality of life and good death for patients with terminal cancer in a hospital-based palliative care setting.¹⁰ It has been suggested that dying patients suffer from many types of loss, including physical, psychological, social, and existential (spiritual); and that they adapt to their losses through a cycle of three processes: comprehension, creative adaptation, and reintegration.¹¹ Physical comfort especially free from pain is the most important need for patients with terminal cancer,¹² then follows the higher order needs such as emotional and spiritual aspects.¹³ Understanding different trajectories of losses and adaptation in dying patients is helpful to individualize intervening strategies.

In a prospective cohort study among 11 hospice/palliative care organizations, researchers found that quality of life remained stable and physical wellness tended to improve at the end of life.¹⁴ A study among three palliative care units in Hong Kong revealed significant improvement in the worst physical symptoms and eating but deterioration in physical well-being, meaning of existence and self-satisfaction in the last 2 weeks of life.¹⁵ Both studies failed to reveal any improvement in psychosocial and spiritual domain of good death emphasized in palliative and hospice care.

In Taiwan, previous investigations in terminal cancer patients have found that good death includes peace of body and mind,¹⁶ awareness of the terminal condition,¹⁷ awareness of dying,^18,19 and being fearless about death.²⁰ Dying at home with the company of family members is considered a good death in Chinese traditional beliefs and the main goal of palliative and hospice care.²¹ But patients who died at home had less physical comfort than those who died in the palliative care unit.²¹

Despite the hospice movement around the world, there are still many barriers that prevent terminal patients from receiving palliative or hospice care.²² In the United States, national statistics revealed that the median length of stay in a hospice is approximately 3 weeks, and 10% of patients were admitted on their last day.²³ In Taiwan, the time when patients receive palliative or hospice care is generally late in the course of illness. But palliative and hospice care in Taiwan is limited by the lack of understanding of hospice care by the medical professionals and the public, patients and families being unwilling to give up curative therapy, limitations of national insurance reimbursement, lack of palliative and hospice care facilities are possible barriers. When caring for patients with terminal cancer in palliative and hospice care facilities, professionals confront a difficult situation in which a lot of things should be done in a very limited amount of time. A question that frequently arises for the palliative care team members is: Can the goal of good death be achieved within the present situation and standards of care, especially for the psychosocial and spiritual dimensions of good death, which depend on the rapport and trust among patients, families and the care team? To answer this question, we evaluated the quality of life and good death of patients with terminal cancer and compared the differences between different survival times after admission to a palliative care unit.

Methods

Study sample and setting

Participants in this prospective study were patients with terminal cancer who had been admitted to the palliative care unit of the National Taiwan University Hospital from October 2002 to December 2003. Patients who were unable to communicate (consciousness disturbance and aphonia, etc.), too sick to complete the interview at admission, and deceased within 1 week after admission were excluded. The study setting was a palliative care inpatient ward exclusively used for terminal cancer patient care. A multidisciplinary team, consisting of physicians, nurses, psychologists, social workers, Buddhist chaplains, and volunteer workers, was responsible to provide total care to the inpatients. The Buddhist chaplains provide company, life reviewing, fulfillment of spiritual needs, praying to God, etc., for the patients and their family.

This study was approved by the ethical committee at the National Taiwan University Hospital. Consent to participate in this study was obtained from patients or their legal proxy at admission.

Instruments and procedure

A complete set of data was obtained at the time of admission from patients, caregivers and medical records including patients' demography, primary cancer sites, Eastern Cooperative Oncology Group (ECOG) performance status,²⁴ quality of life, and conditions related to good death.

Quality of life measures were assessed daily after admission. The assessment form was the instrument used in our previous study, which demonstrated good psychometric properties.²⁵ It contains four dimensions: physical symptoms, psychological distress, social support, and spirituality. The physical symptom dimension has 16 items such as pain, weakness, fatigue, anorexia, etc., which were reported by each patient and recorded by the same medical staff. Pain, weakness, and fatigue were rated by a scale from 0 (none) to 10 (extreme), while the other symptoms were rated by a scale from 0 (none) to 3 (severe). Psychological distress items including anxiety, depression and aggression were rated on a scale from 1 (almost none) to 5 (extreme). Social support items including intimacy with family members and visit by friends or relatives were rated on a scale from 0 (very unlikely) to 6 (very likely). Information about psychological distress and social support was assessed by clinical psychologists. The spirituality dimension was evaluated by a 7-item cumulative scale with “fear and unaccepted death” at the lowest level and “spiritual sublimation” at the highest level. The items in the spiritual domain were assessed by Buddhist chaplains. Research team meetings were held once per week to review and collect the data for the study. The physical symptom score was calculated by collapsing the 10-point scale of the pain, weakness, and fatigue assessment into a 3-point scale equivalent to the scale of the other symptoms. All physical, psychological, social items were summed up into a total score for each area. The physical, psychological, social, and spirituality scores were then transformed into a scale of 0 to 15 for final analysis, with a higher score representing less physical and psychological difficulties, more social support and higher satisfaction in spirituality. Good death was assessed by a Good Death Scale (GDS)¹⁹ derived from the principles of appropriate death proposed by Weisman.²⁶ It contains 5 items including awareness that one is dying (0 = complete ignorance, 3 = complete awareness), accepting death peacefully (0 = complete unacceptance, 3 = complete acceptance), propriety (0 = no reference to the patient's will, 1 = following the family's will alone, 2 = following the patient's will alone, 3 = following the will of both the patient and their family), death timing (0 = no preparation, 1 = the family alone had prepared, 2 = the patient alone had prepared, 3 = both the patient and their family had prepared), and degree of physical comfort (0 = a lot of suffering, 1 = suffering, 2 = a little suffering, 3 = no suffering). The possible score ranges from 0 to 15. In a study of 366 patients, the internal consistency reliability of the GDS was 0.71.¹⁹ Following the death of each patient, a score was recorded to summarize the extent to which it was a good death. The opinion of each member was considered and the final score for each patient was decided jointly at the team meeting.

Statistical analysis

All data were analyzed by using SPSS 11.0 for Windows (SPSS Inc., Chicago, IL). Descriptive data were summarized as frequencies and percentages for categorical variables, mean and standard deviation for age and mean, standard deviation and median for survival time. Since there was a very wide range in survival time and mean value tended to deviate by extreme values, we used the median of survival time as a cutoff point for long and short survival. Chi-square statistics was used to compare differences in demographic data, location of primary cancer, and ECOG status between the short and the long survival group. Analysis of variance (ANOVA) was used to examine the equality of the means of physical symptoms, psychosocial distress, social support, spirituality, and GDS between the short and long survival groups. Change of quality of life and good death with time in each survival group was evaluated by paired t test.

Results

A total of 281 patients with terminal cancer (52.0% women) who fulfilled the enrollment criteria were admitted and died in our palliative ward in the study period and were included in the final analysis. Mean age of the patients was 64.4 ± 13.8 years. Mean survival time after admission was 35.4 ± 44.0 days (ranging from 7 to 418 days) and the median time of survival was 20 days. Eighty-nine patients (31.7%) died in the second week, 56 (19.9%) in the third week, 37 (13.2%) between the third week and 1 month, 55(19.6%) between 1 and 2 months, 25 (8.9%) between 2 and 3 months, and 19 patients (6.8%) died after 3 months. We defined patients surviving less than 3 weeks after admission as “short survival” and those surviving for 3 or more weeks as “long survival” according to the median time of survival in the following analysis. Comparisons between short and long survival patients revealed no statistical significant difference in age, gender, primary cancer sites, and ECOG performance status at admission (Table 1).

Table 1.

Demographic Data and Disease Characteristics of Terminal Cancer Patients

Variable	Survive < 3 weeks	Survive ≥ 3 weeks
Number of patients	145 (51.6%)	136 (48.4%)
Age	64.37 ± 14.03	64.40 ± 13.70
Survival days (range)	12.60 (7–20)	59.76 (21–418)^a
Gender
Male	76 (52.4%)	59 (43.4%)
Female	69 (47.6%)	77 (56.6%)
Primary cancer sites
Liver	27 (18.6%)	18 (13.2%)
Lung	20 (13.8%)	32 (23.5%)
Colorectal	20 (13.8%)	16 (11.8%)
Stomach/esophagus	17 (11.8%)	11 (8.1%)
ENT	9 (6.2%)	13 (9.6%)
Breast	8 (5.5%)	7 (5.1%)
Pancreas	7 (4.8%)	6 (4.4%)
Cervical/uterine/ovary	12 (8.3%)	10 (7.3%)
Urinary tract	7 (4.9%)	8 (5.9%)
Others	18 (12.5%)	15 (11.0%)
ECOG score at admission
1	8 (5.5%)	17 (12.5%)
2	23 (15.9%)	22 (16.2%)
3	56 (38.6%)	55 (40.4%)
4	58 (40.0%)	42 (30.9%)

p < 0.001.

Percentages in parenthesis.

ANOVA for age and survival days; χ² for gender, cancer sites, and ECOG.

ENT, ear, nose, and throat; ECOG; Eastern Cooperative Oncology Group; ANOVA; analysis of variance.

Comparison of quality of life dimensions between the two survival groups and with time

Table 2 compares the differences in quality-of-life dimensions in the palliative care unit for patients with terminal cancer of short and long survival. The groups showed no difference in physical symptoms, psychological symptoms, perceived social support or spirituality at admission. Patients with short survival had lower physical symptom scores than the long survival group at 1 week after admission, though this difference disappeared at 2 days before death. Table 3 presents the changes in quality of life dimensions from admission to the first week and from the first week to 2 days before death. Except for the change of physical symptom in the short survival group from admission to the first week, both groups have significant changes in all four quality of life dimensions in these two time periods. Among the four quality-of-life dimensions, the change of spirituality was the most obvious.

Table 2.

Comparison of Quality of Life Dimensions between the Two Survival Groups

	Quality of life dimensions	Survival < 3 weeks Mean ± SD (95% CI)	Survival ≥ 3 weeks Mean ± SD (95 % CI)
Admission	Physical	10.32 ± 1.85 (10.00–10.64)	10.92 ± 1.68 (10.62–11.21)
	Psychological	10.76 ± 2.96 (10.26–11.26)	11.20 ± 2.75 (10.73–11.68)
	Social support	10.76 ± 3.55 (10.25–11.28)	10.40 ± 2.97 (9.89–10.91)
	spirituality	7.35 ± 3.78 (6.74–7.96)	7.27 ± 3.78 (6.58–7.96)
One week	Physical	10.44 ± 1.87 (10.09–10.80)	11.69 ± 1.64 (11.40–11.99)^a
	Psychological	11.49 ± 3.14 (10.91–12.07)	12.07 ± 2.52 (11.62–12.53)
	Social support	11.40 ± 2.84 (10.90–11.91)	10.59 ± 2.79 (10.09–11.08)
	spirituality	9.72 ± 3.16 (9.15–10.28)	9.36 ± 3.48 (8.70–10.02)
Two days before death	Physical	9.94 ± 1.83 (9.58–10.30)	9.42 ± 2.00 (9.03–9.82)
	Psychological	12.44 ± 2.86 (11.91–12.98)	13.25 ± 1.96 (12.87–13.63)
	Social support	11.89 ± 2.71 (11.43–12.35)	11.57 ± 2.60 (11.09–12.04)
	spirituality	11.38 ± 2.69 (10.91–11.84)	11.83 ± 2.82 (11.31–12.35)

Statistical significance if no overlapping between the two confidence intervals.

(ANOVA) (2-tailed); analysis of variance; SD, standard deviation; CI, confidence interval.

Table 3.

Change of Quality of Life Dimensions from Admission to the First Week and from the First Week to Two days before Death

Survival	QOL dimensions	Admission—1 week Δ ± SD (95% CI)	1 week–2 days before death Δ ± SD (95% CI)
<3 weeks	Physical	−0.11 ± 1.47 (−0.38 – 0.17)	0.41 ± 0.98 (0.21 – 0.62)^a
	Psychological	−0.92 ± 2.08 (−1.31 – −0.54)^a	−0.89 ± 1.82 (−1.25 – −0.52)^a
	Social	−0.50 ± 1.34 (−0.74 – −0.26)^a	−0.68 ± 1.60 (−0.97 – −0.39)^a
	Spiritual	−2.31 ± 2.48 (−2.76 – −1.87)^a	−1.66 ± 2.38 (−2.10 – −1.22)^a
≥3 weeks	Physical	−0.79 ± 1.22 (−1.01 – −0.56)^a	2.11 ± 2.11 (1.68 – 2.54)^a
	Psychological	−0.84 ± 2.18 (−1.24 – −0.45)^a	−1.30 ± 2.66 (−1.84 – −0.76)^a
	Social	−0.26 ± 1.35 (−0.50 – −0.14)^a	−1.24 ± 2.54 (−1.72 – −0.76)^a
	Spiritual	−2.13 ± 2.31 (−2.57 – −1.69)^a	−2.72 ± 2.99 (−3.31 – −2.13)^a

Statistical significance if 95% confidence interval not includes zero.

Paired t-test (2-tailed); Δ: difference between two means; SD: standard deviation; QOL, quality of life.

Comparison of good death dimensions between the two survival groups and with time

Table 4 presents the differences of the five good death dimensions and the good death score between the short and long survival groups. At admission, patients with long survival had a higher score in awareness of terminal disease, acceptance of death, timeliness of death, comfort and the GDS. However, all the above differences disappeared and the two groups had a similar good death experience at 2 days before death. Table 5 presents the change of good death dimensions from admission to 2 days before death. All good death dimensions and the GDS improved with time after admission despite survival time.

Table 4.

Comparison of Good Death Dimensions between the Two Survival Groups

	Survival < 3 weeks Mean ± SD (95% CI)	Survival ≥ 3 weeks Mean ± SD (95% CI)
At admission
Awareness	2.18 ± 0.83 (2.03–2.32)	2.53 ± 0.73 (2.40–2.66)^a
Acceptance	2.26 ± 0.72 (2.14–2.38)	2.52 ± 0.56 (2.42–2.62)
Propriety	2.64 ± 0.75 (2.52–2.77)	2.66 ± 0.79 (2.52–2.80)
Timeliness	2.43 ± 0.93 (2.28–2.59)	2.72 ± 0.66 (2.60–2.84)^a
Comfort	2.02 ± 0.67 (1.91–2.14)	2.32 ± 0.67 (2.20–2.44)^a
GDS	11.51 ± 3.00 (11.00–12.03)	12.74 ± 2.36 (12.31–13.17)^a
Two days before death
Awareness	2.82 ± 0.44 (2.74–2.89)	2.89 ± 0.38 (2.82–2.96)
Acceptance	2.77 ± 0.49 (2.69–2.86)	2.87 ± 0.34 (2.81–2.93)
Propriety	2.87 ± 0.48 (2.79–2.95)	2.86 ± 0.51 (2.77–2.95)
Timeliness	2.85 ± 0.52 (2.76–2.94)	2.92 ± 0.35 (2.86–2.98)
Comfort	2.79 ± 0.43 (2.71–2.86)	2.71 ± 0.52 (2.61–2.80)
GDS	14.09 ± 1.70 (13.80–14.38)	14.26 ± 1.38 (14.01–14.50)

Statistical significance if no overlapping between the two confidence intervals.

ANOVA (2-tailed), analysis of variance; SD, standard deviation; CI, confidence interval; GDS, Good Death Score.

Table 5.

Change of Good Death Dimensions and Good Death Score from Admission to Two Days before Death

Survival	Good death dimensions	Admission–2 days before death Δ ± SD (95% Confidence interval)
<3 weeks	Awareness	−0.64 ± 0.72 (−0.76 – −0.52)^a
	Acceptance	−0.51 ± 0.64 (−0.62 – −0.040)^a
	Propriety	−0.23 ± 0.62 (−0.33 – −0.12)^a
	Timeliness	−0.42 ± 0.81 (−0.56 – −0.28)^a
	Comfort	−0.76 ± 0.69 (−0.88 – −0.64)^a
	GDS	−2.57 ± 2.60 (−3.01 – −2.12)^a
≥3 weeks	Awareness	−0.36 ± 0.66 (−0.48 – −0.25)^a
	Acceptance	−0.35 ± 0.53 (−0.44 – −0.25)^a
	Propriety	−0.20 ± 0.69 (−0.32 – −0.07)^a
	Timeliness	−0.20 ± 0.58 (−0.30 – −0.09)^a
	Comfort	−0.39 ± 0.60 (−0.50 – −0.29)^a
	GDS	−1.51 ± 2.11 (−1.90 – −1.13)^a

Statistical significance if 95% confidence interval not includes zero.

Paired t test (2-tailed); Δ; difference between two means; SD, standard deviation; GDS, Good Death Score.

Discussion

In this observational study, we found that although physical symptoms inevitably increased towards the end of the life, other quality of life dimensions such as psychological well-being, social supports, and spirituality could be improved and good death could be achieved by comprehensive care in a hospital palliative care setting. Contrary to the views of some of the health workers, different time trajectories i.e., short (<3 weeks) or long (≥3 weeks) survival, did not affect the acquisition of quality of life or good death in patients with terminal cancer. Both survival groups had a similar amount of improvement under the care of a palliative care team. Although the short trajectory posed a challenge to the delivery of care in the palliative ward, with the effort of the palliative care team, good death could be achieved within 3 weeks.

Physical symptoms are a major concern for patients and their families, and are generally the cause of admission to palliative ward. However, our results revealed that patients' physical symptoms continue to develop until death. Despite some initial success with pain control in the palliative ward, when the patients become sicker and closer to death, other physical symptoms inevitably progressed.

Unlike physical symptoms, we found that psychological well-being, social supports, and spirituality continuously improved, so most patients could achieve good death. This result has great clinical implications in palliative care. It implies that a holistic care approach including the physical, psychological, social, and spiritual aspects should be emphasized and steps should be taken to ensure all patients are able to receive holistic care service. In our palliative care team, clinical psychologists and Buddhism chaplains visits patients actively to assess their psychological and spiritual needs and arrange further visits according to patients' needs and choices.

Not all palliative care settings provide the same kinds of service. In a study about hospice providers' perceptions of symptom burdens at the end of life, most of the symptoms identified were physical distress such as lack of energy and pain, though hospice providers' generally lacked information to rate psychological symptoms and spiritual sufferings were not mentioned.²³ Kutner's observation reflected that some palliative care providers focus more on physical than psychological and spiritual suffering of terminal cancer patients, even though psychological and spiritual issues seemed to be more relevant to quality of death at the end of life. This discrepancy highlights the need for more education on psychosocial and spiritual care for palliative care providers.

The dying trajectory identified in this study for terminal cancer patients after palliative ward admission reveals some similarities and some differences with previous studies in other hospice/palliative care settings.^14,15 Kutner et al.¹⁴ showed that all symptoms and quality-of-life domain, including psychological and existential well-being, remained stable during the study period from 0 to 17 days after admission. In direct contrast to our study, the study by Lo¹⁵ found an improving trend in the physical domain, a deteriorating trend in the existential domain, and no significant changes in psychological, support, and sexuality domains. Differences in the philosophy of care, health care settings, patient characteristic, quality of care, and study methods may be responsible to the variation found in different studies. Great caution should be taken in generalizing the results of this study since this study was done in only one palliative care unit with a competent care team caring for patients with terminal cancer. Although we collected clinical data and held group meetings every week to discuss patients' conditions, we only included data at admission, at 1 week and 2 days before death in this study. In our study cohort, there was naturally continuous attrition due to death with time after one week of admission (those who survived less then 1 week were excluded in this study). Thus comparison of different time intervals after 1 week may be biased by comparing groups with different numbers of patients and the experience of patients who survived to the next time interval may differ from those who died in the previous interval.

In summary, this study suggests that most patients have the potential to achieve good death with the support of an interdisciplinary palliative care team. Our results need to be confirmed by future study with other well-established evaluation tools, in other patient groups and other palliative care settings. Factors that may affect the achievement of good death and life quality including patient characteristics, care settings, and quality of care, should be further studied.

Footnotes

Acknowledgments

We thank the patients and their families, and all the members of the palliative care unit for their contribution of this research. This study was supported by the National Science Council (NSC 91-2314-B-002-224) and the Department Of Health (DOH92-HP-1506), Executive Yuan, Taipei, Taiwan.

Author Disclosure Statement

No competing financial interests exist.

References

Sepúlveda

, Marlin

, Yoshida

, Ullrich

. Palliative care: The World Health Organization's Global Perspective. J Pain Symptom Manage, 2002; 24:91–96.

Stewart

, Teno

, Patrick

. The concept of quality of life of dying persons in the context of health care. J Pain Symptom Manage, 1999; 17:93–108.

Committee of Care at the End of Life, Institute of Medicine. Approaching Death: Improving Care at the End of Life. Washingtion, D.C.: National Academy Press, 1997.

Chochinov

. Dying, dignity, and new horizons in palliative end-of-life care. CA Caner J Clin, 2006; 56:84–103.

McNamara

. Good enough death: Atonomy and choice in Australian palliative care. Soc Sci Med, 2004; 58:929–38.

Hart

, Sainsbury

, Short

. Whose dying? A sociological critique of the good death. Mortal, 1998; 3:65–67.

McNamara

, Waddell

, Colvin

. The institutionalization of the good death. Soc Sci Med, 1994; 39:1501–1508.

Kehl

. Moving toward peace: An analysis of the concept of a good death. Am J Hospice & Palliat Med, 2006; 23:277–286.

Francke

. Evaluating research on palliative support teams: A literature review. Patient Education and Counseling, 2000; 41:83–91.

10.

Higginson

, Finlay

, Goodwin

, Cook

, Hood

, Edwards

, Douglas

, Norman

. Do hospital-based palliative teams improve care for patients or families at the end of life? J Pain Symptom Manage, 2002; 23:96–106.

11.

Knight

, Emanuel

. Processes of adjustment to end-of-life losses: A reintegration model. J Palliat Med, 2007; 10:1190–1198.

12.

Steinhauser

, Christakis

, Clipp

, McNeilly

, McIntyre

, Tulsky

. Factors considered important at the end of life by patients, family, physicians, and other care providers. JAMA, 2000; 284:2476–2482.

13.

Mako

, Galek

, Poppito

. Spiritual pain among patients with advanced cancer in palliative care. J Palliat Med, 2006; 9:1106–1113.

14.

Kutner

, Bryant

, Beaty

, Fairclough

. Time course and characteristics of symptom distress and quality of life at the end of life. J Pain Symptom Manage, 2007; 34:227–236.

15.

, Woo

, Zhoc

, Li

, Yeo

, Johnson

, Mak

, Lee

. Quality of life of palliative care patients in the last two weeks of life. J Pain Symptom Manage, 2002; 24:388–397.

16.

Chao CSC. The Meaning of Good Dying of Chinese Terminally Ill Cancer Patients in Taiwan[dissertation]Cleveland, OH: Frances Payne Bolton School of Nursing, Case Western Reserve University, 1993.

17.

Leung

, Chiu

, Chen

. The influence of awareness of terminal condition on spiritual well-being in terminal cancer patients. J Pain Symptom Manage, 2006; 31:449–456.

18.

Shao-Yi

Cheng

, Tai-Yuan

Chiu

, Wen-Yu

, Fei-Ran

Guo

, Yu

Wang

, Lin-Lin

Chou

, Ching-Yu

Chen

. A pilot study of good death in terminal cancer patients. Chin J Fam Med, 1996; 6:83–92.

19.

Cheng

, Hu

, Liu

, Yao

, Chen

, Chiu

. Good death study of elderly patients with terminal cancer in Taiwan. Palliat Med, 2008; 22:626–632.

20.

Tsai

, Wu

, Chiu

, Hu

, Chen

. Fear of death and good death among the young and elderly with terminal cancers in Taiwan. J Pain Symptom Manage, 2005; 29:344–351.

21.

Yao

, Hu

, Lai

, Cheng

, Chen

, Chiu

. Does dying at home influence the good death of terminal cancer patients? J Pain Symptom Manage, 2007; 34:497–504.

22.

Casarett

, Quill

. “I'm not ready for hospice”: Strategies for timely and effective hospice discussions. Ann Intern Med, 2007; 146:443–449.

23.

National Trend Summary 2004. Washington, DC: National Hospice and Palliative Care Organization, 2006.

24.

Oken

, Creech

, Tormey

, Horton

, Davis

, McFadden

, Carbone

. Toxicity and response criteria of the Eastern Cooperative Oncology Group. Am J Clin Oncol, 1982; 5:649–655.

25.

Tsai

, Wu

, Chiu

, Hu

, Chen

. Symptom patterns of advanced cancer patients in a palliative care unit. Palliat Med, 2006; 20:617–622.

26.

Weisman

. Appropriate death and the hospice program. Hosp J, 1988; 4:65–77.