Being Aware of the Prognosis: How Does It Relate to Palliative Care Patients' Anxiety and Communication Difficulty with Family Members in the Hong Kong Chinese Context?

Abstract

Chinese family members play an important role in end-of-life decision making. Culturally, they may believe that a patient's awareness of a poor prognosis could harm his or her well-being, and thus they may avoid telling the truth to the patient. To examine this cultural belief empirically, this study aimed to explore the relationships among the patient's awareness of the prognosis, the family's awareness of the prognosis, the patient's anxiety, and difficulty in communicating with family members. Clinical data mining was the research method. Clinical records of deceased cancer patients in the palliative care unit of a hospital during 2003–2005 were reviewed, and 935 patients were included in this study. Logistic regression analyses indicated that patients who did not have a clear awareness of their prognosis were more likely to experience anxiety (odds ratio [OR] = 1.44, 95% CI = 1.14–1.82, p = 0.002) and difficulty communicating with family members (OR = 2.11, 95% CI = 1.62–2.76, p < 0.001). Patients whose family members were less aware of the prognosis were more likely to experience difficulty communicating with family members (OR = 4.48, 95% CI = 1.93–10.42, p < 0.001). This study further suggests that awareness of prognosis may not harm Chinese patients. Instead, unrealistic incomplete awareness of prognosis by patients and family members may be associated with patient anxiety and difficulty communicating.

Introduction

Concerns about whether patients are aware of their diagnosis and prognosis in end-of-life care have been paramount for many years, since Glaser and Strauss's study on awareness of dying among terminally ill patients.¹ Recent efforts have been put into conducting a systematic review of the previous studies of end-of-life communication. For example, one review concluded that health care professionals often underestimated patients' need for information but overestimated their understanding and awareness of the prognosis.² Another stated that the majority of patients would like health care professionals to disclose prognostic information honestly in end-of-life care. However, some might prefer ambiguity in disclosure, and it was suggested that health care professionals conduct individualized assessments of patients' preferences.³ Although these reviews have clear criteria for inclusion of the studies, irrespective of the ethnicity of patients, most have mainly included studies done in Western countries. However, the matter of disclosure of diagnosis and prognosis in end-of-life care in other cultures is equally important, considering that the goal of achieving a “good death” is universal.^4,5 For example, in the Chinese context, the number of research studies on truth telling and patients' awareness of diagnosis and prognosis has been increasing.^6–11 Culturally, it is believed that Chinese patients should not be told the diagnosis or prognosis in end-of-life care.¹² Empirically, such a claim violates the preference of Chinese patients.^7,8,13 Specifically, these studies suggest that the majority of Chinese patients want to be informed of the diagnosis and prognosis,¹³ but some may want to know the diagnosis in the earlier stage of illness rather than in the terminal stage.⁷ In addition, Chinese patients have been shown to be more eager to know the diagnosis and prognosis than are their family members, and they think they should be informed directly instead of via family members.^7,8

Although these studies seem to indicate that Chinese patients welcome the disclosure of diagnosis and prognosis in end-of-life care, our clinical experience has told us that Chinese family members avoid telling patients the truth. The main reason for nondisclosure is often nonmaleficence.^12,14 Chinese family members and patients are often interdependent.^15,16 Patients often rely on family members for emotional support and decision making,^17,18 and caregivers' support and acceptance may be considered an essential personal resource that may be associated with better psychosocial outcomes of Chinese palliative care patients.¹⁹ As a result, family members still play a crucial role in deciding whether patients should know the truth. As long as they consider disclosure of diagnosis or prognosis harmful to the patient, they may object to the disclosure and persuade health care professionals not to tell the patient the truth.¹⁰

However, it has been argued that nondisclosure of the truth may be even more harmful to palliative care patients.²⁰ To examine this notion empirically, Hinton²¹ found that depression was associated only with caring relatives' greater awareness of patients dying. No significant relationship was found between patients' awareness of dying and their level of depression. In another study, findings showed that depression was almost 3 times greater in patients who did not acknowledge their prognosis than in those who acknowledged or partially acknowledged their prognosis.²² In the Chinese context, some studies have been conducted to show the benefits of truth telling. For example, Leung, Chiu, and Chen²³ showed that awareness of terminal illness was associated significantly with better spiritual well-being in terminal cancer patients in Taiwan. Another study showed that cancer patients in Taiwan who were aware of the diagnosis had higher levels of hope and internal locus of control than those who were not informed.¹¹ Moreover, patients who were aware of the diagnosis of cancer also experienced significantly lower level of pain intensity, pain interference, and higher level of satisfaction with pain management.²⁴ All these studies were conducted in the Taiwan Chinese context, but little is known about the situation in other Chinese contexts such as Hong Kong. Also, these studies mainly dichotomized patients into either being aware or unaware of the diagnosis and/or prognosis, and did not consider different levels of awareness. In addition, the relationship of Chinese patients' awareness of diagnosis and/or prognosis with other key psychosocial outcomes such as anxiety and communication with family members in end-of-life care^20,25–28 remains unexplored. Considering the interdependence between Chinese family members and patients, in addition to exploring the role of patients' awareness of diagnosis and prognosis on the psychosocial outcomes of patients, the role of family members' awareness of diagnosis and prognosis on the psychosocial outcomes of patients should be explored. Previous studies have shown limited attention to the effect of family members' awareness of illness on patient outcomes.

Therefore, this study aimed at exploring the relationships among patients' awareness of prognosis, family members' awareness of prognosis, and two psychosocial outcomes of patients: patients' anxiety and communication with family members.

Methods

A practice-based research approach called clinical data mining was adopted in this study.^29,30 The essence is to retrieve and analyze the clinical data from practice for research purposes.

Participants

Data of the participants in this study came from patients' records in a palliative care unit of a public hospital in Hong Kong. All patients who used the services of that unit and died between 2003 and 2005 were included in this study. Approval was granted by the research ethics committee of the hospital. A data retrieval form was developed to code the clinical data from the patients' records and input them into the computer program for statistical analysis.

Types of clinical data included

In this study, the following clinical data were retrieved: demographics such as age and gender, type of terminal illness, and length of service in palliative care; and independent variables such as Palliative Performance Score (PPS), level of financial difficulty, family's insight into prognosis, and patient's insight into prognosis. Dependent variables include patient's anxiety level and communication between patient and family. PPS is often used to assess the severity of a palliative care patient's physical condition according to the patient's performance in five areas—ambulation, activity and evidence of disease, self-care, intake, and consciousness level—and is used to predict the prognosis, such as length of survival.³¹ The score ranges from 0 (death) to 100 (fully functional in all 5 areas). Other variables (level of financial difficulty, family's insight into illness, patient's insight into illness, patient's anxiety level, and communication between patient and family) were measured by the Support Team Assessment Schedule (STAS).^32,33 STAS is a clinical assessment tool commonly used in palliative care,^32–36 including in the Hong Kong Chinese context.³⁷ Validity of STAS was supported by the correlations between professionals' ratings, and patients' and families' ratings, as well as by the score of the Hebrew Rehabilitation Centre for the Aged Quality of Life index (HRCA-QL). The reliability, including the inter-rater reliability, internal consistency, and test-retest reliability were also found to be satisfactory.^33,38 The existing version used in this study is the modified one, which suits the needs of inpatient palliative care use.³³ It assesses the level of condition in different physical (e.g., pain, dyspnea) and psychosocial symptoms. In this study, we focused only on the psychosocial variables mentioned above. Each variable was rated 0 to 4. A higher score means a poorer condition in that item. For example, a patient's insight into illness was rated as: 0 = full awareness of prognosis, 1 = prognosis under- or overestimated by 200%, 2 = uncertain about becoming well or long-term prognosis, 3 = unrealistic (e.g., expecting to return to normal activity or work for a year, when the prognosis is only 3 months), 4 = expecting to become completely well. Details of the rating of each variable measured by STAS in this study are shown in Table 1. All clinical data were collected on each patient's admission to palliative care. Nurses in the palliative care unit were responsible for completing the STAS assessments through the clinical interviews with patients. A nurse specialist was responsible for training and supervising these nurses, to ensure the validity and reliability of assessments.

Table 1.

Definition of Ratings in STAS

Items in STAS		Definition of ratings
Level of financial difficulty	0=	all entitled benefits received, managing
	1=	one benefit desirable, patient and family managing, matter may be in hand
	2=	one benefit required urgently, several benefits desirable, patient and family experiencing difficulties managing
	3=	urgent need for several benefits, barely managing
	4=	not managing, entitled to many benefits, in chaos and none in hand
Patient's/Family's insight into illness	0=	full awareness of prognosis
	1=	prognosis under- or overestimated by 200%
	2=	uncertain about becoming well or long prognosis
	3=	unrealistic (e.g., expecting to return to normal activity or work for a year, when the prognosis is for 3 months)
	4=	expecting to become completely well
Patient's anxiety level	0=	none
	1=	worry over changes, no physical or behavioral symptoms of anxiety; concentration not affected
	2=	waiting for changes or problems: on edge; occasional physical or behavioral symptoms of anxiety
	3=	often anxious, physical/behavioral symptoms; concentration markedly affected
	4=	completely and continuously preoccupied with anxiety and worries; unable to think of other matters
Communication between patient and family	0=	communicating openly and honestly, verbally and nonverbally
	1=	communicating openly at times or with some family members
	2=	acknowledge condition but discussion does not satisfy patient or family, who feel full implications are not discussed; maybe with partner but not family
	3=	out of step, all discussion guarded
	4=	pretending

Data analysis

SPSS version 11.5 (SPSS Inc., Chicago, IL) was used for statistical analysis. Descriptive statistics were conducted to show demographics and other independent and dependent variables. Due to the skewed distribution of two dependent variables, patient's anxiety level and communication between patient and family, they were recoded to dichotomous variables. The original score of 0 remained unchanged, but the scores of 1 to 4 were recoded as 1. Such categorization may reflect whether patients experienced any distress in anxiety and communication with family members (e.g., 0 = no difficulty in communicating with family members, 1 = having difficulty in communicating with family members). No recoding was conducted for the other two independent variables, patient's insight into prognosis and family's insight into prognosis. The reason is that the distribution of ratings in patient's insight into prognosis was relatively less skewed and more dispersed (e.g., 49.5% of patients were rated a score of 0, but still 25.5% and 10.7% of patients were rated 1 and 2, respectively) and thus this variable was treated as continuous. In turn, family's insight into prognosis was deliberately treated as continuous, to better compare the relationships with dependent variables between these two independent variables (i.e., patient's insight into prognosis versus family's insight into prognosis). Finally, logistic regression was conducted to explore the associations among the patient's insight into prognosis, family's insight into prognosis, and these two recoded dependent variables. Demographics, including age, gender, PPS, and level of financial difficulty, were entered first as control variables. Family's insight into prognosis was then entered to explore the influence of the family on patient outcomes. Finally, patient's insight into prognosis was entered to explore the associations with the anxiety level and communication with family members.

Results

A total of 935 Chinese palliative care patients were included in this study. There were more male patients (57.1%) than female (42.9%). The majority were older adults (mean age = 74.1) and had limited education. Nearly one-third were illiterate (32.4%). Details of patients' demographics are shown in Table 2.

Table 2.

Demographics of Palliative Care Patients

Items	N	%
Gender
Male	534	57.1
Female	401	42.9
Age	Mean = 74.1 (SD 11.2)
Length of time in palliative care	Mean = 53.6 days; Median = 27 days
Type of cancer
Lung cancer	400	42.9
Colorectal cancer	139	14.9
Liver cancer	88	9.4
Stomach cancer	58	6.2
Breast cancer	23	2.5
Other	227	24.1
Marital status
Single	84	9
Married	493	52.8
Widowed	312	33.4
Separated/Divorced	44	4.7
Education level
Illiterate	266	32.4
Primary education	360	43.8
Secondary education	158	19.2
Tertiary education	38	4.6

Due to missing data on education level and marital status, total number of patients on those items does not equal 935. Percentage was rounded to nearest 0.1, and thus the total percentage of some items does not equal exactly 100%.

Patient's and family's insights into prognosis

About half the patients were fully aware of their prognosis (49.5%). Only one patient was totally unaware of the prognosis and expected to become completely well. Some patients were too weak to have their insight into prognosis assessed upon admission to palliative care (10.7%). Most family members were fully aware of patients' prognosis (82.5%). Distribution of ratings for patient's and family's insight into prognosis is shown in Table 3.

Table 3.

Distribution of Ratings in Patient's and Family's Insight into Prognosis, Patient's Anxiety Level, and Communication between Patient and Family

STAS score	Patient's insight into prognosis (%)	Family's insight into prognosis (%)	Patient's anxiety level (%)	Communication between patient and family (%)
0	49.5	82.5	42.6	65.6
1	25.5	3.3	40.6	18.2
2	10.7	0.4	5.3	4.7
3	1.4	0.1	0.1	0.3
4	0.1	0	0	0.1
Unable to assess (e.g., too weak)	10.7	Not applicable	9.3	8.8
Missing data	2.1	13.7	2.0	2.4

Percentage was rounded to the nearest 0.1, and thus the total percentage of some items does not equal exactly 100%.

Patient's anxiety level and communication between patient and family

Original distribution of ratings in patient's anxiety level and communication between patient and family is shown in Table 3. After recoding to dichotomous variables, slightly more patients experienced anxiety (46%) than those who did not experience anxiety (42.6%). In addition, fewer patients experienced difficulty communicating with family members (23.3% versus 65.6%).

Relationships among patient's insight into prognosis, family's insight into prognosis, patient's anxiety, and difficulty communicating between patient and family

After controlling the effects of physical and sociodemographic variables, it was shown that patients who were less aware of the prognosis of their illness (higher scores in patient's insight into illness) were associated significantly with patient's anxiety (OR = 1.44, 95%; CI = 1.14–1.82, p = 0.002). In other words, they were more likely to experience anxiety than patients who were aware of the prognosis. The family's insight into the illness did not show significant association with patient's anxiety. Similarly, after controlling the effects of physical and sociodemographic variables, patients who were less aware of the prognosis of their illness were associated significantly with difficulty communicating between patient and family (OR = 2.11, 95% CI = 1.62–2.76, p < 0.001), that is, they were more likely to experience difficulty communicating with family members. In addition, higher scores in family's insight into illness were associated significantly with communication difficulty between patient and family (OR = 4.48, 95%CI = 1.93–10.42, p < 0.001). This means that patients were more likely to experience communication difficulty with family members when the latter were less aware of the prognosis of the illness. Detailed results of logistic regression are shown in Table 4.

Table 4.

Relationships among Awareness of Prognosis, Patient's Anxiety, and Communication Difficulty between Patient and Family

	Patient's anxiety					Communication difficulty between patient and family
	β	Wald	OR	95% CI	P	β	Wald	OR	95% CI	P
Age	−0.03	13.44	0.97	0.95–0.97	<0.001	−0.01	0.23	1.00	0.98–1.02	0.63
Gender	0.20	1.00	1.22	0.83–1.78	0.317	0.09	0.15	1.10	0.69–1.74	0.70
Palliative Performance Score	−0.02	3.42	0.99	0.97–1.00	0.064	−0.01	0.80	0.99	0.97–1.01	0.37
Marital status
Single	−0.47	0.74	0.63	0.22–1.82	0.391	0.02	0.002	1.03	0.32–3.27	0.97
Married	0.38	0.87	1.47	0.65–3.29	0.352	−0.78	2.85	0.46	0.19–1.13	0.09
Widowed	0.37	0.74	1.45	0.62–3.37	0.391	−0.67	1.87	0.51	0.20–1.34	0.17
Education level
No education received	0.05	0.06	1.05	0.71–1.55	0.806	−0.01	0.003	0.99	0.62–1.57	0.96
Financial difficulty	0.44	12.79	1.55	1.22–1.97	<0.001	0.22	2.34	1.25	0.94–1.65	0.13
Family's insight into illness	0.33	0.76	1.39	0.66–2.92	0.385	1.50	12.14	4.48	1.93–10.42	<0.001
Patient's insight into illness	0.37	9.52	1.44	1.14–1.82	0.002	0.75	30.60	2.11	1.62–2.76	<0.001

Wald refers to the Wald test, which is a kind of parametric statistical test used in the logistic regression. For details refer to http://en.wikipedia.org/wiki/wald_test

Discussion

A key reason for nondisclosure of prognosis to palliative care patients is that such awareness may be harmful to their well-being. This belief is particularly held in Chinese culture, which emphasizes familial collectivism.^39–41 However, results of this study show that palliative care patients who were more aware of their prognosis were not more likely to experience anxiety or communication difficulty with family members. Instead, it was found that patients who were less aware of the prognosis tended to experience anxiety and communication difficulty with family members. Conceptually and philosophically, Chinese scholars have argued that Chinese culture did not necessarily favor nondisclosure of truth.¹² Findings of this study have provided further but preliminary evidence that it may be a cultural myth that Chinese patients' lack of awareness of their prognosis is beneficial.

One of the limitations of this study is that we only assessed patients' awareness of prognosis. We did not know whether limited awareness came from nondisclosure or other reasons, such as patient denial. If the reason is patient denial, such psychological defense may not help to reduce anxiety but rather result in difficulty communicating with family members. However, because many family members still believe that, in the Hong Kong Chinese context, truth telling may harm patients, we argue that at least some limited awareness of the prognosis was due to the nondisclosure of diagnosis and prognosis. In this sense, our findings may suggest that deceit as more harmful in end-of-life communication may also be relevant in non-Western cultures such as Hong Kong.²⁰

One possibility of the associations among patients' prognosis, anxiety, and communication difficulty with family members is that patients who were more aware of their prognosis may be more likely to engage in end-of-life planning and discussion.⁴² This may lead to a lower level of anxiety and better communication with family members. Further studies may also reveal how limited awareness of the prognosis of a terminal illness may affect palliative patients' anxiety level and communication with family members. In addition, future studies may explore whether other factors such as the patient's willingness to know the prognosis³ and peacefulness in knowing the prognosis⁴³ may mediate or moderate the associations.

Our findings also show that patients whose family members were less aware of their prognosis were more likely to experience communication difficulty with their family members. Similarly, such a finding may be explained by the fact that family members who were less aware of the prognosis may be less sensitive to the needs of patients, including their desire for communicating their end-of-life preferences. Yet, an alternative interpretation may be more relevant in the clinical situation of Hong Kong. In Hong Kong, almost all family members know the prognosis of patients before they are admitted to palliative care units. Thus, if they still show limited awareness of the patient's prognosis, this may reflect their difficulty in accepting the reality, that is, the poor prognosis. As a result, one reason for our findings is that family members may be affected by psychological distress, such as difficulty in accepting the patient's illness, and this may affect the quality of communication with patients. For example, one previous study showed that family caregivers' emotional exhaustion, depression, and self-efficacy may predict the level of open communication with terminal cancer patients.⁴⁴

Limitations and future research

Although STAS is commonly used in the palliative care setting in Hong Kong, future studies should pay more attention to its psychometrics in the Hong Kong Chinese context, especially when it is used in the inpatient setting. Such a concern has been expressed by Carson and his colleagues.⁴⁵ STAS was validated originally by the community samples of palliative patients in the UK.⁴⁶ However, the study by Carson and associates has indicated that reliability and validity of STAS may be questionable when it is used in other clinical settings and patient populations, such as the inpatient palliative setting in Canada. In this study, we are aware that the majority of ratings in different STAS items were skewed, and this may raise the question of whether the measurement can adequately assess the range of conditions of patients. Whether this issue of measurement can be improved by implementing more in-depth training in the use of STAS, or whether we should consider other measurements that are more valid in the Hong Kong Chinese context is the next step on which to focus.

Because we were limited by the cross-sectional research design of this study, causal relationships among patients' awareness of prognosis, anxiety, and communication difficulty with family members cannot be established. However, in addition to the previous findings in the literature, this study may provide further preliminary evidence of the benefits of prognostic awareness to patients' psychosocial outcomes in the Chinese context. In the future, studies with more vigorous methodology, such as a longitudinal study of different groups of patients with various levels of prognostic awareness, should be conducted to examine the casual relationships.

Conclusion

For now, the findings suggest that helping professionals should by no means accept nonmaleficence as the one correct reason for nondisclosure of truth to Chinese palliative care patients. Family members should be helped to consider carefully the possible pros and cons of telling patients the diagnosis and prognosis, such as the effect on patient anxiety and quality of communication with family members.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Glaser

, Strauss

. Awareness of Dying. Chicago: Aldine, 1965.

Hancock

, Clayton

, Parker

, Walder

, Butow

, Carrick

, Currow

, Ghersi

, Glare

, Hagerty

, Tattersall

. Discrepant perceptions about end-of-life communication: A systematic review. J Pain Symptom Manage, 2007; 34:190–200.

Innes

, Payne

. Advanced cancer patients' prognostic information preferences: A review. Palliat Med, 2009; 23:29–39.

Abramovitch

. “Good death”and“bad death”: Therapeutic implications of cultural conceptions of death and bereavement. Malkinson

, Rubin

, Witztum

. Traumatic and Non-traumatic Loss and Bereavement: Clinical Theory and Practice. Madison: Psychosocial Press, 2000; 255–272.

Chan

WCH

, Tse

, Chan

THY

. What is good death: Bridging the gap between research and intervention. Chan

CLW

, Chow

AMY

. Death, Dying and Bereavement: A Hong Kong Chinese Experience. Hong Kong: Hong Kong University Press, 2006; 127–135.

, Liu

, Zou

, Huang

, Yu

, You

, Jiang

, Li

, Jiang

. To tell or not to tell: Attitudes of Chinese oncology nurses towards truth telling of cancer diagnosis. J Clin Nurs, 2008; 17:2463–2470.

Jiang

, Liu

, Li

, Huang

, Yao

, Zhang

, Yao

, Du

, Chen

, Xie

, Zhao

, Wei

. Different attitudes of Chinese patients and their families toward truth telling of different stages of cancer. Psycho-Oncology, 2007; 16:928–936.

Tang

, Liu

, Lai

, Liu

, Chen

, Koong

. Congruence of knowledge, experiences, and preferences for disclosure of diagnosis and prognosis between terminally-ill cancer patients and their family caregivers in Taiwan. Cancer Invest, 2006; 24:360–366.

Tang

, Lee

S-Y

. Cancer diagnosis and prognosis in Taiwan: Patient preferences versus experiences. Psycho-Oncology, 2004; 13:1–13.

10.

Kendall

. Being asked not to tell: Nurses' experience of caring for cancer patients not told their diagnosis. J Clin Nurs, 2006; 15:1149–1157.

11.

Lin

, Tsay

. Relationships among perceived diagnostic disclosure, health locus of control, and levels of hope in Taiwanese cancer patients. Psycho-Oncology, 2005; 14:376–385.

12.

Tse

, Chong

, Fok

. Breaking bad news: A Chinese perspective. Palliat Med, 2003; 17:339–343.

13.

Fielding

, Hung

. Preferences for information and involvement in decisions during cancer care among a Hong Kong Chinese population. Psycho-Oncology, 1996; 5:321–329.

14.

Mitchell

. Cross-cultural issues in the disclosure of cancer. Cancer Practice, 1998; 6:153–160.

15.

Chan

WCH

, Epstein

, Reese

, Chan

CLW

. Family predictors of psychosocial outcomes among Hong Kong Chinese cancer patients in palliative care: Living and dying with the “support paradox.” Soc Work Health Care, 2009; 48:519–532.

16.

Markus

, Kitayama

. Culture and the self: Implications for cognition, emotion and motivation. Psycholl Rev, 1991; 98:224–253.

17.

Liu

J-E

, Mok

, Wong

. Perceptions of supportive communication in Chinese patients with cancer: Experiences and expectations. J Adv Nurs, 2005; 52:262–270.

18.

Payne

, Chapman

, Holloway

, Seymour

, Chau

. Chinese community views: Promoting cultural competence in palliative care. J Palliat Care, 2005; 21:111–116.

19.

Chan

WCH

. Can personal resources make a difference? An exploratory study of Chinese cancer patients in palliative care. J Soc Work End-of-Life Palliat Care, 2009; 5:186–200.

20.

Fallowfield

, Jenkins

, Beveridge

. Truth may hurt but deceit hurts more: Communication in palliative care. Palliat Med, 2002; 16:297–303.

21.

Hinton

. The progress of awareness and acceptance of dying assessed in cancer patients and their caring relatives. Palliat Med, 1999; 13:19–35.

22.

Chochinov

, Tataryn

, Wilson

, Enns

, Lander

. Prognostic awareness and the terminally ill. Psychosomatics, 2000; 41:500–504.

23.

Leung

, Chiu

, Chen

. The influence of awareness of terminal condition on spiritual well being in terminal cancer patients. J Pain Symptom Manage, 2006; 31:449–455.

24.

Lin

. Dislosure of the cancer diagnosis as it relates to the quality of pain management among patients with cancer pain in Taiwan. J Pain Symptom Manage, 1999; 18:331–337.

25.

Werth

Jr , Blevins

. Psychosocial Issues Near the End Of Life: A Resource for Professional Care Providers. Washington, DC: American Psychological Association, 2006.

26.

Hack

, Degner

, Parker

. The communication goals and needs of cancer patients: A review. Psycho-Oncology, 2005; 14:831–845.

27.

Mystakidou

, Tsilika

, Parpa

, Katsouda

, Galanos

, Vlahos

. Assessment of anxiety and depression in advanced cancer patients and their relationship with quality of life. Quality Life Res, 2005; 14:1825–1833.

28.

Vachon

MLS

, Kristjanson

, Higginson

. Psychosocial issues in palliative care: The patient, the family, the process and outcome of care. J Pain Symptom Manage, 1995; 10:142–150.

29.

Epstein

. Clinical Data-Mining: Integrating Practice and Research. New York Oxford University Press, 2010.

30.

Epstein

. Using available clinical information in practice-based research: Mining for silver while dreaming of gold. Soc Work Health Care, 2001; 33:17.

31.

Anderson

, Downing

, Hill

. Palliative Performance Scale: A new tool. J Palliat Care, 1996; 12:5–11.

32.

Higginson

, Costantini

. Communication in end-of-life cancer care: A comparison of team asessments in three European countries. J Clin Oncol, 2002; 20:3674–3682.

33.

Higginson

. Audit methods: Validation and in-patient use. Higginson

. Clinical Audit in Palliative Care. Oxford and New York: Radcliffe Medical Press, 1993; 48–54.

34.

Butters

, Higginson

, George

, McCarthy

. Palliative care for people with HIV/AIDS: View of patients, carers and providers. AIDS Care, 1993; 5:105–116.

35.

Edmonds

, Higginson

, Altmann

, McDonnell . Is the presence of dyspnea a risk factor for morbidity in cancer patients? J Pain Symptom Manage, 2000; 19:15–22.

36.

Toscani

, Brunelli

, Miccinesi

, Costantini

, Gallucci

, Tamburini

, Paci

, Di Giulio

, Peruselli

. Predicting survival in terminal cancer patients: Clinical observation or quality-of-life evaluation? Palliat Med, 2005; 19:220–227.

37.

RSK

, Ding

, Chung

, Woo

. Prospective study of symptom control in 133 cases of palliative care inpatients in Shatin Hospital. Palliat Med, 1999; 13:335–340.

38.

Higginson

, McCarthy

. Validity of the support team assessment schedule: Do staff's ratings reflect those made by patients or their families? Palliat Med, 1993; 7:219–228.

39.

Hong

, Ip

, Chiu

. Cultural identity and dynamic construction of the self: Collective duties and individual rights in Chinese and American cultures. Soc Cognition, 2001; 19:251–268.

40.

. Culture, gender and self-close-other(s) connectedness in Canadian and Chinese samples. Euro J Soc Psychol, 2002; 32:93–104.

41.

Yang

. Chinese social orientation: An integrative analysis. Lin

, Tseng

, Yeh

. Chinese Societies and Mental Health. Hong Kong: Oxford University Press, 1995; 19–39.

42.

Jeong

SY-S

, Higgins

, McMillan

. The essentials of advance care planning for end-of-life care for older people. J Clin Nurs, 2010; 19,3–4:389–397.

43.

Ray

, Block

, Friedlander

, Zhang

, Maciejewski

, Prigerson

. Peaceful awareness in patients with advanced cancer. J Palliat Med, 2006; 9:1359–1368.

44.

Bachner

, Carmel

. Open communication between caregivers and terminally ill cancer patients: The role of caregivers' characteristics and situational variables. Health Commun, 2009; 24:524–531.

45.

Carson

, Fitch

, Vachon

MLS

. Measuring patient outcomes in palliative care: Reliability and validity study of the Support Team Assessment Schedule. Palliat Med, 2000; 14:25–36.

46.

Higginson

. Audit methods: A community schedule. Higginson

. Clinical Audit in Palliative Care. Oxford and New York: Radcliffe Medical Press, 1993; 34–47.