Volume 39, Issue 4 , Pages 655-662, April 2010
Significance of Symptom Clustering in Palliative Care of Advanced Cancer Patients
Article Outline
Abstract
Patients with advanced cancer often experience multiple concurrent symptoms. To explore this symptom clustering and its associated parameters, we prospectively surveyed 427 consecutive patients on admission to the Palliative Care Unit. There were 222 males (52.0%) and 205 females (48.0%), with a median age of 66 years (range: 27–93 years). The main tumor sites were lung (19.9%), liver (18.0%), and colorectum (11.0%). The median survival was 13 days (1–418 days). Symptoms were assessed using a face-valid Symptom Reporting Form. We identified five symptom clusters by exploratory factor analysis. Clusters were named “loss of energy,” “poor intake,” “autonomic dysfunction,” “aerodigestive impairment,” and “pain complex.” We used nonhierarchical cluster analysis to divide the 394 patients with complete data into six groups. Each group was characterized by a particular pattern that was composed of different symptom clusters. Survival, functional performance, bone metastasis, and fluid accumulation were significantly associated with symptom clustering in six groups of patients. The severity of psychological distress also related to their physical deterioration. These data suggest that different underlying mechanisms associate with symptom clustering. Further elucidation of these processes may assist in symptom management.
Key Words: Symptom cluster, hospice palliative care, advanced cancer patients
Introduction
Patients with advanced cancer experience multiorgan failure, with many physical and psychological symptoms. Patients seek relief of suffering at the end stage of their lives.1 Consequently, the essential component of palliative cancer care is to provide them with symptom relief and better quality of life.2, 3 This may be challenging, and health care providers may have difficulty in developing symptom management strategies that can be applied across acute and home care settings.4 A lack of sufficient knowledge regarding effective therapeutic strategies is one of numerous factors that could interfere with adequate management of pain, depression and fatigue, and other symptoms.5 Thus, how to alleviate patients' symptoms is still a core issue in palliative care.
Based on our previous reports and those of other groups,6, 7, 8 the symptoms of advanced cancer patients are multiple, concurrent, and tend to be moderate or severe in intensity. These studies used to focus on individual symptoms in various patients with malignancy.9 Individual, symptom-oriented therapeutic strategies raise some important concerns, such as polypharmacy and drug side effects,10 and the tendency to emphasize management of a single symptom at the expense of others may compromise quality of life.9, 11
In recent years, a few studies of symptom management have shifted from individual symptoms to symptom clusters.12, 13 Symptom clusters have been defined as three or more concurrent symptoms that are related to each other13 or two or more related symptoms that occur together.14 Using cluster analysis, a retrospective U.S. study of 1000 advanced cancer patients described seven symptom clusters: fatigue (anorexia-cachexia), neuropsychological, upper gastrointestinal, nausea and vomiting, aerodigestive, debility, and pain.9 This study highlighted the importance of symptom clustering in advanced cancer.9 Another study on 151 cancer patients conducted in Taiwan highlighted three symptom clusters by factor analysis, including sickness, gastrointestinal symptoms, and emotional symptoms.15 The investigators also demonstrated that disease status, chemotherapy, and psychological distress were related to symptom clusters. Two longitudinal studies of cancer patients with bone or brain metastasis reinforced the concept of different patterns of symptom clustering at various time points, which may be associated with radiotherapy.16, 17 A more recent study demonstrated the relationship of symptom cluster with symptom interference with daily life in Taiwanese lung cancer patients.18
Rapid progress has been made in the study of symptom clusters in the past few years. However, little remains known about the underlying mechanisms that lead to aggregation of symptoms, associations of physical signs and psychological distress with symptom clustering, and the relationship of symptom clustering and prognosis in cancer patients. For better symptom control, further study of symptom clustering in advanced cancer patients is warranted.
Methods
Patients and Palliative Care Setting
Participants in this prospective study were selected from consecutive patients with advanced cancer who consented and were admitted to the Palliative Care Unit of the National Taiwan University Hospital during the period from September 2002 through December 2003. The inclusion criteria included age older than 20 years, and level of consciousness had to be clear enough (alert or lethargic consciousness) to report symptoms in Mandarin or Taiwanese on admission. The participants were receiving care provided by a multidisciplinary team consisting of physicians, nurses, psychologists, social workers, clinical Buddhist chaplains, and volunteers. The study was approved by the ethics committee at the National Taiwan University Hospital.
Instruments
The assessment tool was a “Symptom Reporting Form” used in our previous study,19 designed by experienced specialists, which assessed physical and psychosocial distress by different scale systems. Physical symptoms included fatigue, weakness, and pain, and were graded on a scale of 0–10 (0
=none and 10=extreme). Anorexia, nausea/vomiting, taste alteration, dysphagia, restlessness/heat, abdominal fullness, constipation, diarrhea, dry mouth, dizziness, dyspnea, insomnia, and night sweating were graded on a scale of 0–3 (0=none, 1=mild, 2=moderate, and 3=severe). Based on the physical examination, functional performance was assessed by the Eastern Cooperative Oncology Group (ECOG) Scale. Psychological assessments included depression, anxiety, and aggression, and were graded on a scale of 1–5 (1=almost none, 2=mild, 3=moderate, 4=severe; 5=extreme).
Symptom Assessment and Data Collection
According to patient report, symptoms and their severities were recorded on admission and checked by the same staff members. Information about psychological distress, including depression, anxiety, and aggression, were collected by medical staff and clinical psychologists. A consensus of psychological distress rating was obtained after thorough discussion in team meetings. Team meetings were held once per week, and data for this study consisted of routine records, including patients' demographic data (age, gender, primary site of cancer, and survival days) and Symptom Reporting Forms on admission.
Statistical Analysis
All data were analyzed by using SPSS 11.0 (SPSS Inc., Chicago, IL) statistical software. To compare with other symptoms, the scale scores of fatigue, weakness, and pain were linearly transformed from a scale of 0–10 to 0–3. Descriptive statistical data were summarized as frequencies and percentages for categorical variables, medians and ranges for number of symptoms, and means and standard deviations for other continuous variables. To understand the latent constructs of the 15 symptoms on admission, we conducted an exploratory factor analysis using the principal component method with promax rotation. Based on the factor scores from the exploratory factor analysis, we first performed hierarchical cluster analysis to explore the suitable number of clusters. Ward's method was used, because it maximized the within-group homogeneity. The number of clusters was determined by considering the root mean square standard deviation, semipartial R-squared, the case number of each cluster, and clinical phenomena. After the number of clusters was determined, the K-means nonhierarchical clustering method was used to cluster the patients. Finally, we used analysis of variance (ANOVA) to test the differences among patient groups in subjective symptoms and associated clinical parameters, and Scheffé's test was used for post hoc examinations. A probability of less than 0.05 (P<0.05) was considered statistically significant for ANOVA, and probability of less than 0.01 (P<0.01) was considered statistically significant in post hoc tests.
Results
The subjects of this study included 427 hospitalized patients with advanced cancer; 222 of them were males. The median age was 66 years (range: 27–93). With regard to the primary cancer origins, lung cancer was the largest group (19.9%), followed by liver (18.0%) and colorectal (11.0%) cancers. The median survival was 13 days (ranging from 1 to 418 days). More than three-fourths of these patients had an ECOG performance status of 3 or 4 (Table 1).
Table 1. Demographic and Diagnostic Data on 427 Patients on Admission
| Variable | Statisticsa |
|---|---|
| Sex | |
| Male | 222 (52.0) |
| Female | 205 (48.0) |
| Age (years) | 66 (27–93) |
| Survival (days) | 13 (1–418) |
| Functional performance (ECOG) | |
| 0 | 2 (0.5) |
| 1 | 35 (8.2) |
| 2 | 61 (14.4) |
| 3 | 145 (34.3) |
| 4 | 182 (42.6) |
| Primary cancer sites | |
| Lung | 85 (19.9) |
| Liver | 77 (18.0) |
| Colon/rectum | 47 (11.0) |
| Stomach | 35 (8.2) |
| Head and neck | 29 (6.8) |
| Breast | 23 (5.4) |
| Pancreas | 17 (4.0) |
| Cervix/uterus | 16 (3.7) |
| Others | 98 (23.0) |
an (%) for categorical variables and median (range) for continuous variables. |
Fatigue (94.2%) was the most common symptom, followed by weakness (93.9%), anorexia (87.1%), pain (83.8%), and constipation (64.2%). With regard to the severity of symptoms, anorexia (1.75±0.89) was the most serious symptom, followed by fatigue (1.73±0.70), weakness (1.72±0.69), pain (1.35±0.81), constipation (1.19±1.06), and abdominal fullness (1.13±1.13) (Table 2).
Table 2. Prevalence and Severity of Symptoms on Admission
| Variable of Symptom (Range of Severity) | With Symptom | Mean±SD | |
|---|---|---|---|
| n | % | ||
| Fatigue (0–3) | 402 | 94.15 | 1.73±0.70 |
| Weakness (0–3) | 401 | 93.91 | 1.72±0.69 |
| Anorexia (0–3) | 372 | 87.12 | 1.75±0.89 |
| Pain (0–3) | 358 | 83.84 | 1.35±0.81 |
| Constipation (0–3) | 274 | 64.17 | 1.19±1.06 |
| Abdominal fullness (0–3) | 243 | 56.91 | 1.13±1.13 |
| Insomnia (0–3) | 226 | 52.93 | 0.83±0.87 |
| Dyspnea (0–3) | 216 | 50.59 | 0.90±1.03 |
| Dry mouth/thirsty (0–3) | 207 | 48.48 | 0.70±0.81 |
| Dysphagia (0–3) | 201 | 47.07 | 0.81±0.98 |
| Dizziness (0–3) | 148 | 34.66 | 0.48±0.71 |
| Nausea/vomiting (0–3) | 187 | 43.79 | 0.75±0.97 |
| Taste alteration (0–3) | 135 | 31.62 | 0.47±0.75 |
| Restless/heat (0–3) | 90 | 21.08 | 0.29±0.61 |
| Night sweats (0–3) | 79 | 18.50 | 0.26±0.58 |
As mentioned previously, to understand the latent constructs of the 15 symptoms on admission, we conducted an exploratory factor analysis using the principal component method with promax rotation. In the process of analysis, pain complex (PC) on various sites was separated as an independent symptom. We further conducted factor analyses for the other 14 symptoms and obtained four clusters, named as “loss of energy” (LE), “poor intake” (PI), “autonomic dysfunction” (AD), and “aerodigestive impairment” (AI). LE included fatigue and weakness. PI included anorexia, taste alteration, dysphagia, constipation, and dry mouth/thirst. AD included restlessness/heat, dizziness, insomnia, and night sweats. AI included nausea/vomiting, abdominal fullness, and dyspnea (Table 3).
Table 3. Factor Analysis of Symptoms on Admission
| Factor Loading | |||||
|---|---|---|---|---|---|
| Cluster | Symptom | Factor 1 | Factor 2 | Factor 3 | Factor 4 |
| Loss of energy | Fatigue | 0.949 | |||
| Weakness | 0.964 | ||||
| Poor intake | Anorexia | 0.377 | |||
| Taste alteration | 0.655 | ||||
| Dysphagia | 0.799 | ||||
| Constipation | 0.394 | ||||
| Dry mouth/thirsty | 0.533 | ||||
| Autonomic dysfunction | Restlessness/heat | 0.790 | |||
| Dizziness | 0.371 | ||||
| Insomnia | 0.460 | ||||
| Night sweats | 0.832 | ||||
| Aerodigestive impairment | Nausea/vomiting | 0.553 | |||
| Abdominal fullness | 0.734 | ||||
| Dyspnea | 0.539 | ||||
Based on the factor scores of five symptom factors, we further conducted cluster analysis on 394 patients with complete data selected from 427 participants. We decided to divide the 394 patients into six groups after considering the results of the hierarchical cluster analysis and clinical phenomena. We then used the K-means nonhierarchical clustering method to divide the 394 patients into six groups. The first group included 93 patients, the second 49, the third 75, the fourth 52, the fifth 67, and the sixth 57. We used ANOVA to illustrate that the five factor scores of the six patient groups were different statistically (P<0.05) (Table 4). The pattern of symptom clustering in Group 1 was characterized by LE and PC, Group 2 was characterized by all symptom clusters except PC, Group 3 was characterized by all relatively mild symptom clusters, Group 4 was characterized by PC, Group 5 was characterized by LE, and Group 6 was characterized by all symptom clusters. According to the severity of symptom clusters presented in six groups of patients, LE was the most severe one, followed by PC and then AD. In general, the patients in Groups 2 and 6 had more symptom clusters with higher severity. The severity of symptom clusters in Groups 4 and 5 was relatively lower.
Table 4. Cluster Analysis Based on Factor Scores on Admission
| Cluster | Group 1 (n=93) | Group 2 (n=49) | Group 3 (n=75) | Group 4 (n=52) | Group 5 (n=67) | Group 6 (n=57) | F | Post Hoc |
|---|---|---|---|---|---|---|---|---|
| Loss of energy | 2.00±0.34 | 2.15±0.50 | 1.06±0.43 | 0.88±0.44 | 2.13±0.36 | 2.11±0.45 | 124.91a | 1,2,5,6>3,4 |
| Poor intake | 0.85±0.38 | 1.23±0.43 | 0.70±0.41 | 0.68±0.54 | 0.86±0.46 | 1.73±0.49 | 48.83a | 6>2>1,3,4,5 |
| Autonomic dysfunction | 0.35±0.32 | 0.79±0.55 | 0.34±0.35 | 0.28±0.43 | 0.25±0.25 | 0.90±0.50 | 29.45a | 2,6>1,3,4,5 |
| Aerodigestive impairment | 0.74±0.41 | 1.72±0.38 | 0.94±0.47 | 0.32±0.36 | 0.60±0.40 | 1.44±0.62 | 74.73a | 2,6>1,3,5>4 3>5 |
| Pain complex | 1.98±0.39 | 0.91±0.51 | 0.63±0.47 | 1.69±0.44 | 0.57±0.48 | 2.21±0.37 | 176.73a | 1,6,>4>2>3,5 |
aP |
To examine the biological and psychological parameters in the six patient groups, we conducted ANOVA and summarized the significant parameters (Table 5). Patients in Group 4 had the longest survival and best functional performance. Patients in Group 6 had the shortest survival and worst functional performance. There were about 40% of patients with bone metastasis in Groups 1, 4, and 6. Both ascites and pleural effusion were key physical parameters. Patients in Groups 2 and 6 had more severe pleural effusion and ascites. Regarding psychological assessment, patients in Groups 2 and 6 had more severe anxiety, and those in Groups 1 and 6 had more severe depression.
Table 5. Clinical Parameters Associated With Symptom Clustering on Admission
| Parameter | Group 1 | Group 2 | Group 3 | Group 4 | Group 5 | Group 6 | F/χ2 | Post Hoc |
|---|---|---|---|---|---|---|---|---|
| Survival (days) | 25.48±37.36 | 21.66±32.54 | 21.05±24.02 | 40.35±39.54 | 29.63±37.43 | 19.02±27.73 | 2.69a | 4>3,6 |
| Functional performance (ECOG: 0–4) | 3.22±0.88 | 3.16±0.92 | 2.78±1.00 | 2.42±1.16 | 3.31±0.82 | 3.38±0.80 | 9.10b | 1,2,5,6>4 |
| With bone metastasis, n (%) | 37 (39.8) | 14 (28.6) | 21 (28.0) | 22 (40.4) | 12 (17.9) | 23 (40.4) | 12.85a | 5,6>3 |
| Physical examination | ||||||||
| Pleural effusion | 0.28±0.63 | 0.65±0.81 | 0.39±0.73 | 0.12±0.43 | 0.25±0.53 | 0.47±0.71 | 4.45c | 2,6>4 2 >5 |
| Ascites | 0.43±0.77 | 1.20±1.15 | 0.49±0.86 | 0.31±0.76 | 0.30±0.60 | 0.65±0.94 | 8.50b | 2>1,3,4,5 |
| Psychological distress | ||||||||
| Anxiety | 2.35±1.02 | 2.60±1.01 | 2.27±1.00 | 2.00±0.87 | 2.07±0.88 | 2.50±0.76 | 3.08c | 2,6>4 |
| Depression | 2.43±1.05 | 2.43±0.94 | 2.12±1.05 | 1.94±0.93 | 1.95±0.78 | 2.50±0.89 | 4.03c | 1,6>4,5 |
aP |
bP |
cP |
Discussion
To investigate symptom clustering in patients with advanced cancer, we identified five major symptom clusters: LE, PI, AD, AI, and PC. We then classified patients into six groups by cluster analysis and evaluated symptom clusters in terms of finding what may relate to underlying pathophysiological mechanisms. We demonstrated that survival, functional performance, bone metastasis, pleural effusion, and ascites were associated with the symptom clustering pattern in advanced cancer patients.
Cluster 1 (LE) includes fatigue and weakness. The concurrence of these two symptoms is consistent with the results of a previous study;9 they represent similar meanings in Taiwanese cancer patients.19 The cluster 2 (poor intake) includes symptoms of the anorexia-cachexia syndrome, which has been thought to be mediated by proinflammatory cytokines, such as interleukin-1, interleukin-6 and tumor necrosis factor-alpha (TNF-α).20 Cluster 3 (AD) includes the symptoms that may result from AD, which has been documented in cancer.21 The paraneoplastic processes may be one of the causes of AD.22 Cluster 4 (AI) includes nausea, vomiting, abdominal fullness, and dyspnea. The apparent tendency of some gastrointestinal symptoms to occur together was also noted in a previous study.23 Dyspnea occurred together with nausea or vomiting, suggesting that ascites may be an important contributing factor.24 Cluster 5 is a PC affecting various sites, which is a special symptom in advanced cancer patients. Our previous studies showed that pain in advanced cancer patients possesses some characteristics, including being a controllable symptom, and is frequently linked with depression.6, 19 Thus, pain was separated as an independent symptom in the process of factor analysis.
Based on the results of cluster analysis, LE and PC seem to be the most common symptom clusters in advanced cancer patients, which is consistent with the results of another study.8 In patient Groups 2 and 6 with poorer prognosis, there is an association between PI and AI. Their concurrence may aggravate the expression of AD25 and appear to relate to PC and LE.26 When all the clusters are present, as in patient Group 6, PI, AD, and PC are the most prominent ones.
Based on our data, survival, functional performance, and bone metastasis are associated with symptom clustering. Symptom burden is a concept that encompasses both the severity of the symptoms and the patient's perception of the impact of the symptoms.27 The patients in Groups 2 and 6 had shorter survival, which reinforces that symptom burden is associated with survival.28 In addition to symptom burden, LE, including fatigue and weakness, is associated with survival,29 which supports the patients in Group 3 having shorter survival. Symptom burden and LE are associated with impaired functional performance,30, 31 which supports the finding that patients in Groups 1, 2, 5, and 6 have worse functional performance. This highlights the need to manage symptoms associated with LE. Bone metastasis is associated with PC,32 which supports the association of Groups 1, 4, and 6 with pain. The patients in Group 4 had the longest survival and best functional performance, which shows that pain can exist independently and be controlled.6, 19, 33 On physical examination, pleural effusion and ascites were found to be more severe in Groups 2 and 6, because they are associated with symptom burden, functional performance, and survival in advanced cancer patients.34, 35 Thus, fluid accumulation, including ascites and pleural effusion, is also associated with symptom clustering. Primary cancer origin, age, and gender are not significant parameters in symptom clustering, because multiple organ failure is the common pathway in these patients with far advanced cancer.6, 36
On psychological assessment, the levels of anxiety and depression in Group 6 were found to be higher than those of other groups, suggesting that subjective symptoms are associated with psychological distress.37 Although LE and depression are associated,38 the depression levels of patients in Group 5 with prominent LE were lower than those of patients in Groups 1 and 6, which means that the associated factors of depression are multiple,39 and more active pain control is indicated for the patients with depression.40 Subjective symptoms are correlated with organ failure, indicating that symptoms of advanced cancer patients result mainly from physical deterioration, and psychological distress is secondary.41 Therefore, aggressive symptom management is very important for advanced cancer patients.
A limitation of the present study is that the checklist of symptom assessment is a relatively short instrument, which assesses only 15 symptoms. It should be considered to use a more comprehensive tool to assess symptoms when exploring symptom clustering. In addition, the study design is cross-sectional, whereas patients' symptoms are dynamic over time along the disease trajectory. Studies on longitudinal follow-up should be designed to explore symptom clustering at different time points.
In conclusion, symptom clustering exists in advanced cancer patients. Survival, functional performance, bone metastasis, and fluid accumulation are associated with particular types of symptom clustering. Psychological distress is secondary to physical deterioration. To come to grips with the complex manifestations of symptoms in advanced cancer, symptom clusters should be considered in developing clinical guidelines for palliative care.
Acknowledgments
The authors are indebted to the faculty of the Department of Family Medicine, National Taiwan University Hospital, for its full support of this study.
References
- . Symptom burden in the last week of life. J Pain Symptom Manage. 2004;27:5–13
- . New directions in supportive care. Support Care Cancer. 2005;13:135–137
- . The symptom monitor. A diary for monitoring physical symptoms for cancer patients in palliative care: feasibility, reliability and compliance. J Pain Symptom Manage. 2004;27:24–35
- Advancing the science of symptom management. J Adv Nurs. 2001;33:668–676
- National Institutes of Health State-of-the-Science Conference Statement: symptom management in cancer: pain, depression, and fatigue, July 15-17, 2002. J Natl Cancer Inst Monogr. 2004;32:9–16
- . Prevalence and severity of symptoms in terminal cancer patients: a study in Taiwan. Support Care Cancer. 2000;8:311–313
- . Common symptoms in patients with advanced cancer. J Palliat Care. 1991;7:25–29
- . The symptoms of advanced cancer: relationship to age, gender, and performance status in 1,000 patients. Support Care Cancer. 2000;8:175–179
- . Symptom clustering in advanced cancer. Support Care Cancer. 2006;14:831–836
- . Polypharmacy in palliative care: can it be reduced?. Singapore Med J. 2002;43:279–283
- . Symptom clusters: the new frontier in symptom management research. J Natl Cancer Inst Monogr. 2004;32:17–21
- . Occurrence of symptom clusters. J Natl Cancer Inst Monogr. 2000;32:76–78
- . Symptom clusters and their effect on the functional status of patients with cancer. Oncol Nurs Forum. 2001;28:465–470
- . Symptom clusters: concept analysis and clinical implications for cancer nursing. Cancer Nurs. 2005;28(4):270–282
- . Symptom clusters in cancer patients. Support Care Cancer. 2006;14:825–830
- . Symptom clusters in cancer patients with bone metastases. Support Care Cancer. 2007;15:1035–1043
- Symptom clusters in cancer patients with brain metastases. Clin Oncol. 2008;20:76–82
- . Symptom clusters and relationships to symptom interference with daily life in Taiwanese lung cancer patients. J Pain Symptom Manage. 2008;35:258–266
- . Symptom patterns of advanced cancer patients in a palliative care unit. Palliat Med. 2006;20:617–622
- Impact of TNF-alpha and IL-6 levels on development of cachexia in newly diagnosed NSCLC patients. Am J Clin Oncol. 2006;29:328–335
- . Autonomic nervous system dysfunction in advanced cancer. Support Care Cancer. 2002;10:523–528
- . Symptom clusters in advanced illness. Semin Oncol Nurs. 2005;21:20–28
- . Gastrointestinal symptoms among inpatients with advanced cancer. Am J Hosp Palliat Care. 2002;19:351–355
- . Temporary drainage of symptomatic malignant ascites by a catheter inserted under computerized tomography. J Pain Symptom Manage. 1998;15:374–378
- . Chronic nausea and anorexia in advanced cancer patients: a possible role for autonomic dysfunction. J Pain Symptom Manage. 1987;2:19–21
- . Cytokines and advanced cancer. J Pain Symptom Manage. 2000;20:214–232
- . Symptom burden: multiple symptoms and their impact as patient-reported outcomes. J Natl Cancer Inst Monogr. 2007;37:16–21
- Prediction of survival for advanced cancer patients by recursive partitioning analysis: role of Karnofsky performance status, quality of life, and symptom distress. Cancer Invest. 2004;22:678–687
- . Lung carcinoma symptoms—an independent predictor of survival and an important mediator of African-American disparity in survival. Cancer. 2004;101:1655–1663
- . Relationship between symptom change, objective tumor measurements, and performance status during chemotherapy for advanced lung cancer. Clin Lung Cancer. 2008;9:51–58
- . Giving meaning to measure: linking self-reported fatigue and function to performance of everyday activities. J Pain Symptom Manage. 2006;31:229–241
- Prevalence and management of pain in Italian patients with advanced non-small-cell lung cancer. Br J Cancer. 2004;90:2288–2296
- . High dose morphine use in the hospice setting. A database survey of patient characteristics and effect on life expectancy. Cancer. 1999;86:871–877
- Dyspnea and its correlates in Taiwanese patients with terminal cancer. J Pain Symptom Manage. 2004;28:123–132
- . Prediction of survival in terminal cancer patients in Taiwan: constructing a prognostic scale. J Pain Symptom Manage. 2004;28:115–122
- . A prospective study on the dying process in terminally ill cancer patients. Am J Hosp Palliat Care. 1998;15:217–222
- . Psychological distress in patients with advanced cancer. Clin Med. 2006;6:148–150
- . Assessment of symptom clusters in people with cancer. J Natl Cancer Inst Monogr. 2004;32:98–102
- Depression, hopelessness, and desire for hastened death in terminally ill patients with cancer. JAMA. 2000;284:2907–2911
- Psychological distress and pain significantly increase before death in metastatic breast cancer patients. Psychosom Med. 2003;65:416–426
- Quality of life and survival prediction in terminal cancer patients: a multicenter study. Cancer. 2004;101:1090–1098
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.
PII: S0885-3924(10)00078-3
doi:10.1016/j.jpainsymman.2009.09.005
© 2010 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved.
Volume 39, Issue 4 , Pages 655-662, April 2010
