Effects of pain, fatigue, insomnia, and mood disturbance on functional status and quality of life of elderly patients with cancer

Transcription

1 Critical Reviews in Oncology/Hematology 78 (2011) Effects of pain, fatigue, insomnia, and mood disturbance on functional status and quality of life of elderly patients with cancer Karis K.F. Cheng, Diana T.F. Lee The Nethersole School of Nursing, Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong Accepted 10 March 2010 Contents 1. Introduction Patients and methods Instruments Statistical analysis Results Discussion Reviewers Conflicts of interest Acknowledgement References Biographies Abstract Background: Most elderly patients with cancer suffer from a multitude of intense physical and psychological symptoms regardless of the stage of disease. The current paper describes the prevalence of pain, fatigue, insomnia, and mood disturbance, alone and in combination in elderly cancer patients, as well as the inter-correlations among these four symptoms, and the relationship of the symptom cluster to functional status and quality of life (QoL) during cancer therapy. Patients and methods: This cross-sectional study used secondary data from a convenience sample of 120 patients, 65 years of age and older, with colorectal, lung, head/neck, breast, gynecological, prostate or esophageal cancer receiving chemotherapy or radiotherapy. Measuring instruments included the Karnofsky Performance Scale (KPS), the respective items from the Chinese version of the Symptom Distress Scale (SDS-C), and the Functional Assessment of Cancer Therapy-General (FACT-G [C]). The influence of the symptom cluster on patients functional status and QoL was determined by hierarchical multiple regression. Results: Twenty percent and 29.2% of patients reported co-occurrence of any two and any three symptoms of pain, fatigue, insomnia, and mood disturbance, respectively. About one-third of patients (31.2%) reported co-occurrence of all of the four symptoms. The inter-correlations among pain, fatigue, insomnia, and mood disturbance were mild to moderate (r = , p < 0.01). In terms of functional status, the KPS showed a moderate negative correlation with the four symptoms (r = 0.29 to 0.55, p < 0.01). Correlations between the FACT-G (C) subscale/total scores and symptom cluster showed moderate negative correlations (r = 0.23 to 0.55, p < 0.01). About % of variance in functional status and QoL is explained by the symptom cluster of pain, fatigue, insomnia, and mood disturbance in elderly cancer patients receiving cancer therapy after adjustment for gender, age, co-morbidity, stage of disease, and treatment modality. Corresponding author at: Room 732, Esther Lee Building, The Nethersole School of Nursing, Faculty of Medicine, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong. Tel.: ; fax: address: kariskwong@cuhk.edu.hk (K.K.F. Cheng) /$ see front matter 2010 Elsevier Ireland Ltd. All rights reserved. doi: /j.critrevonc

2 128 K.K.F. Cheng, D.T.F. Lee / Critical Reviews in Oncology/Hematology 78 (2011) Conclusions: Our results suggest that pain, fatigue, insomnia, and mood disturbance are highly prevalent in elderly patients who undergone cancer therapy. These four symptoms may occur in a cluster and may negatively influence elderly patients functional status and QoL during cancer therapy Elsevier Ireland Ltd. All rights reserved. Keywords: Symptom cluster; Elderly cancer patients; Pain; Fatigue; Insomnia; Mood disturbance 1. Introduction The incidence of cancer increases progressively throughout the age span and is a major burden of disease for people aged 65 years or older. Coupled with the increased incidence of cancer among the older people is the recognition of unprecedented growth of the elder demographic worldwide [1]. However, most elderly patients with cancer present with advanced disease and suffer from a multitude of intense physical and psychological symptoms. Some symptoms are related to the progression of the disease, while others are associated with the early and late medical effects of cancer treatment, as well as psychosocial adjustment [2,3]. The consequences that symptoms exert have far reaching and negative effects on health care costs and other social resources, as well as many aspects of a patient s life, including functional status, roles, quality of life (QoL), and length of survival [4,5]. Complicating the symptom experience for the elderly patients is the fact that many are entering the cancer trajectory with pre-existing co-morbidity [6] as well as susceptibility to the progressive accumulation of multiple chronic diseases and a decline in functional ability [7]. Thus, elderly patients make up an important group of patients in the cancer setting that need special attention. However, our understanding of this disproportionate symptom experience for elderly cancer patients is very limited at present. Pain, fatigue, insomnia, and mood disturbance are the four most common and distressing symptoms reported by patients with cancer throughout the disease and treatment trajectories [8,9]. Nevertheless, symptoms seldom occur in isolation in patients with cancer [10]. Contemporary research evidence reveals compelling support for the notion that symptoms such as pain, fatigue, insomnia, mood disturbance, and others occur in clusters and can negatively influence patient outcomes [8,11 13]. The concept of symptom clusters is becoming increasingly recognized as an important platform for symptom management for cancer patients. Several studies have demonstrated small to moderate inter-correlations among pain, fatigue, and insomnia [11,14,15]. Depression which refers to the symptom of low mood is often part of a cluster of interrelated symptoms, including fatigue and insomnia [13,16,17]. Bower et al. surveyed fatigue in breast cancer survivors and found that fatigue was associated with depression, pain, and insomnia [18]. Glover et al. studied pain in adults with cancer and revealed that it was associated with depression, fatigue, and anxiety [19]. Gastron-Johansson et al. demonstrated that pain, fatigue, and depression were significantly correlated with total health status in patients with breast cancer [20]. Redeker et al. indicated that fatigue, insomnia, anxiety, and depression together explained 47% of the variance in QoL in patients undergoing initial chemotherapy for cancer [13]. In Dodd et al. s study, the cluster of pain, fatigue, and insomnia was associated with a lower functional status in patients receiving three cycles of chemotherapy [11]. In addition, a number of researchers have examined the factors contributing to patients symptom experiences. While numerous factors including age, gender, co-morbidity, and treatment modality have been shown to influence the symptom severity and distress as well as patients level of functioning, the findings are conflicting [4,9,21 23]. To date, there have been few attempts to fully characterize the interrelationships of pain, fatigue, insomnia, and mood disturbance in a cluster and to examine their synergistic effects on functional status and QoL. Furthermore, the majority of studies on symptom clusters deal with young adults. Little is known about symptom clustering in elderly cancer patients, and even less is known about how pain, fatigue, insomnia, and mood disturbance in a cluster are related to each other, and how the symptom cluster affects elderly patients functional status as well as their QoL. As pointed out by Cohen, for some older patients, maximizing functional status and QoL is the goal of cancer treatment [7]. It is therefore important for gerontology/oncology health care professionals to understand QoL in the context of cancer occurring in the setting of an increased prevalence of multiple symptoms. The purpose of this secondary data analysis was to determine the prevalence of pain, fatigue, insomnia, and mood disturbance, alone and in combination, in elderly cancer patients, to examine the relationships among these four symptoms, and to examine the relationship of the symptom cluster to functional status and QoL after adjusting for age, gender, number of pre-existing co-morbid conditions, stage of disease, and treatment modality. 2. Patients and methods This cross-sectional study used secondary data from a convenience sample of 120 patients 65 years of age and older with colorectal, lung, head/neck, breast, gynecological, prostate or esophageal cancer receiving chemotherapy or radiotherapy at an oncology unit of a regional hospital in Hong Kong. The study sample was drawn from a previously conducted observational validation study [24]. The original database consisted of 370 subjects who undergone cancer therapy or were at the early post-treatment stage for

3 K.K.F. Cheng, D.T.F. Lee / Critical Reviews in Oncology/Hematology 78 (2011) Table 1 Characteristics of subjects (N = 120). n (%) Age (years) (mean ± SD) ± (84.2%) (15.8%) Male 70 (58.3) Female 50 (41.7) Educational level (n = 116) No formal education 25 (21.6) Primary 53 (45.7) Secondary 30 (25.9) Tertiary 8 (6.9) No. of co-morbidity (n = 110) 0 58 (52.7) 1 30 (25.4) 2 14 (11.9) 3 5 (4.2) 4 3 (2.5) Cancer diagnosis Head and neck 19 (15.8) Breast 13 (10.8) Lung 24 (20) Colorectal 38 (31.7) Gynaecological 7 (5.8) Prostate 7 (5.8) Esophagus 4 (3.3) Others 8 (6.7) (n = 109) Early 72 (66.1) Late 37 (33.9) Previous use of surgery (n = 120) No 56 (46.7) Yes 64 (53.3) Chemotherapy 49 (40.8) Radiotherapy 39 (32.5) Chemoradiotherapy 32 (26.7) any diagnosis of cancer. Inclusion criteria for the original sample were that subjects (a) were 18 years of age or older, (b) had been diagnosed with solid tumors, (c) were undergoing chemotherapy or radiotherapy or were in the first 12 months of the post-treatment stage, (d) without medical diagnosis of encephalopathy or psychiatric disease, and (e) were able to understand the study and give informed consent. Patients were included in this secondary analysis if they were 65 years of age or older and were being treated with cancer therapy. The study was conducted in accordance with the Declaration of Helsinki; all the subjects provided written informed consent before enrolling in the study Instruments The observer rated Karnofsky Performance Scale (KPS) was used to measure functional status. This is an 11-point rating scale ranging from 0 to 100 (0, dead; 100, normal function) widely used to assess patients physical functional level related to cancer and its treatment [25]. A validation study strongly suggests that the score reflects physical functioning of the patient [26]. Pain, fatigue, insomnia, and mood disturbance were measured using the respective items from the Chinese version of the Symptom Distress Scale (SDS-C). The SDS-C provides a valid measure of global symptom distress in Chinese cancer patients. This scale evaluates 10 symptoms that are scored from 1 (no problem with a particular symptom at the present time) to 5 (worst possible problem). The possible score on pain, fatigue, insomnia, and mood disturbance ranges from 1 to 5, with 5 indicating a high level of distress [27]. For this study, a symptom was considered present if the SDS-C score was 2. Patients QoL was assessed using the Chinese version of the Functional Assessment of Cancer Therapy-General (FACT-G [C]). This is a 29-item general measure of HRQoL for Chinese cancer patients. Each item is scored from 0 to 4, anchored from not at all to very much. The physical wellbeing (Phy), social/family well-being (Soc/Fam), emotional well-being (Emt), and functional well-being (Fnt) subscale and total scores were computed as previously described. A lower score indicates a poorer QoL [28]. Both the SDS-C and FACT-G (C) are attributed with valid and reliable psychometrics and acceptable cultural equivalence with the original version [29,30]. In the original study, the data were collected by face-to-face interview in a private room of the clinic. Face-to-face interview was used because of the high level of illiteracy among older Hong Kong Chinese people. Two part-time research assistants were employed; they were briefed on the questionnaire and interview skills in order to achieve consistency in data collection Statistical analysis Pearson s simple correlation test was performed for the correlations among pain, fatigue, insomnia, and mood disturbance, and to determine the relationships between symptoms, the KPS scores, and the FACT-G (C) subscale/total scores. One-way analyses of variance were used to determine if there were significant differences among patients with different number of combinations of symptoms on KPS scores, and the FACT-G (C) subscale and total scores. The influence of the symptom cluster on patients functional status and QoL was determined by a two-stage hierarchical multiple regression. In the first step, gender (male 0, female 1), age, number of pre-existing co-morbid conditions, stage of disease (early [0-II] 0, late [III-IV] 1), and treatment modality (chemotherapy 0, radiotherapy 1, chemoradiotherapy 1) were entered into the regression model as covariates because of their potential confounding effect with the four selected symptoms based on prior literature [4,9,21 23]. In this study, medical co-morbidity information including hypertension, diabetes, heart and lung problems, arthritis, as well as vision and hearing problems was gathered prospectively from the

4 130 K.K.F. Cheng, D.T.F. Lee / Critical Reviews in Oncology/Hematology 78 (2011) Table 2 The mean KPS and FACT-G(C) subscale/total scores by the number of combined pain (P), fatigue (F), insomnia (I), and mood disturbance (M) symptoms (N = 120). F p-value All of the four symptoms of PFIM (n = 38) Any three symptoms of PFIM (n = 35) Any two symptoms of PFIM (n = 24) No PFIM (n = 9) Any one symptom of PFIM (n = 14) Mean ± SD (95% CI) KPS a 94.7 ± 7.0 ( ) 93.6 ± 11.5 ( ) 92.1 ± 6.6 ( ) 87.1 ± 10.5 ( ) 76.8 ± 12.2 ( ) <0.001 FACT-G(C) scores Physical b 26.9 ± 2.2 ( ) 24.1 ± 3.3 ( ) 23.4 ± 3.4 ( ) 21.4 ± 4.4 ( ) 17.5 ± 5.3 ( ) <0.001 Social b 23.4 ± 2.9 ( ) 20.3 ± 3.5 ( ) 20.2 ± 4.3 ( ) 18.7 ± 4.2 ( ) 17.8 ± 3.7 ( ) 5.15 =0.001 Emotion c 21.9 ± 2.7 ( ) 19.2 ± 2.9 ( ) 18.9 ± 2.8 ( ) 17.0 ± 4.3 ( ) 16.8 ± 4.4 ( ) 6.97 <0.001 Functional b 22.3 ± 4.9 ( ) 18.1 ± 6.5 ( ) 16.1 ± 5.4 ( ) 12.6 ± 5.9 ( ) 9.1 ± 5.4 ( ) <0.001 Total d ± 6.5 ( ) 88.0 ± 12.3 ( ) 86.2 ± 12.3 ( ) 75.4 ± 13.2 ( ) 66.3 ± 14.9 ( ) <0.001 Post hoc comparisons of KPS and FACT-G(C) physical subscale scores: no, any one, any two, any three PFIM > all PFIM. Post hoc comparisons of FACT-G(C) social subscale score: no PFIM > any three/all PFIM. Post hoc comparisons of FACT-G(C) emotional and functional subscale and total scores: no, any one, any two PFIM > any three/all PFIM. a The mean KPS score range 0 100; higher score represents poor functional status. b The mean FACT-G(C) physical, social, and functional subscale scores range 0 28; higher score represents a better quality of life. c The mean FACT-G(C) emotion subscale score range 0 24; higher score represents a better quality of life. d The mean FACT-G(C) total score range 0 116; higher score represents a better quality of life. medical record by the research assistants. To analyze whether a symptom cluster existed and contributed to the patient s functional status and QoL over and above the influences of gender, age, co-morbidity, stage of disease, and treatment modality, the four symptoms were entered into the hierarchical analysis as a second step. Statistical significance was present at p < Multicollinearity was examined with tolerance and variance inflation factor (VIF) statistics and found to be acceptable in all cases. Highest VIF values were 1.61, and lowest tolerance values were 0.88, which fall within most recommended acceptable guidelines for multiple regression [31]. 3. Results As shown in Table 1, the mean age of the patients was ± 5.6 years (range years); 58.3% (n = 70) were males, and 32.8% (n = 38) had completed secondary or tertiary education. Less than half of the patients had at least one co-morbid condition (47.3%, n = 52). About 32% were diagnosed with colorectal cancer (n = 38), and 20% with lung cancer (n = 24). The top three most frequently recorded co-morbid conditions included: (1) hypertension (58.2%, n = 64), (2) diabetes (20%, n = 22), and (3) heart problems (6.4%, n = 7). More than half of the patients were in the early stage of disease (66.1%). About half of the patients had undergone surgery to excise their tumors (53.3%, n = 64). The majority of patients were receiving chemotherapy (40.8%, n = 49). It is noteworthy that 86.7% of the patients reported mood disturbance, 62.5% reported fatigue, 59.2% reported insomnia, and 49% reported pain. Twenty percent (n = 24) and 29.2% (n = 35) of patients reported co-occurrence of any two and any three symptoms of pain, fatigue, insomnia, and mood disturbance, respectively. About one-third of patients (31.2%, n = 38) reported co-occurrence of all of the four symptoms. Pain was the most distressing symptom, with a mean score of 3.29 ± 0.9, while mood disturbance was the least distressing symptom, with a mean score of 2.7 ± 0.7. The mean distress scores for fatigue and insomnia were 3.01 ± 0.9 and 2.87 ± 0.9, respectively. The mean KPS score was ± Eighty-four percent had a KPS score >80. The mean FACT-G (C) physical, social, and functional subscale scores were 22.0 ± 4.9, 19.5 ± 4.2, and ± 7.0, respectively (score range 0 28), with lower scores representing a poorer QoL. The emotional subscale score was ± 4.1 (score range 0 24). The mean FACT-G (C) total score was ± 16.8 (score range 0 116). Table 2 reveals the KPS and FACT-G(C) subscale/total scores for patients with a different number of combinations of the pain, fatigue, insomnia, and mood disturbance. Significant differences were found in all KPS and FACT-G(C) subscale/total scores among the five subgroups of patients (p < 0.01). Post hoc comparisons with Bonferroni corrections show that the KPS and FACT- G(C) subscale/total scores of patients with co-occurrence of

5 K.K.F. Cheng, D.T.F. Lee / Critical Reviews in Oncology/Hematology 78 (2011) Table 3 Correlation coefficients among pain, fatigue, insomnia, and mood disturbance, and among KPS and FACT-G (C) subscale/total scores (N = 120). SDS-C KPS FACT-G(C) Pain Fatigue Insomnia Physical Social Emotional Functional Total Pain *** *** ** ** *** *** Fatigue *** *** *** ** *** *** *** Insomnia *** *** *** *** ** *** *** *** Mood disturbance ** *** *** ** *** ** *** *** *** ** p < *** p < all of the four symptoms (KPS: 76.8; FACT-G(C) subscales: , FACT-G(C) total: 66.3) were significantly lower than those without any symptoms (KPS: 94.7; FACT-G(C) subscales: , FACT-G(C) total: 103.2), and those with any one symptom (KPS: 93.6; FACT-G(C) subscales: , FACT-G(C) total: 88) and any two of the four symptoms (KPS: 92.1; FACT-G(C) subscales: , FACT-G(C) total: 86.2) (p < 0.01). As shown in Table 3, the inter-correlations among pain, fatigue, insomnia, and mood disturbance were mild to moderate (r = , p < ). Only those with r > 0.3 are reported here. Pain showed moderate positive correlations with fatigue (r = 0.384, p < 0.001), and insomnia (r = 0.312, p < 0.001). Fatigue had moderate correlations with insomnia (r = 0.411, p < 0.001), and mood disturbance (r = 0.341, p < 0.001). Furthermore, insomnia was moderately correlated with mood disturbance (r = 0.434, p < 0.001). In terms of functional status, the KPS scores showed a mild to moderate negative correlation with the four symptoms (r = to 0.551, p < ). Correlations between the FACT-G (C) subscale/total scores and the symptom cluster showed moderate negative correlations of the physical (r = to 0.547, p < 0.001), emotional (r = to 0.503, p < ), and functional (r = to 0.536, p < 0.001) subscales and total (r = to 0.548, p < 0.001) scores with the four symptoms. Table 4. The hierarchical multiple regression for the KPS scores as a dependent variable revealed that gender, age, co-morbidity, stage of disease, and treatment modality together explained 11.1% of the variance in the KPS scores (p = 0.047) in the first step. Age showed significant independent effects for the KPS scores (β = 0.227, p < 0.05). Regression of the KPS scores against the symptom cluster as the second step revealed that the increase in explained variance of 36.8% was significant (F change = 19.0, p < 0.001). Pain (β = 0.353, p < 0.001) and fatigue (β = 0.32, p < 0.001) showed significant independent effects for the KPS scores, whereas mood disturbance and insomnia were not seen as influencing the KPS scores independently (p > 0.05). As for the QoL, gender, age, co-morbidity, stage of disease, and treatment modality together explained 7.6% of the variance in the FACT-G (C) physical subscale score (p > 0.05) in the first step of the hierarchical multiple regression. Regression of the FACT-G (C) physical subscale score against the symptom cluster as the second step revealed that the increase in explained variance of 52.9% was significant (F change = 33.8, p < 0.001). Fatigue (β = 0.360, p < 0.001), mood disturbance (β = 0.355, p < 0.001), and insomnia (β = 0.169, p < 0.05) showed significant independent effects for the FACT-G (C) physical subscale score. As the β coefficients of fatigue and mood disturbance are similar in value, their effects on patients physical well-being are roughly equal. The hierarchical multiple regression for the FACT-G (C) social subscale score as a dependent variable revealed that gender, age, co-morbidity, stage of disease, and treatment modality together explained 9.6% of the variance in the social subscale score (p > 0.05) in the first step. Regression of the FACT-G (C) social subscale score against the symptom cluster as the second step revealed that the increase in explained variance of 8.7% was significant (F change = 2.6, p = 0.041). However, none of the symptoms were found to influence the social subscale score independently (p > 0.05)., age, co-morbidity, stage of disease, and treatment modality together explained 6.2% of the variance in the FACT-G (C) emotional subscale score (p > 0.05) in the first step of the hierarchical multiple regression. Regression of the FACT-G (C) emotional subscale score against the symptom cluster as the second step revealed that the increase in explained variance of 30.1% was significant (F change = 12.1, p < 0.001). Mood disturbance (β = 0.338, p < 0.001) and fatigue (β = 0.249, p < 0.01) showed significant independent effects for the FACT-G (C) emotional subscale score, whereas insomnia and pain were not seen as influencing the FACT-G (C) emotional subscale score independently (p > 0.05). The hierarchical multiple regression for the FACT-G (C) functional subscale score as a dependent variable revealed that gender, age, co-morbidity, stage of disease, and treatment modality together explained 10.9% of the variance in the functional subscale score (p > 0.05) in the first step. Age showed significant independent effects for the functional subscale score (β = 0.218, p < 0.05). Regression of the FACT-G (C) functional subscale score against the symptom cluster as the second step revealed that the increase in explained variance of 41.1% was significant (F change = 21.6, p < 0.001). All the four symptoms of fatigue (β = 0.24, p < 0.01), mood disturbance (β = 0.228, p < 0.01), insomnia (β = 0.223,

6 132 K.K.F. Cheng, D.T.F. Lee / Critical Reviews in Oncology/Hematology 78 (2011) Table 4 Regression coefficients of the KPS and FACT-G(C) subscale/total scores against pain, fatigue, insomnia, and mood disturbance (N = 120). Step 1 Step 2 B SE β B SE β KPS Female Age * Co-morbidity Late Radiotherapy Chemoradiotherapy Pain *** Fatigue *** Insomnia Mood disturbance R 2 = for step 1; R 2 = 0.368, F=19.0 for step 2, p<0.001 FACT-G(C) Physical Female ** Age Co-morbidity Late *** Radiotherapy Chemoradiotherapy Pain Fatigue *** Insomnia * Mood disturbance *** R 2 = for step 1; R 2 = 0.529, F=33.8 for step 2, p<0.001 Social Female Age Co-morbidity Late Radiotherapy Chemoradiotherapy Pain Fatigue Insomnia

7 K.K.F. Cheng, D.T.F. Lee / Critical Reviews in Oncology/Hematology 78 (2011) Table 4(Continued ) Step 1 Step 2 B SE β B SE β Mood disturbance R 2 = for step 1, R 2 = 0.087, F=2.6 for step 2, p<0.05 Emotional Female Age Co-morbidity Late Radiotherapy Chemoradiotherapy Pain Fatigue ** Insomnia Mood disturbance *** R 2 = for step 1, R 2 = 0.301, F=12.1 for step 2, p<0.001 Functional Female Age * Co-morbidity Late Radiotherapy Chemoradiotherapy Pain ** Fatigue ** Insomnia ** Mood disturbance ** R 2 = for step 1, R 2 = 0.411, F=21.6 for step 2, p<0.001 Total Female Age Co-morbidity Late Radiotherapy Chemoradiotherapy Pain * Fatigue ** Insomnia *

8 134 K.K.F. Cheng, D.T.F. Lee / Critical Reviews in Oncology/Hematology 78 (2011) Table 4(Continued ) Step 1 Step 2 B SE β B SE β Mood disturbance *** R 2 = for step 1, R 2 = 0.494, F=25.6 for step 2, p<0.001 B, unstandardized coefficients; SE, standard error of unstandardized coefficients;, standardized coefficients; R 2, R 2 change; F, F change. * p < ** p < *** p < p < 0.01), and pain (β = 0.219, p < 0.01) showed significant independent effects for the FACT-G (C) functional subscale score. As the β coefficients of all of these four symptoms are similar in value, their effects on patients functional wellbeing are roughly equal., age, co-morbidity, stage of disease, and treatment modality together explained 4.7% of the variance in the FACT-G (C) total score (p > 0.05) in the first step of the hierarchical multiple regression. Regression of the FACT-G (C) total score against the symptom cluster as the second step revealed that the increase in explained variance of 49.4% was significant (F change = 25.6, p < 0.001). All the four symptoms of mood disturbance (β = 0.337, p < 0.001), fatigue (β = 0.278, p < 0.01), insomnia (β = 0.199, p < 0.05), and pain (β = 0.193, p < 0.05) showed significant independent effects for the FACT-G (C) total score. As demonstrated by the relative sizes of beta weights, the symptoms explaining the greatest proportion of the variances in overall well-being were mood disturbance and fatigue. 4. Discussion A diagnosis of cancer and its associated treatment not only presents physical symptoms but also a multitude of psychological symptoms to patients. Our data have shown that pain, fatigue, insomnia, and mood disturbance are highly prevalent, alone and in combination, in elderly patients who have undergone cancer therapy. Estimates from other studies have indicated that adult patients reported pain, fatigue, insomnia, and depression as high as 80%, 75%, 50%, 58%, respectively, during or after treatment for cancer [32 35], Kozachik revealed that % of the elderly patients had co-occurrence of pain, fatigue and insomnia during their first year after a cancer diagnosis [15]. The current study also showed that pain, fatigue, insomnia, and mood disturbance to be moderately distressing for elderly cancer patients during cancer therapy. A cross-sectional study of 263 cancer patients who were undergoing chemotherapy also found a moderate distress level for fatigue (mean = 2.64) and insomnia (mean = 2.08) as measured by a 5-point SDS [13]. Our findings also provide initial support for the notion that pain, fatigue, insomnia, and mood disturbance occur in a cluster. The mild to moderate inter-correlations among pain, fatigue, insomnia, and mood disturbance were greater than those found by previous studies. Dodd et al. showed small inter-correlations among pain, fatigue, and insomnia (r = 0.06 to 0.22) in 93 adults during their first three cycles of chemotherapy [11]. A secondary analysis of a sample of elderly patients (n = ) revealed small inter-correlations among pain, fatigue, and insomnia at 6 (r = ), 12 (r = ), 24 (r = ), and 52 (r = ) weeks following cancer diagnosis [15]. This discrepancy is likely to be related to the differences in the scales used to measure the symptoms and the dimensions used to determine the symptom experience. Other investigators have used a 3-point scale (mild, moderate, or severe) to measure the severity of pain, fatigue, and insomnia [14], and an 11-point Quality of Life-Cancer (QOL-CA) scale as a proxy measure for pain, fatigue, and insomnia [11]. Our study employed a 5-point SDS to capture distress associated with pain, fatigue, insomnia, and mood disturbance. Currently there is limited consensus regarding the dimensions of the symptoms to be evaluated for their inter-correlations and existence of symptom cluster. Some studies have used the dichotomous ratings of symptom occurrence (i.e., present or absent), while others have used severity or distress ratings. A study of 160 patients who were at the end of radiation therapy revealed that symptom clusters derived from ratings of severity rather than occurrence provide a more stable factor structure [36]. None of the studies compared the identified symptom clusters derived using the occurrence or severity ratings with those derived using the distress scores, an evaluation that needs to be considered in future studies. Of note there is also a lack of compelling evidence to guide researchers about the optimal size of the inter-correlations among symptoms for the symptoms to be considered a cluster. Loge et al. suggested that it is a necessary requirement of a symptom cluster that two or more concurrent symptoms be moderately correlated with one another (in the range of r = ) [37]. Nevertheless, strong inter-correlations among symptoms in a cluster do not necessarily make it a clinically meaningful symptom cluster. Consideration should be given to the effect of a symptom cluster on outcomes or consequences. Indeed, the effect of a symptom cluster on patient and clinical outcomes is crucial in validating the scientific construct of the particular symptom cluster, as well as identifying clinically important and useful symptom clusters. The literature has indicated that functional status and QoL are important measures of patient-reported outcomes; they

9 K.K.F. Cheng, D.T.F. Lee / Critical Reviews in Oncology/Hematology 78 (2011) can complement traditional biomedical outcomes in conveying important additional information for assessing the burden of a disease condition, and can help in the making of informed medical decisions [38]. Our results suggest an association between the symptom cluster, functional status, and QoL. Patients who had co-occurrence of pain, fatigue, insomnia, and mood disturbance reported the worst functional status and poorest QoL. Miaskowski et al. and Pud et al. analyzed 191 and 228 adults undergoing active cancer treatment, respectively, and also found that the subgroup of patients who had high levels of pain, fatigue, insomnia, and depression reported poor functional status and QoL [12,39]. In the current study, one-third to half of the variance ( %) in functional status as well as physical, emotional, functional well-being and total QoL was explained by the symptom cluster of pain, fatigue, insomnia, and mood disturbance in elderly patients receiving cancer therapy after adjustment for gender, age, co-morbidity, stage of disease, and treatment modality. On the other hand, all these four symptoms could explain only 8.7% of the variance in social well being. The variables explaining the greatest proportion of the variance in functional status were pain and fatigue. Dodd et al. also revealed that the symptoms of pain and fatigue contributed the most in explaining the change in functional status for younger patients (mean age 55 years) receiving chemotherapy [11]. Findings in this study also suggest that pain, fatigue, insomnia, and mood disturbance appear to be equally important in explaining the change in functional sphere of QoL for elderly cancer patients. More work is needed to test the robustness of this result and to determine the causal nature of the relationships among these symptoms on patients functional well-being. It was notable that mood disturbance had the greatest influence on physical, emotional and total QoL. Redeker et al. studied 263 patients undergoing chemotherapy (mean age 57 years) using the Profile of Mood States (POMS). They found that depression, fatigue, and anxiety affected patients QoL, with depression being the largest contributor [13]. Fox and Lyon found that depression and fatigue negatively influenced cancer survivors QoL (mean age 57 years), with depression, measured by the mental health subscale SF-36, being the largest contributor [40]. Undoubtedly, the experience of cancer to the individual is a life episode presenting emotional challenges. There is evidence that about one-third of elderly cancer patients may experience psychological distress. Prevalence rates of clinically relevant levels of depression in elderly cancer patients have been estimated to be up to 25% [41]. Although the literature suggests that the psychological impact of cancer may be less negative in the elderly compared to younger patients, the psychological distress that cancer and its treatment can produce in the elderly is real and should not be neglected [41]. Our data supported the notion that mood disturbance can have a profound effect on the physical, psychological, and overall well-being of elderly patients during cancer therapy. Nevertheless, it seems possible that mood disturbance was not merely additive in its influence on patients outcomes. Instead it occurred with pain, fatigue, and insomnia as a cluster of symptoms that reinforce each other in an interactive manner. Lenz et al. asserted that concurrent symptoms are likely to result in an experience that is multiplicative rather than additive [42]. In future, more empirical data are needed to support this theoretical proposition. In this study, age, gender, pre-existing co-morbidity, stage of disease, and treatment modality together explained about 10% of the functional status and QoL subscale/total scores, results similar to those of a previous study with younger patients (mean age 57 years) [13]. Results from this study also showed that age had an independent effect on functional status, as well as on the functional sphere of QoL. Neither gender nor the number of co-morbid conditions had significant independent effects on functional status and QoL. Dodd et al. also reported no difference in functional status between women and men [11]. Nevertheless, this finding diverges somewhat from Given et al. s report that being female and having 3 co-morbid conditions each had an independent effect on compromising physical function in elderly patients during the year after diagnosis [9]. This discrepancy may, in part, be a result of differences in sample size, the measurement techniques and the variety of co-morbid conditions being assessed, and the proportion of patients in the oldest age group. In this study, the proportion of those in the older age cohort that were age 75 or above was 15.8%, which is lower than those reported by Given et al. (32%) [9]. Piccirillo et al. indicated that the number and severity of co-morbid conditions increased with patients over the age of 80 [43]. In conclusion, our results suggest that pain, fatigue, insomnia, and mood disturbance may occur in a cluster and may negatively influence elderly patients functional status and QoL during cancer therapy. Data also highlight the high prevalence of pain, fatigue, insomnia, and mood disturbance, alone and in combination in elderly cancer patients. Nevertheless, the mechanism to explain the interrelationships among these four symptoms in a cluster is complex and multifactorial. Further prospective and longitudinal studies with larger sample sizes are clearly needed to test the robustness of these results and to explain the full spectrum of mechanisms underlying the occurrence and persistence of pain, fatigue, insomnia, and mood disturbance in the elderly cancer patient population. Further studies should also be directed towards assessing the long-term impact of co-occurrence of these four symptoms with more comprehensive evaluations including cognitive functioning, roles, and length of survival. A study using a triangulation approach combining quantitative and qualitative methods is also warranted to obtain in-depth information about the symptom cluster experience in elderly patients. Although cross-sectional, results from this study may be useful in directing comprehensive assessment and integrated intervention efforts for pain, fatigue, insomnia, and mood disturbance in elderly patients during cancer therapy.

10 136 K.K.F. Cheng, D.T.F. Lee / Critical Reviews in Oncology/Hematology 78 (2011) Reviewers Dr. Christine Miaskowski, Sharon A. Lamb Endowed Chair in Nursing and Associate Dean, UCSF School of Nursing, Dept. of Physiological Nursing, San Francisco, CA , United States. Dr. Winnie Kwok-wei So, Assistant Professor, The Chinese University of Hong Kong, The Nethersole School of Nursing, 7/F, Esther Lee Building, New Territories, Hong Kong, SAR China. Dr. Geoffrey A. Sonn, Stanford University, Department of Urology, Stanford, California , United States. Conflicts of interest None of the authors have any financial and personal relationships with other people or organizations that could inappropriately influence their work. Acknowledgement This study was partially supported by the Lee Hysan Foundation Fund, United College of the Chinese University of Hong Kong. References [1] Yancik R, Ries LAG. Cancer in older persons: an international issue in an aging world. Semin Oncol 2004;31(2): [2] Chang VT, Hwang SS, Feuerman M, et al. Symptom and quality of life survey of medical oncology patients at a Verterans Affairs Medical Center: a role for symptom assessment. Cancer 2000;88: [3] Donnelly S, Walsh D, Rybicki L. The symptoms of advanced cancer: identification of clinical and research priorities by assessment of prevalence and severity. J Palliat Care 1995;11(1): [4] Given B, Given C, Azzouz F, et al. Physical functioning of elderly cancer patients prior to diagnosis and following initial treatment. Nurs Res 2001;50(4): [5] Miaskowski C, Lee KA. Pain, fatigue, and sleep disturbances in oncology outpatients receiving radiation therapy for bone metastasis: a pilot study. J Pain Symptom Manage 1999;17(5): [6] Yates JW. Comorbidity considerations in geriatric oncology research. CA Cancer J Clin 2001;51: [7] Cohen HJ. Cancer and the functional status of the elderly. Cancer 1997;15: [8] Barsevick AM. The elusive concept of the symptom cluster. Oncol Nurs Forum 2007;35(5): [9] Given CW, Given B, Azzouz F, et al. Comparison of changes in physical functioning of elderly patients with new diagnoses of cancer. Med Care 2000;38: [10] Paice JA. Assessment of symptom clusters in people with cancer. J Natl Cancer Inst Monogr 2004;32: [11] Dodd MJ, Miaskowski C, Paul SM. Symptom clusters and their effect on the functional status of patients with cancer. Oncol Nurs Forum 2001;28(3): [12] Miaskowski C, Copper BA, Paul SM, et al. Subgroups of patients with cancer with different symptom experiences and quality of life outcomes: a cluster analysis. Oncol Nurs Forum 2006;33(5):E [13] Redeker NS, Lev EL, Ruggiero J. Insomnia, fatigue, anxiety, depression, and quality of life of cancer patients undergoing chemotherapy. Sch Inq Nurs Pract 2000;14(4): [14] Beck SL, Dudley WN, Barsevick A. Pain, sleep disturbance, and fatigue in patients with cancer: using a mediation model to test a symptom cluster. Oncol Nurs Forum 2005;32:E [15] Kozachik SL. Symptom clusters in elderly cancer patients. PhD Dissertation. Johns Hopkins University; [16] Theobald DE. Cancer pain, fatigue, distress, and insomnia in cancer patients. Clin Cornerstone 2004;6:S [17] So WKW, Marsh G, Ling WM, et al. The symptom cluster of fatigue, pain, anxiety, and depression and the effect on the quality of life of women receiving treatment for breast cancer: a multicenter study. Oncol Nurs Forum 2009;36:E [18] Bower JE, Ganx PA, Desmond KA, et al. Fatigue in breast cancer survivors: occurrence, correlates, and impact of quality of life. J Clin Oncol 2000;18: [19] Glover J, Dibble SL, Dodd MJ, et al. Mood states of oncology outpatients: does pain make difference? J Pain Symptom Manage 1995;10: [20] Gaston-Johansson F, Fall-Dickson JM, Bakos AB, et al. Fatigue, pain, and depression in pre-autotransplant breast cancer patients. Cancer Pract 1999;7: [21] Degner LF, Sloan JA. Symptom distress in newly diagnosed ambulatory cancer patients and as a predictor of survival in lung cancer. J Pain Symptom Manage 1995;10: [22] Gift AG, Jablonski A, Stommel M, et al. Symptom clusters in elderly patients with lung cancer. Oncol Nurs Forum 2004;31(2): [23] Sarna L. Correlates of symptom distress in women with lung cancer. Cancer Pract 1993;1:21 8. [24] Cheng KKF, Wong EMC, Ling WM, et al. Measuring the symptom experience of Chinese cancer patients: a validation of the Chinese Version of the Memorial Symptom Assessment Scale. J Pain Symptom Manage 2009;37(1): [25] Yates JW, Chalmer B, McKegney FP. Evaluation of patients with advanced cancer using the Karnofsky Performance Scale. Cancer 1980;45(8): [26] Mor V, Laliberte L, Morris JN, et al. The Karnofsky Performance Status Scale: an examination of its reliability and validity in a research setting. Cancer 1984;53: [27] Chan WH. The Chinese translation of the revised Piper Fatigue Scale and Symptom Distress Scale. Research Report. Hong Kong: The Nethersole School of Nursing, The Chinese University of Hong Kong; [28] Yu CLM, Fielding R, Chan CLW, et al. Measuring quality of life of Chinese cancer patients: a validation of the Chinese version of the functional assessment of cancer therapy-general (FACT-G) scale. Cancer 2000;88: [29] McCorkle R, Young K. Development of a symptom distress scale. Cancer Nurs 1978;1: [30] Cella DF, Tulsky DS, Gray G, et al. The functional assessment of cancer therapy scale: development and validation of the general measure. J Clin Oncol 1993;11: [31] Keith TZ. Multiple regression and beyond. Boston: Pearson Educ; [32] Savard J, Morin CM. Insomnia in the context of cancer: a review of a neglected problem. J Clin Oncol 2001;19: [33] Servaes P, Verhagen C, Bleijenberg G. Fatigue in cancer patients during and after treatment: prevalence, correlates and intervention. Eur J Cancer 2002;38: [34] Meuser T, Pietruck C, Radbruch L, et al. Symptoms during cancer pain treatment following WHO-guidelines: a longitudinal follow-up study of symptom prevalence, severity and etiology. Pain 2001;93: [35] Massie MJ. Prevalence of depression in patients with cancer. J Natl Cancer Inst Monogr 2004;32: [36] Kim E, Jahan T, Aouizerat BE, et al. Differences in symptom clusters identified using occurrence rates versus symptom severity ratings in patients at the end of radiation therapy. Cancer Nurs 2009;32:

11 K.K.F. Cheng, D.T.F. Lee / Critical Reviews in Oncology/Hematology 78 (2011) [37] Loge JH, Abrahamsen AF, Ekeberg O, Kassa S. Fatigue and psychiatric morbidity among Hodgkin s Disease survivors. J Pain Symptom Manage 2000;19:91 9. [38] Lipscomb J, Gotay CC, Snyder CF. Patient-reported outcomes in cancer: a review of recent research and policy initiatives. CA Cancer J Clin 2007;57: [39] Pud D, Ami SB, Cooper BA, et al. The symptom experience of oncology outpatietns has a different impact on quality-of-life outcomes. J Pain Symptom Manage 2008;35: [40] Fox SW, Lyon DE. Symptom clusters and quality of life in survivors of lung cancer. Oncol Nurs Forum 2006;33(5): [41] Kua J. The prevalence of psychological and psychiatric sequelae of cancer in the elderly how much do we know? Ann Acad Med 2005;34: [42] Lenz ER, Pugh LC, Milligan RA, et al. The middle-range theory of unpleasant symptoms: an update. Adv Nurs Sci 1997;19(3): [43] Piccirillo JF, Vlahiotis A, Barrett LB, et al. The changing prevalence of comorbidity across the age spectrum. Crit Rev Oncol/Hemat 2008;67: Biographies Prof. Karis Cheng, an internationally known researcher on cancer therapy-related complication and side effect management, has built several external competitive research grants studying symptom management, instrument development and validation, and quality of life issues. She has published over 80 refereed journal articles, book chapters and conference papers in the past 8 years. Her latest publications include papers in the Cancer and the Journal of Pain and Symptom Management. She is also the reviewer for a number of international cancer journals. Prof. Diana Lee is Chair Professor of Nursing and Director of the Nethersole School of Nursing and Assistant Dean of the Faculty of Medicine at the Chinese University of Hong Kong. She is also a Visiting Professor of the University of London in the United Kingdom and Advisory Professor of Fudan University and Guangzhou Medical College in China. Besides her teaching and administrative roles, Professor Lee has researched and published widely, especially in the area of gerontological nursing. She has successfully obtained over HK$ 95 million from various competitive research grants and has just been honored with an excellent research award in health services research by the Hong Kong Government. Most of her research efforts are focused on promoting the health and well-being of elderly people in residential care homes and in the community. She has expertise in a range of research methodologies. She has also established international joint research work, especially in China and the UK, on elderly care. Professor Lee has published over 160 refereed journal articles, book chapters and conference papers. Her latest publications include papers in the Journal of the American Geriatrics Society, the Gerontologist and Quality of Life Research. She is also the editor and reviewer for a number of international nursing and gerontological journals. Professor Lee has been invited to deliver keynote addresses in numerous international and national conferences.