Predictive Factors of Overall Well-Being Using the EORTC QLQ-C15-PAL Extracted from the EORTC QLQ-C30

Abstract

Objective:

The European Organization of Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire – Core 15 Palliative (EORTC QLQ-C15-PAL) was developed to assess quality of life (QOL) for the palliative cancer population to decrease patient burden. The purpose of this study was to compare predictive factors for well-being in the QLQ-C15-PAL extracted from the EORTC Quality of Life Questionnaire – Core 30 (QLQ-C30) with the QLQ-C30 itself.

Methods and Materials:

Patients with advanced cancer referred for treatment of bone metastases completed the QLQ-C30. Fifteen items from the QLQ-C15-PAL were extracted from the QLQ-C30. Univariate and multivariate analyses were used to determine predictive factors of the global QOL/health score in both tools. In the multivariate analyses, a p value of <0.003 indicated statistical significance.

Results:

Overall, predictive factors were similar when analyzing data from both tools. Predictive factors for the QLQ-C30 were role functioning (p<0.0001), fatigue (p<0.0001), nausea/vomiting (p<0.0001), and financial problems (p<0.0001) and factors for the extracted QLQ-C15-PAL were physical functioning (p<0.0001) and fatigue (p<0.0001).

Conclusions:

Extraction of the QLQ-C15-PAL items from the QLQ-C30 resulted in similar predictive QOL domains for all patient subgroups analyzed individually. The QLQ-C15-PAL is reflective of the QLQ-C30 domains and is recommended for future studies involving patients in a palliative setting, as this shorter questionnaire reduces patient burden and may increase accrual and compliance, while maintaining a similar breadth of coverage and achieving the same predictive ability.

Introduction

For patients with advanced cancer, survival may not be the most appropriate endpoint due to the poor life expectancy and great disease burden. Instead, preservation of quality of life (QOL) and symptom management arguably become more important.¹ QOL is a multidimensional construct that involves both physical and psychosocial aspects.¹ Standardized measurements of these subjective endpoints are collected through self-reported questionnaires completed by the patient themselves or by proxy, though there has been some controversy on the reliability of the latter.² This becomes the focus of treatment and response evaluation, as it gives health care providers a better understanding of the impact of cancer on the patient's QOL and may assist clinicians in prioritizing their approach in providing appropriate of care.³ Patients with advance cancer with bone metastases commonly receive radiotherapy for symptom management.⁴

Various valid questionnaires are available for the assessment of QOL in patients with cancer. The well-established European Organization of Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire – Core 30 (QLQ-C30) is a 30-item questionnaire commonly used in the evaluation of QOL in clinical trials.⁵ QLQ-C30 scores have not been limited only to evaluation of patient-reported treatment response; they have also been utilized to investigate the relationships between QOL and other factors, such as prognosis^6,7 and predicted treatment responses in women with advanced breast cancer.⁸

Significant symptom burden and poor performance status are common characteristics in patients with advanced cancer. Therefore, it may be onerous to complete the QLQ-C30, especially in studies where repeated assessments are required. This may lead to lower accrual rates and discontinuation of patient participation in studies. The EORTC has developed an abbreviated version (with all items extracted from the QLQ-C30) for palliative cancer patients, the Quality of Life Questionnaire – Core 15 Palliative (QLQ-C15-PAL), for the purpose of decreasing patient burden.⁹ A study by Steinmann and colleagues showed that use of the less burdensome QLQ-C15-PAL resulted in higher patient compliance and study participation for palliative patients with primary brain tumors.¹⁰

Despite the availability of this new assessment tool, the QLQ-C15-PAL is not frequently used to evaluate patient QOL in studies of palliative patients.^10,11 This may be due to the lack of reference data (i.e., published studies using the QLQ-C15-PAL) or due to the lack of studies documenting experience using this tool. In a survey conducted by Harding and coworkers that aimed to identify the assessment tools used in end-of-life care in Europe, it was found that the QLQ-C30 was used by clinicians 21 times as opposed to 15 for the QLQ-C15-PAL. Researchers also reported using the QLQ-C30 almost three times as often (48 versus 18).¹²

Since the original publication,⁹ there have been a few publications that have evaluated the use of the QLQ-C15-PAL.^11,13 In a cohort of 32 patients, Miyazaki et al. used intraclass correlation coefficients and equivalent form reliability to show that domain scores of QLQ-C15-PAL items taken from the QLQ-C30 showed sufficient reliability, with a high degree of coincidence with the original QLQ-C30.¹⁴ The extracted and shortened domains were able to explain original QLQ-C30 scores approximately 87% of the time. To our knowledge, there is scarce published literature validating the QLQ-C15-PAL using similar methods. The present study attempts to add to the literature by extracting the 15 items from the QLQ-C30 and comparing the respective predictive factors of overall QOL.

Methods and Materials

Patient population

Patients with advanced cancer referred for the treatment of bone metastases completed the QLQ-C30 at baseline across seven countries between March 2010 and January 2011. Baseline data collected included age, Karnofsky Performance Status (KPS), gender, education level, employment status, cohabitants, marriage status, primary cancer site, out/in patient status and whether they had previous systemic treatment. Research ethics approval was obtained prior to starting this research from all relevant ethics committees from all participating centers.

Quality-of-life instruments and scoring

The QLQ-C30 consists of 30 questions: five functioning scales including physical, role, cognitive, emotional, and social assessments; three symptom scales including fatigue, nausea/vomiting, and pain; five single-item scales that assess additional symptoms (appetite loss, constipation, diarrhea, dyspnea, and sleep disturbance); and a global QOL scale, consisting of two items. Patients rated each question/item on a numeric scale from 1 (not at all) to 4 (very much), with the exception of global QOL (Q29 and Q30), which was rated from 1 (very poor) to 7 (excellent). All scales are linearly transformed to a 0–100 scale. For the functioning scales and the global QOL scales, a higher score represents a better level of functioning. A high score for a symptom scale item represents greater symptom severity.

In the present study, items that were used in the QLQ-C15-PAL were extracted from the QLQ-C30 questionnaire. Fifteen items (Q1, Q2, Q6, Q7, Q10, Q15, Q17, Q20, Q22, Q23, Q25, Q26, Q27, Q28, and Q30) were dropped and the remaining items were renamed Q1-Q15. After the extraction, the following domains remained: physical functioning, emotional functioning, fatigue, nausea/vomiting, pain, dyspnea, insomnia, appetite loss, and constipation. Questions grouped in the same domain remained unchanged and the overall QOL question was based off a single item (Q15). Table 1 shows the QLQ-C30 questions used for the study and the items used for the scoring procedure.

Table 1.

EORTC QLQ-C30

QLQ-C30			Domain	Q# after extraction
Q1.	Do you have any trouble doing strenuous activities, like carrying a heavy shopping bag or a suitcase?		PF
Q2.	Do you have any trouble taking a long walk?		PF
Q3.	Do you have any trouble taking a short walk outside of the house?	√	PF	1
Q4.	Do you need to stay in bed or a chair during the day?	√	PF	2
Q5.	Do you need help with eating, dressing, washing yourself, or using the toilet?	√	PF	3
	During the past week:
Q6.	Were you limited in doing either your work or other daily activities?		RF
Q7.	Were you limited in pursuing your hobbies or other leisure time activities?		RF
Q8.	Were you short of breath?	√	DY	4
Q9.	Have you had pain?	√	PA	5
Q10.	Did you need to rest?		FA
Q11.	Have you had trouble sleeping?	√	SL	6
Q12.	Have you felt weak?	√	FA	7
Q13.	Have you lacked appetite?	√	AP	8
Q14.	Have you felt nauseated?	√	NV	9
Q15.	Have you vomited?		NV
Q16.	Have you been constipated?	√	CO	10
Q17.	Have you had diarrhea?		DI
Q18.	Were you tired?	√	FA	11
Q19.	Did pain interfere with your daily activities?	√	PA	12
Q20.	Have you had difficulty in concentrating on things, like reading a newspaper or watching television?		CF
Q21.	Did you feel tense?	√	EF	13
Q22.	Did you worry?		EF
Q23.	Did you feel irritable?		EF
Q24.	Did you feel depressed?	√	EF	14
Q25.	Have you had difficulty remembering things?		CF
Q26.	Has your physical condition or medical treatment interfered with your family life?		SF
Q27.	Has your physical condition or medical treatment interfered with your social activities?		SF
Q28.	Has your physical condition or medical treatment caused you financial difficulties?		FI
Q29.	How would you rate your overall health during the past week?	√	QOL	15
	1 2 3 4 5 6 7
	Very poor Excellent
Q30.	How would you rate your overall quality of life during the past week?		QOL
	1 2 3 4 5 6 7
	Very poor Excellent

√ signifies items that were used in the EORTC QLQ-C15-PAL.

AP, appetite loss; CF, cognitive functioning; CO, constipation; DI, diarrhea; DY, dyspnea; EF, emotional functioning; FA, fatigue; FI, financial problems; NV, nausea/vomiting; PA, pain; PF, physical functioning; QOL, quality of life; RF, role functioning; SL, insomnia.

Statistical analysis

Results were expressed as mean, standard deviation (SD), median and range for age, and KPS, as proportions for other sociodemographic and baseline clinical data. Normality tests were carried out on the domain scores of extracted items of QLQ-C30. Skewness and kurtosis were calculated, where significant p values rejected the null hypothesis, and natural log transformation was applied for domain scores to normalize the distribution. A simple univariate linear regression model was applied for detecting significant relationships between Q15 (overall QOL) and domain scores (nine domains). After conducting univariate linear regression analysis, backward selection procedure was used to select the most significant items related to Q15 in a multivariate model. Variables that were significant from univariate (p<0.05) were put in the multivariate analysis. Due to multiple comparisons, a Bonferroni-adjusted p value of <0.003 (0.05/15 items) indicated statistical significance. The coefficient, standard error (SE) of the coefficient, p value, and mean square error (MSE) were calculated.

To compare predictive factors obtained from extracted Q15 with original QLQ-C30, we conducted the same univariate and multivariate analyses as described above on the following three major outcomes: global health status, Q29, and Q30. All analyses were conducted using Statistical Analysis Software (SAS version 9.2 for Windows; SAS Institute Inc., Cary, NC).

Results

There were 447 patients from seven countries who responded to the QLQ-C30 questionnaire, and 91% of them were from Canada, Taiwan, or Cyprus. To obtain the homogeneity regarding the patients' perception and understanding of the QLQ-C15-PAL questions, patients from Brazil, Egypt, India, and France were excluded in the analysis. Therefore, a total of 364 patients from three countries with advanced cancer were accrued to complete the QLQ-C30 questionnaire prior to treatment for bone metastases (Table 2). Among the 364 patients, 361 patient scores were collected at baseline. Of these patients, 78.3% were female and 21.7% were male. Median age was 56 years (range 26–95) and median KPS was 80 (range 30–100). Of the patients enrolled, 61.8% were from Taiwan, 33.0% from Canada, and the remaining 5.2% from Cyprus. The most common primary cancer sites were breast (n=262, 72.0%), prostate (n=41, 11.3%), and lung (n=23, 6.3%). Three hundred and forty-nine patients (96.1%) were treated on an outpatient basis.

Table 2.

Demographics in Patients with Bone Metastases (n=364)

Sociodemographic data
Country	N (%)
Taiwan	225 (61.81%)
Canada	120 (32.97%)
Cyprus	19 (5.22%)
Age at baseline (years)
Mean (SD)	57.7 (13.0)
Median (range)	56 (26–95)
KPS at baseline
Mean (SD)	85.2 (15.4)
Median (range)	80 (30–100)
Gender
Female	285 (78.30%)
Male	79 (21.70%)
Education level
Elementary school	16 (11.43%)
High school	54 (38.57%)
University	33 (23.57%)
Masters or PhD	11 (7.86%)
Other	26 (18.57%)
Employment status
Retired	78 (54.93%)
Employed	42 (29.58%)
Unemployed	21 (14.79%)
Others	1 (0.70%)
Cohabitants
Spouse	61 (42.96%)
Spouse and child(ren)	32 (22.54%)
Alone	24 (16.90%)
Child(ren)	14 (9.86%)
Others	11 (7.74%)
Married status
Married	94 (66.20%)
Widowed	12 (8.45%)
Single	13 (9.15%)
Other	23 (16.20%)
Primary cancer site
Breast	262 (71.98%)
Lung	23 (6.32%)
Prostate	41 (11.26%)
Gastrointestinal	8 (2.20%)
Renal cell	8 (2.20%)
Other	22 (6.04%)
Patient status
Outpatient	349 (96.14%)
Inpatient	14 (3.86%)
Previous systemic treatment
Yes	107 (75.35%)
No	35 (24.65%)

KPS, Karnofsky Performance Status; SD, standard deviation.

For our study, the mean baseline symptom scores (Table 3) ranged from 11.7±21.26 (nausea/vomiting) to 46.3±31.92 (pain). Pain, fatigue, and insomnia were the most severe symptoms with mean domain scores of 46.3±31.92, 42.3±28.04, and 39.2±33.43, respectively. The mean baseline extracted QLQ-C30 scores for physical functioning, emotional functioning, and overall QOL were 63.6±26.24, 69.9±24.26, and 49.5±24.39, respectively.

Table 3.

Domain Scales of QLQ-C15-PAL Extraction from QLQ-C30 at Baseline

	N	Mean	SD	Median	Max.
Overall QOL	359	49.5	24.39	50.0	100
Physical functioning	353	63.6	26.24	73.3	93
Emotional functioning	361	69.9	24.26	66.7	100
Fatigue	360	42.3	28.04	33.3	100
Nausea/Vomiting	361	11.7	21.26	0.0	100
Pain	361	46.3	31.92	33.3	100
Dyspnea	359	22.1	25.30	33.3	100
Insomnia	361	39.2	33.43	33.3	100
Appetite loss	361	28.0	32.43	33.3	100
Constipation	360	26.6	31.38	33.3	100

QOL, quality of life; SD, standard deviation.

As in the original QLQ-C30 study looking at predictive factors of Q29, Q30 (QOL), and global health score (GHS), simple univariate and multivariate analyses were carried out for each of the QLQ-C15-PAL extracted domains. A positive coefficient between QOL and functioning scores indicated a higher QOL with increased functioning, and a negative coefficient between QOL and symptom scores indicated a lower overall QOL with increased symptom experience. Upon establishing the relationship between the domains scores with overall QOL, the significant predictive domains of the present study were compared with the predictive factors in the QLQ-C30 study (Table 4).

Table 4.

Predictive Factors of Q15 (Extracted from Original QLQ-C30) Compared with the Predictive Factors Obtained from Original QLQ-C30 Global Health Status (GHS) Scale and Q29, Q30 items, by Multivariate Linear Regression Analysis

QLQ-C15-PAL	Coefficient	SE	P value	QLQ-C30	Coefficient	SE	P value	Coefficient	SE	P value	Coefficient	SE	P value
	Intercept=0.920 (SE 0.139) MSE=0.164				Q29 Intercept=1.377 (SE 0.103), MSE=0.159			Q30 Intercept=1.443 (SE 0.105), MSE=0.045			GHS Intercept=3.237 (SE 0.124), MSE=0.684
PF	0.181	0.028	<0.0001	RF	0.087	0.014	<0.0001	0.089	0.015	<0.0001	0.216	0.028	<0.0001
FA	−0.109	0.017	<0.0001	FA	−0.090	0.021	<0.0001	−0.105	0.020	<0.0001
				NV	−0.050	0.012	<0.0001				−0.095	0.025	0.0001
				FI				−0.047	0.011	<0.0001	−0.085	0.022	0.0001

A p value of <0.003 indicates statistical significance.

FA, fatigue; FI, financial problems; PF, physical functioning; MSE, mean standard error; NV, nausea/vomiting; RF: role functioning; SE, standard error.

From the original QLQ-C30, predictive factors for overall QOL (Q29 and Q30) were role functioning (p<0.0001), fatigue (p<0.0001), nausea/vomiting (p<0.0001), and financial problems (p<0.0001) (Table 4). Role functioning (p<0.0001), nausea/vomiting (p=0.0001), and financial problems (p=0.0001) were significantly related to global health status from the original QLQ-C30. From the extracted QLQ-C15-PAL domains, in univariate analysis, overall QOL (Q15) resulted in a positive coefficient with regards to physical functioning (p<0.0001) and emotional functioning (p<0.0001) and a negative coefficient with regards to fatigue, nausea/vomiting, pain, dyspnea, insomnia, appetite loss (p<0.0001), and constipation (p=0.0003). From the multivariate analysis of the QLQ-C15-PAL, physical functioning (p<0.0001) and fatigue (p<0.0001) were significantly predictive of QOL (Q15).

Discussion

This is the first international study where items that were reflective of the QLQ-C15-PAL were extracted from the QLQ-C30 questionnaire and analyzed for their predictive coefficients with respect to self-reported QOL. We demonstrate that predictive factors were similar in the patients referred for bone metastases when compared with the original QLQ-C30 study.

For patients with bone metastases, complications include pathological fractures, spinal cord compression, hypercalcemia, and intractable pain. These complications lead to discomfort, limited physical functioning, fatigue, and a decreased QOL. External beam radiotherapy improves these aforementioned complications and results in overall pain relief response rates of about 60%.¹⁵

Fatigue was predictive of QOL in the QLQ-C30 and then again in the extracted QLQ-C15-PAL analysis. Other factors that were found to be predictive in the QLQ-C30 but not in the QLQ-C15-PAL, including role functioning and financial problems, were not part of our analysis because they were not reflective of items on the QLQ-C15-PAL and were dropped during the extraction. However, a study conducted by Lee and colleagues demonstrated that role functioning and physical functioning were highly correlated in a multitrait scaling analysis.¹⁶ Other authors have recommended combining these two scales¹⁷ on the basis that the items of the EORTC QLQ-C30 role functioning domain refer to limitations regarding work or household jobs, and are thus fairly closely associated with physical functioning.¹⁸ When compared with the Functional Assessment of Cancer Therapy – General (FACT-G), social domains of the QLQ-C30 have been shown to be more closely related to physical aspects rather than social supports and emotional closeness.¹⁵ Thus, with the exception of financial problems, the domains that were significant predictors of QOL, namely fatigue and physical functioning, concur with the previous results found in the QLQ-C30 analysis.

Fatigue is the most frequent symptom in palliative care.¹⁹ It can significantly decrease QOL in patients with advanced disease²⁰ and can be difficult to manage, often requiring a multidisciplinary approach. In our patient cohort, 40% reported at least moderate fatigue at baseline. Of the predictive domains determined in the present study, fatigue was the most significant predictor symptom for overall well-being. In addition, Zeng et al. demonstrated that fatigue was influenced by many factors (pain, nausea, depression, drowsiness, and dyspnea) and had a significant relationship with overall well-being.²⁰

We are limited in that the patient group selected to participate was determined based on the site of treatment, but these patients may have multiple sites of disease that are not accounted for and could confound our results. For example, a proportion of the patients referred for palliative radiotherapy for painful bone metastases may have had prior treatment for brain metastases, and may experience significant symptoms from both bone and brain metastases.

Gender differences among the patient population contributed to another limitation in the present study. Of the total cohort, 78.3% was female and 21.7% was male, generating a skewed patient population distribution. Studies have shown that gender differences have an impact on overall QOL within the same patient population.^21,22 Maino and coworkers found that in patients with primary brain tumors, females had lower QOL measurements and more incidences of depression when compared with male patients both pre- and post-surgical treatment.²³ The uneven distribution of females to males in the present cohort could potentially skew the results and may not be representative of the total palliative population as a whole.

The current study extracts QLQ-C15-PAL scores from the QLQ-C30 and analyzes for the predictive factors of QOL from the questionnaire. This method creates another potential drawback, because this study looks for overlap in predictive factors and coefficients between items that are essentially from the same data. This can be accounted for by the multivariate analyses, because domains were dropped in the extraction, therefore creating a different pool of variables that affect the overall QOL. Miyazaki et al¹⁴ conducted a similar study involving a QLQ-C15-PAL extraction and utilizing intraclass correlation coefficients to compare domains with the original QLQ-C30. As the extracted domains did not include all the items from the original QLQ-C30, the inputs for the intraclass correlation coefficients would have been different.

Although there have been several editorials discussing the use of the QLQ-C15-PAL,^24,25 along with many papers encouraging the planned use of QLQ-C15-PAL in upcoming studies,^26,27 only a few studies to date have assessed QOL using an abbreviated questionnaire.^10,11,13 In a study comparing use of the QLQ-C30 with the QLQ-C15-PAL in a patient population with brain metastases, Steinman et al.¹⁰ demonstrated that the QLQ-C15-PAL was more practical and generated higher patient compliance.

Our results confirm the validity of the QLQ-C15-PAL, in that predictive domains of overall QOL remained the same despite the removal of some items. The shorter QLQ-C15-PAL is a more time efficient and appropriate measure of QOL. Previous studies have already found better compliance with the QLQ-C15-PAL compared with the QLQ-C30.¹⁰ Unfortunately, the QLQ-C15-PAL is unable to assess certain domains of the QLQ-C30 due to the shortening. The present study validates the reliability of the QLQ-C15-PAL questionnaire as an effective assessment tool of QOL, in comparison with the QLQ-C30. Future studies in patients in a palliative setting should continue to use the QLQ-C15-PAL over the span of patients' treatment as it reduces patient burden and may increase compliance and decrease attrition in the clinical trial setting.

Footnotes

Acknowledgments

We thank the generous support of the Bratty Family Fund, the Michael and Karyn Goldstein Cancer Research Fund, the Joseph and Silvana Melara Cancer Research Fund, and the Ofelia Cancer Research Fund.

Author Disclosure Statement

No competing financial interests exist.

References

Kirkbride

, Tannock

. Trials in palliative treatment-have the goal posts been moved? Lancet Oncol, 2008; 9:186–187.

Nekolaichuk

, Bruera

, Spachynski

, MacEachern

, Hanson

, Maguire

. A comparison of patient and proxy symptom assessments in advanced cancer patients. Palliat Med, 1999; 13:311–323.

Gotay

. Assessing cancer-related quality of life across a spectrum of applications. J Natl Cancer Inst Monogr, 2004; 33:126–133.

Chow

, Wu

JSY

, Hoskin

, Coia

, Bentzen

, Blitzer

. International consensus on palliative radiotherapy endpoints for future clinical trials in bone metastases. Radiother Oncol, 2002; 64:275–280.

Aaronson

, Ahmedzai

, Bergman

et al. The European Organization for Research and Treatment of Cancer QLQ-C30: A quality-of-life instrument for use in international clinical trials in oncology. JNCI Cancer Spectrum, 1993; 85:365–376.

Mauer

, Taphoorn

, Bottomley

et al. Prognostic value of health-related quality-of-life data in predicting survival in patients with anaplastic oligodendrogliomas, from a Phase III EORTC brain cancer group study. J Clin Oncol, 2007; 25:5731–5737.

Quinten

, Coens

, Mauer

et al. Baseline quality of life as a prognostic indicator of survival: A meta-analysis of individual patient data from EORTC clinical trials on behalf of the EORTC Clinical Groups. Lancet Oncol, 2009; 10:865–871.

Kramer

, Curran

, Piccart

et al. Identification and interpretation of clinical and quality of life prognostic factors for survival and response to treatment in first-line chemotherapy in advanced breast cancer. Eur J Cancer, 2000; 36:1498–1506.

Groenvold

, Petersen

et al.; EORTC Quality of Life Group:The development of the EORTC QLQ-C15-PAL: A shortened questionnaire for cancer patients in palliative care. Eur J Cancer, 2006; 42:55–64.

10.

Steinmann

, Schäfer

, Oorschot

et al.

Effects of radiotherapy for brain metastases on quality of life (QoL)

Strahlenther Onkol, 2009; 185:190–197.

11.

Suárez-Del-Real

, Allende-Pérez

, Alférez-Mancera

et al. Validation of the Mexican–Spanish version of the EORTC QLQ-C15- PAL questionnaire for the evaluation of health-related quality of life in patients on palliative care. Psychooncology, 2011; 20:889–896.

12.

Harding

, Simon

, Benalia

et al. The PRISMA symposium 1: Outcome tool use. Disharmony in European outcomes research for palliative and advanced disease care: Too many tools in practice. J Pain Symptom Manage, 2011; 42:493–500.

13.

Shin

, Choi

, Miyashita

et al. Cross-cultural application of the Korean version of the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire-Core 15-Palliative Care. J Pain Symptom Manage, 2011; 41:478–484.

14.

Miyazaki

, Suzukamo

, Shimozuma

, Nakayama

. Verification of the psychometric properties of the Japanese version of the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire Core 15 Palliative (EORTCQLQ-C15-PAL) Qual Life Res, 2012; 21:335–340.

15.

Chow

, Zeng

, Salvo

, Dennis

, Tsao

, Lutz

. Update on the systematic review of palliative radiotherapy trials for bone metastases. Clin Oncol, 2012; 24:112–124.

16.

Lee

, Chun

, Wang

, Lim

, Choi

. Multidimensional constructs of the EORTC Quality of Life Questionnaire (QLQ-C30) in Korean cancer patients with heterogeneous diagnoses. Cancer Res Treat, 2005; 37:148–156.

17.

Ringdal

, Ringdal

. Testing the EORTC quality of life questionnaire on cancer patients with heterogeneous diagnoses. Qual Life Res, 1993; 2:129–140.

18.

Kemmler

, Holzner

, Kopp

, Dunser

, Margreiter

, Greil

et al. Comparison of two quality of life instruments for cancer patients: The functional assessment of cancer therapy-general and the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire-C30. J Clin Oncol, 1999; 17:2932–2940.

19.

Stone

, Hardy

, Broadley

, Tookman

, Kurowska

, A'Hern

. Fatigue in advanced cancer: a prospective controlled cross-sectional study. Br J Cancer, 1999; 79:1479–1486.

20.

Zeng

, Koo

, Zhang

et al. Fatigue in advanced cancer patients attending an outpatient palliative radiotherapy clinic as screened by the Edmonton Symptom Assessment System. Support Care Cancer, 2012; 20:1037–1042.

21.

Norris

, Ghali

, Galbraith

, Graham

, Jensen

, Knudtson

. Women with coronary artery disease report worse health-related quality of life outcomes compared to men. Health Qual Life Outcomes, 2004; 5:21

22.

Agewall

, Berglund

, Henareh

. Reduced quality of life after myocardial infarction in women compared with men. Clin Cardiol, 2004; 27:271–274.

23.

Mainio

, Hakko

, Niemela

, Koivukangas

, Rasanen

. Gender difference in relation to depression and quality of life among patients with a primary brain tumor. Eur Psychiatry, 2006; 21:194–199.

24.

Echteld

, Onwuteaka-Philipsen

, van der Wal

, Deliens

, Klein

. EORTC QLQ-C15-PAL: The new standard in the assessment of health-related quality of life in advanced cancer? Palliat Med, 2006; 20:1–2.

25.

Groenvold

, Petersen

, Aaronson

, Arraras

, Blazeby

, Bottomley

, Fayers

et al.

EORTC QLQ-C15-PAL: The new standard in the assessment of health-related quality of life in advanced cancer?

Palliat Med, 2006; 20:59–61.

26.

Caissie

, Zeng

, Nguyen

et al. Assessment of health-related quality of life with the European Organization for Research and Treatment of Cancer QLQ-C15-PAL after palliative radiotherapy of bone metastases. Clin Oncol, 2012; 24:125–133.

27.

Caissie

, Culleton

, Nguyen

et al. EORTC QLQ-C15-PAL quality of life scores in patients with advanced cancer referred for palliative radiotherapy. Support Care Cancer, 2012; 20:841–848.