Single- vs. Multiple-Item Instruments in the Assessment of Quality of Life in Patients with Advanced Cancer

Article Outline

• Abstract
• Introduction
• Methods
  • Patients
  • Assessment Scales
  • Data Collection
  • Statistical Analysis
• Results
• Discussion
• Acknowledgments
• References
• Copyright

Abstract 

Although multidimensional instruments are usually used to measure quality of life in advanced cancer patients, recent research suggests that single-item assessments can provide a reliable measure. Using the Functional Assessment of Cancer Therapy-General (FACT-G) instrument as a gold standard, we assessed the performance of the Edmonton Symptom Assessment System “feeling of well-being” (ESAS WB) item. We reviewed the data from 213 patients enrolled in six clinical trials. We determined the association between baseline ESAS WB and FACT-G total and subscale domain scores (Physical Well-being [PWB], Social/Family Well-being [SWB], Emotional Well-being [EWB], and Functional Well-being [FWB]. We also calculated the association between baseline (T1) and second (T2) observations of ESAS WB and of FACT-G total score. In addition, we predicted the change in FACT-G predicted by the ESAS WB score using regression analysis. Mean age was 60 (SD 12) years and 48% were female. The Spearman correlation coefficient of ESAS WB and FACT-G was −0.48 (P<0.0001). Correlations with FACT-G subscale domains were also highly significant, except for the SWB domain (P=0.08). The Pearson correlation coefficient for T1–T2 in ESAS WB and FACT-G for 146 patients was −0.36 (P<0.0001). The change in ESAS WB corresponding to FACT-G published minimally important difference was −0.24 for 3, −1.55 for 5, and −2.87 for 7, respectively. These results suggest that the single-item measure ESAS WB best reflects the total score on the FACT-G and PWB, EWB, and FWB domains but not on the SWB domain.

Key Words: Palliative care, quality of life, Edmonton Symptom Assessment System, FACT-G

Back to Article Outline

Introduction 

Patients with advanced cancer may experience physical, psychosocial, and spiritual difficulties throughout their illness that impact on their overall quality of life (QOL). The main objective of palliative care for advanced cancer patients is to improve the QOL for both patients and their families.¹ Therefore, it is necessary to systematically assess QOL to ascertain if this goal is being met.

QOL is a subjective experience that remains poorly defined.², ³ It is a complex multidimensional concept, which includes many domains or dimensions, such as physical well-being, functional ability, emotional and social well-being.⁴, ⁵ The importance of the existential domain and spiritual well-being also has been highlighted.⁶

Traditionally, it has been stated that because multiple dimensions impact on the construct of QOL, there is a need to use multidimensional assessments.⁴ Two commonly used core questionnaires are the Functional Assessment of Cancer Therapy-General (FACT-G)⁷ and the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-C30).⁸ As QOL is so diverse and dynamic in nature, it is possible that even standardized multidimensional instruments may not be comprehensive enough to cover all the possible QOL domains, which are important for an individual palliative care patient at a particular point in time. In view of this, some authors have suggested that single-item global assessments can provide a reliable measure of overall QOL.⁹, ¹⁰, ¹¹, ¹² Single-item tools for physical and emotional symptoms have been suggested as effective screening instruments in outpatients with cancer when compared with much longer comprehensive assessment tools.¹³, ¹⁴

As QOL is a dynamic process, regular systematic assessment of QOL by health care professionals is important to detect changes early¹⁵ and to open up communication between physician and patient.¹⁶ A practical, brief QOL assessment tool is invaluable in the busy clinical setting, particularly when patients are fatigued or frail, such as in the advanced cancer and palliative care population. A useful single-item tool must not, however, sacrifice the reliable representation of QOL in an effort to be less burdensome.¹⁷

There have been few studies comparing the QOL total score (index measure) and subsection scores (profile measure) from a valid and reliable multidimensional QOL questionnaire to a single-item question that directly assesses overall QOL or well-being.¹⁸ To our knowledge, no studies have formally assessed the performance of the “feeling of well-being” item in the Edmonton Symptom Assessment System (ESAS), a brief and widely used assessment tool for multiple symptoms in palliative care patients,¹⁹ and compared it with a validated multidimensional QOL instrument, such as the FACT-G. The purpose of the present study was to determine the correlations of the single-item measure ESAS “feeling of well-being” (WB) item with the longer multidimensional FACT-G.

Back to Article Outline

Methods 

This retrospective study was approved by the institutional review board at the University of Texas M. D. Anderson Cancer Center, with waiver of informed consent, as it was a secondary analysis of previously collected data.

Patients 

We analyzed the data collected from the charts of patients with advanced cancer who had been seen by the palliative care team at M. D. Anderson Cancer Center, a tertiary level comprehensive cancer center in Houston, Texas, or had been admitted to hospice care in the vicinity, and had participated in one of the six clinical trials conducted by our group in the time period of March 2006–June 2008. (All these trials are still currently open to the accrual of new patients.) For all these prospective trials, patients had provided written informed consent before participation. Patients were excluded from our study if they had not answered ESAS WB, or answered fewer than 22 of the 27 of the FACT-G (total score) items, or fewer than four items in each FACT-G subscale. For patients responding above these cutoff values, any missing data were prorated according to scoring guidelines.²⁰

For the first trial, the patients (n=44) were recruited if they had been admitted to hospice care, resided within 75 miles of the M. D. Anderson Cancer Center, and had reduced oral fluid intake. This was a randomized controlled trial to determine whether parenteral hydration is superior to placebo in improving symptoms associated with dehydration and in delaying the onset or reducing the severity of delirium in this patient population. The second trial (n=69) was a randomized placebo-controlled trial evaluating the as-needed use of the psychostimulant methylphenidate in the treatment of cancer-related fatigue. The third trial (n=18) was a randomized, double-blind, placebo-controlled study comparing the effect of dexamethasone vs. placebo. The fourth trial (n=29) was a randomized, double-blind, placebo-controlled study to determine the effect of low-dose thalidomide on the cancer cachexia symptom cluster (anorexia, fatigue, and weight loss). The fifth trial (n=43) was a randomized clinical trial of melatonin vs. placebo for appetite stimulation in advanced cancer patients. The sixth trial (n=18) was a randomized placebo-controlled trial examining the effect of mirtazapine on the appetite of patients with anorexia and weight loss. The day of primary outcome (T2) for all these trials was Day 15, except for the first trial (Day 7) and the fifth trial (Day 29).

Assessment Scales 

The ESAS is a 10-item tool that is widely used clinically to assist in the assessment of symptoms common to palliative care patients. In addition to the “feeling of well-being” item (ESAS WB item), the ESAS records the patient's current subjective perception of multiple symptoms (pain, fatigue, nausea, depression, anxiety, drowsiness, appetite, shortness of breath, and sleep).¹⁹ The patient rates each of these symptoms on a scale of 0–10, where zero indicates that the symptom is absent and 10 rates it at the worst possible severity. It should be noted that a score of zero out of 10 for the ESAS WB item represents the best “feeling of well-being” and 10 out of 10 represents the worst possible “feeling of well-being.” Hence, a higher score on ESAS WB signifies a worse “feeling of well-being,” which is the inverse to the FACT-G QOL scoring system. The ESAS was designed to be self-administered, but if needed, it can be completed with the assistance of the patient's caregiver (family or health care professional). The ESAS has been validated in a cancer patient population for internal consistency, criterion validity, and concurrent validity.²¹

The FACT-G is a validated QOL measure that was developed to evaluate patients receiving cancer treatment⁷ and is commonly used in oncology research. It is now a generic core questionnaire for the Functional Assessment of Chronic Illness Therapy (FACIT) measurement system.²² The palliative care department at our institution routinely incorporates these copyrighted tools into its research studies, including the FACIT-Fatigue (FACIT-F) measure and Functional Assessment of Anorexia/Cachexia Treatment (FAACT) measure. The FACT-G is designed to be answered in reference to the previous seven days. It was originally developed for patient self-administration but may be administered by an interviewer.

The FACT-G (version 4) contains 27 items covering four primary QOL domains in four subscales: Physical Well-being (PWB), Social/Family Well-being (SWB), Emotional Well-being (EWB), and Functional Well-being (FWB).²² The last statement on the FWB subscale (item GF7) is “I am content with the quality of my life right now.” All the FACT-G items have a five-point scale from zero to four for responses ranging from “not at all” to “very much.” The highest possible score for the EWB subscale is 24, and 28 for the other three subscales. Thus, the total FACT-G score can range from zero to 108, with higher scores indicating better QOL. Subscale scores can be prorated for defined missing data.²⁰ The FACT-G needs an overall item response rate greater than 80% to be an acceptable indicator of a patient's QOL.²⁰

Data Collection 

The FACT-G core questionnaire measures were derived from the FACIT-F measure in the first four of the included trials and from the FAACT measure in the remaining two trials. For this study, FACT-G total and four subscale domain scores and ESAS WB and nine ESAS symptom intensity scores at baseline (T1) and on the day of primary outcome for the study participant (T2) were reviewed. Demographic information included age, sex, marital status, race, and cancer diagnosis and was obtained from the databases of the original prospective trials or by chart review.

Statistical Analysis 

We determined the strength of associations between ESAS WB and the FACT-G total score using Pearson or Spearman correlation coefficients. We used the latter if the data were not approximately normally distributed. Similar correlation analyses were used to determine the strength of associations among ESAS WB, the FACT-G total score, each of the four FACT-G subscale domain scores, and the nine ESAS symptom intensity scores. In addition, we determined how changes in ESAS WB predicted changes in the FACT-G total score using a regression analysis. As part of this analysis, we also reported the values of the changes in ESAS WB that corresponded to the minimally important difference (MID) of the FACT-G total score. We also report the correlation coefficients between these change scores.

We powered the study assuming that we would have 220 patients included, and, therefore, would be able to declare correlations that were at least ±0.19 or greater as statistically significant, with a two-sided significance level of 0.05 and 80% power. At T1, these power considerations still apply. For T2 correlations and for correlations of differences between T1 and T2, including approximately 146 patients, we would be expected to be able to declare correlations that were at least ±0.23 or greater as statistically significant, with a two-sided significance level of 0.05 and 80% power.

We also determined the level of association between ESAS WB and item GF7 on the FACT-G (“I am content with the quality of my life right now”) using a correlation coefficient. In addition, we determined the associations between ESAS WB and also FACT-G and its subscales and the variable gender using a Wilcoxon two-sample test.

Back to Article Outline

Results 

There were 221 trial participants for the six clinical trials in the designated time frame. We evaluated information from baseline (T1) and the day of primary outcome (T2). Data were analyzed for 218 trial participants, as there were three exclusions at T1 (one patient had no EWB and thus no total FACT-G score and the other two excluded patients did not have any T1 data recorded). At T1, five patients had participated in two different trials on different occasions. Thus, demographic data for 213 unique patients were collected for this study. There were no T2 data for 59 trial participants as they had withdrawn from their respective trials before this time point. Subsequently, there were 13 further exclusions from our study because of incomplete T2 data. This resulted in 146 trial participants being evaluable at T2.

Patient characteristics are shown in Table 1. Mean age was 60 years and 48% were female. Table 2 reports the median ESAS symptom intensity scores, median ESAS WB score, FACT-G total and subscale domain scores at T1 and T2, and the difference between these two time points. In addition, the sum of the nine ESAS symptoms (excluding ESAS WB) is reported as the ESAS-9 score. It should be noted that for the first evaluated trial (parenteral hydration randomized placebo-controlled trial) ESAS sleep was not routinely collected, hence the lower number of patients (174) scoring this symptom. No significant associations were found between gender and ESAS WB or FACT-G and its subscales (all P>0.35).

Table 1. Patient Characteristics

	n (%)
Mean age, years (SD)	60 (12)
Female	102 (48)
Marital status
Married	135 (63)
Divorced/separated	37 (18)
Single	28 (13)
Widowed	13 (6)
Total	213 (100)
Ethnicity
Caucasian	137 (64)
African American	41 (19)
Hispanic	34 (16)
Asian	1 (<1)
Total	213 (100)
Diagnosis
Gastrointestinal	55 (26)
Lung	50 (23)
Breast	25 (12)
Urologic	23 (11)
Gynecologic	14 (7)
Head and neck	11 (5)
Other	35 (16)
Total	213 (100)

Table 2. Median ESAS and FACT-G Scores at Baseline (T1) and Day of Primary Outcome (T2) and Median Difference Between These Two Time Points

	Baseline (T1)		Day of Primary Outcome (T2)		Median difference (T1–T2)
	n	Median (IR)	n	Median (IR)	n	Median (IR)
ESAS scores
Pain	218	4 (2 to 6)	146	3 (0 to 5)	146	0 (−2 to 1)
Fatigue	218	6 (5 to 8)	146	4.5 (3 to 6)	146	−1 (−3 to 1)
Nausea	218	1 (0 to 4)	146	0 (0 to 2)	146	0 (−1 to 0)
Depression	218	2 (0 to 5)	146	0.5 (0 to 3)	146	0 (−2 to 0)
Anxiety	218	2 (0 to 5)	146	0 (0 to 3)	146	0 (−3 to 0)
Drowsinessa	216	4 (1 to 6.5)	146	2 (0 to 4)	145	0 (−3 to 0)
Dyspnea	218	2.5 (0 to 6)	146	1 (0 to 3)	146	0 (−2 to 1)
Appetite	218	6 (4 to 8)	146	4 (1 to 6)	146	−1 (−3 to 0)
Sleepb	174	4 (2 to 7)	131	3 (1 to 5)	131	−1 (−3 to 1)
ESAS WB	218	5 (3 to 7)	146	3.5 (2 to 5)	146	−1 (−3 to 1)
ESAS-9 scorec	173	31 (22 to 42)	131	21 (15 to 31)	130	−6 (−17 to 2)
FACT-G scores
PWB score	218	14.5 (11 to 18)	146	19 (14 to 22)	146	2 (−1 to 6)
SWB score	218	23 (20 to 27)	146	23 (19 to 27)	146	0 (−2 to 1.2)
EWB score	218	17 (12 to 20)	146	18 (15 to 20)	146	1 (−1 to 3)
FWB score	218	12 (9 to 18)	146	14 (10 to 18)	146	1 (−3 to 4)
Total FACT-G score	218	65 (55 to 78)	146	72 (58 to 83)	146	5 (−3 to 10)

IR=interquartile range.

Associations between ESAS WB and FACT-G total and subscale domain scores at T1 and T2 are shown in Table 3. The correlation coefficient of ESAS WB and FACT-G total score was −0.48 (P<0.0001). In consideration of the subscale scores, only the FACT-SWB domain was not significantly or highly correlated with ESAS WB (P=0.08 and 0.02 at T1 and T2, respectively).

Table 3. Associations^a Between ESAS WB Score and FACT-G Total and Subscale Scores at Baseline (T1) and Day of Primary Outcome (T2)

Table 4 shows the associations of difference scores (the difference between T1 and T2) of ESAS WB and FACT-G total score and subscale scores for 146 patients. These were all highly significant except for the SWB subscale domain, P=0.21.

Table 4. Associations^a Between the Change Scores (Between Day of Primary Outcome [T2] and Baseline [T1]) for FACT-G Total and Subscale Scores as Compared with ESAS WB (n=146)

The regression analysis, regressing ESAS WB difference scores on FACT-G difference scores, was highly significant (P<0.0001). The MID for FACT-G total score has not been completely established, but a range of possible MIDs (3–7) have been published.²⁰, ²³ We calculated that the average change in ESAS WB corresponding to FACT-G published MIDs was −0.24 for 3, −1.55 for 5, and −2.87 for 7, respectively. Using the same regression analysis, a change in ESAS WB of −1, 0, 1, 2, 3, 4, and 5 corresponded to a change in FACT-G total score of 4.2, 2.6, 1.1, −0.4, −1.9, −3.4, and −5, respectively. To further summarize the FACT-G MID data descriptively, in patients in who the FACT-G difference was ≥3, ≥5, and ≥7, the average difference in the ESAS WB was −1.75, −1.25, and −1.75, respectively.

We also determined the level of association between ESAS WB and item GF7 on the FACT-G (“I am content with the quality of my life right now”). We separately analyzed the GF7 item in the FWB domain of FACT-G because this item does ask a global question of QOL. The Pearson correlation coefficients between GF7 and ESAS WB at T1 and T2 were −0.3 and −0.4, respectively (each P<0.0001).

Back to Article Outline

Discussion 

In our study, we found that the correlation of ESAS WB, when administered as part of the 10-item ESAS instrument, and FACT-G total score was highly significant but only modest. The correlations of ESAS WB with the PWB, EWB, and FWB subscale domains also were modest. Other authors have examined the correlations of single-item QOL tools with a range of multiple-item assessment tools. In a four-week study comparing five QOL measures in 157 patients with advanced cancer, Donnelly et al.²⁴ have also shown a good correlation between a single-item QOL measure (a 10-point visual analog scale) with FACT-G, with a Pearson correlation coefficient of 0.6, P<0.0001. The 17-item McGill QOL questionnaire (MQOL) contains five distinct submeasures of PWB, physical and psychological symptoms, existential WB, and support.⁶ In its validation in 143 palliative care cancer patients, the MQOL total score significantly correlated with a single-item numerical rating 0–10 scale (SIS), but the correlation value varied according to the order of instrument administration. The Spearman correlation was 0.48 for SIS administered pre-MQOL compared with 0.66 if SIS followed the multi-item instrument.⁶ The 25-item Missoula-VITAS^® QOL index (MVQOLI) also contains five dimensions: symptom, function, interpersonal, well-being, and transcendent. The Pearson correlation coefficient of the MVQOLI total score with a five-point global QOL item, “How would you rate your overall QOL?” (with anchors of “best possible” and “worst possible”), was reported as 0.43 in its validation study in 257 community hospice patients.²⁵

In the present study, ESAS WB did not significantly correlate with the FACT-SWB subscale domain. Other authors, using other QOL measures, also have reported less agreement occurring with the SWB domain of FACT-G. In a study comparing the FACT-G and the EORTC QLQ-C30 in a population of 244 patients with a diagnosis of breast cancer or Hodgkin's disease, the Pearson correlation for the social domain of both instruments indicated very poor agreement, r=0.14.²⁶ Chang et al.,²¹ in their validation paper of the ESAS in a predominantly male cancer population, reported that the ESAS distress score (calculated by the sum of nine symptoms, which included ESAS WB but not sleep) had the lowest correlation with the SWB subscale (of FACT-G, version 3), with r=−0.25, compared with the three other FACT-G subscales. The SWB subscale of FACT-G emphasizes emotional closeness and includes communication and sexuality. In the ESAS format used in these studies, ESAS WB was the final item.²⁷ We hypothesize that ESAS WB may not be capturing the social domain as the ESAS focuses on physical and psychological symptoms, and the ESAS WB item may be being interpreted by respondents in the domains of the preceding questions. The position of a global QOL item has been shown to influence its score.⁶ Further research is needed to clarify how patients interpret “feeling of well-being” on the ESAS, and the impact of a change of item order of ESAS, moving the ESAS WB item toward the beginning of the tool.

In the present study, we found that the two global questions, GF7 on FACT-G (“I am content with the quality of my life right now”) and ESAS WB (“feeling of well-being”), correlated significantly (P<0.0001 at both time points) but not at a high level. One explanation may be that patients interpreted these two questions in a different manner. Satisfaction with overall QOL on GF7 may be interpreted in the psychosocial domain, whereas ESAS WB as part of the ESAS tool may focus interpretation more in the physical and functional domains. More research is needed.

This study had several limitations and, therefore, additional study is needed. The patients were a selected population who had already been enrolled in another prospective clinical trial. It is not known whether our results are applicable to patients with advanced cancer who elect not to participate in clinical trials and to noncancer palliative care patients. These patient groups may differ in terms of the self-weighting of values, personal preferences, and summation of factors important to their QOL at a particular time point. In addition, this study was retrospective—data quality was verified to ensure its accuracy and consistency of results. Future studies comparing QOL measures should be prospective and include noncancer palliative care patients and those who are not participating in other clinical trials. Our study only compared QOL measures at two defined time points (T1 and T2). Further studies are needed to assess the responsiveness and psychometric performance of ESAS WB to changes over time throughout the illness trajectory, including both patient improvement and patient deterioration.

To strengthen future research, prospective studies should be conducted assessing the correlation of ESAS WB with QOL measures, with the ESAS WB administered independently and also when included as an integral part of the ESAS, for example, when using naturalistic populations of varying characteristics and assessing the performance of the ESAS WB item against other QOL measures. Another important issue for future studies is that when ESAS WB is administered as the final item of the ESAS, it may function as a summing-up item, so the impact of changing the position of ESAS WB item toward the beginning of the ESAS questionnaire also should be studied. Better characterization of the relationship between single-item tools, such as ESAS WB, and multiple-item tools, such as FACT-G, will be needed to determine predictive validity, develop cutoffs, and establish case-finding capabilities of single-item tools in busy clinical settings.

In conclusion, our results suggest that the single-item measure ESAS WB, when administered as the final item of the 10-item ESAS instrument, best reflects the PWB, EWB, and FWB domains of FACT-G as compared with the SWB domain. These findings raise the possibility that there may be a role for a social domain item in ESAS. This may be particularly useful in patients still being cared for in the community to enable screening of social functioning. Future studies should also clarify patients' interpretation of “feeling of well-being.” Further research is needed in this area.

Back to Article Outline

Acknowledgments 

The authors thank Kathy R. King for her secretarial assistance and Research Nurse Cheryl Scott, BS,N for her assistance with data collection.

Back to Article Outline

References 

1. World Health Organization . WHO definition of palliative care. 2002. Available from http://www.who.int/cancer/palliative/definition/en/Accessed February 15, 2009
   • View In Article
2. Gill TM, Feinstein AR. A critical appraisal of the quality of quality-of-life measurements. JAMA. 1994;272:619–626
   • View In Article
   • MEDLINE
3. Kaasa S, Loge JH. Quality of life in palliative care: principles and practice. Palliat Med. 2003;17:11–20
   • View In Article
   • MEDLINE
   • CrossRef
4. Cella DF. Quality of life: concepts and definition. J Pain Symptom Manage. 1994;9:186–192
   • View In Article
   • MEDLINE
   • CrossRef
5. Sloan JA, Cella D, Frost M, et al. Assessing clinical significance in measuring oncology patient quality of life: introduction to the symposium, content overview, and definition of terms. Mayo Clin Proc. 2002;77:367–370
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (63 KB)
   • CrossRef
6. Cohen SR, Mount BM, Bruera E, et al. Validity of the McGill Quality of Life Questionnaire in the palliative care setting: a multi-centre Canadian study demonstrating the importance of the existential domain. Palliat Med. 1997;11:3–20
   • View In Article
   • MEDLINE
   • CrossRef
7. 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;11:570–579
   • View In Article
   • MEDLINE
8. Aaronson NK, Ahmedzai S, Bergman B, 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. J Natl Cancer Inst. 1993;85:365–376
   • View In Article
   • MEDLINE
   • CrossRef
9. de Boer AG, van Lanschot JJ, Stalmeier PF, et al. Is a single-item visual analogue scale as valid, reliable and responsive as multi-item scales in measuring quality of life?. Qual Life Res. 2004;13:311–320
   • View In Article
   • MEDLINE
   • CrossRef
10. Sloan JA, Frost MH, Berzon R, et al. The clinical significance of quality of life assessments in oncology: a summary for clinicians. Support Care Cancer. 2006;14:988–998
    • View In Article
    • MEDLINE
    • CrossRef
11. Kowalski C, Pennell S, Vinokur A. Felicitometry: measuring the “quality” in quality of life. Bioethics. 2008;22:307–313
    • View In Article
    • CrossRef
12. Spitzer WO, Dobson AJ, Hall J, et al. Measuring the quality of life of cancer patients: a concise QL-index for use by physicians. J Chronic Dis. 1981;34:585–597
    • View In Article
    • MEDLINE
    • CrossRef
13. Butt Z, Wagner LI, Beaumont JL, et al. Use of a single-item screening tool to detect clinically significant fatigue, pain, distress, and anorexia in ambulatory cancer practice. J Pain Symptom Manage. 2008;35:20–30
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (252 KB)
    • CrossRef
14. Vignaroli E, Pace EA, Willey J, et al. The Edmonton Symptom Assessment System as a screening tool for depression and anxiety. J Palliat Med. 2006;9:296–303
    • View In Article
    • MEDLINE
    • CrossRef
15. Cohen SR. Quality of life assessment in palliative care. In:  Bruera E,  Higginson IJ,  Ripamonti C,  von Gunten CF editor. Textbook of palliative medicine. London, UK: Hodder Arnold; 2006;p. 349–355
    • View In Article
16. Detmar SB, Muller MJ, Schornagel JH, Wever LD, Aaronson NK. Health-related quality-of-life assessments and patient-physician communication: a randomized controlled trial. JAMA. 2002;288:3027–3034
    • View In Article
    • MEDLINE
    • CrossRef
17. Sloan JA, Aaronson N, Cappelleri JC, Fairclough DL, Varricchio C. Assessing the clinical significance of single items relative to summated scores. Mayo Clin Proc. 2002;77:479–487
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (71 KB)
    • CrossRef
18. Cohen SR. Assessing quality of life in palliative care. In:  Portenoy RK,  Bruera E editor. Issues in palliative care research. New York: Oxford University Press; 2003;p. 231–241
    • View In Article
19. Bruera E, Kuehn N, Miller MJ, Selmser P, Macmillan K. The Edmonton Symptom Assessment System (ESAS): a simple method for the assessment of palliative care patients. J Palliat Care. 1991;7:6–9
    • View In Article
    • MEDLINE
20. Webster K, Cella D, Yost K. The Functional Assessment of Chronic Illness Therapy (FACIT) Measurement System: properties, applications, and interpretation. Health Qual Life Outcomes. 2003;1:79
    • View In Article
    • CrossRef
21. Chang VT, Hwang SS, Feuerman M. Validation of the Edmonton Symptom Assessment Scale. Cancer. 2000;88:2164–2171
    • View In Article
    • MEDLINE
    • CrossRef
22. Cella D. Functional assessment of chronic illness therapy (FACIT). Available from http://www.facit.org/Accessed March 2, 2009
    • View In Article
23. Yost KJ, Eton DT. Combining distribution- and anchor-based approaches to determine minimally important differences: the FACIT experience. Eval Health Prof. 2005;28:172–191
    • View In Article
    • MEDLINE
    • CrossRef
24. Donnelly S, Rybicki L, Walsh D. Quality of life measurement in the palliative management of advanced cancer. Support Care Cancer. 2001;9:361–365
    • View In Article
    • MEDLINE
    • CrossRef
25. Byock IR, Merriman MP. Measuring quality of life for patients with terminal illness: the Missoula-VITAS quality of life index. Palliat Med. 1998;12:231–244
    • View In Article
    • MEDLINE
    • CrossRef
26. Kemmler G, Holzner B, Kopp M, 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
    • View In Article
    • MEDLINE
27. Garyali A, Palmer JL, Yennurajalingam S, et al. Errors in symptom intensity self-assessment by patients receiving outpatient palliative care. J Palliat Med. 2006;9:1059–1065
    • View In Article
    • MEDLINE
    • CrossRef