© The Author 2007. Published by Oxford University Press.
Differences in What Quality-of-Life Instruments Measure
Affiliations of author: College of Nursing, University of Illinois at Chicago, Chicago, IL
Correspondence to: Carol Estwing Ferrans, PhD, RN, FAAN, College of Nursing, University of Illinois at Chicago, 845 South Damen Ave (MC 802), Chicago, IL 60612 (e-mail: cferrans{at}uic.edu).
 |
ABSTRACT |
|---|
 Top Abstract Value of Quality-of-Life...References |
|---|
To address the question of what value is added by assessing quality of life (QOL) in symptom management trials in cancer, we used the model of Wilson and Cleary to identify what is measured by the most commonly used QOL instruments. Examples of clinical trials are presented demonstrating the contributions of these broad-based QOL instruments in terms of symptoms, functioning, general health perceptions, and overall QOL. The examples show that QOL instruments can provide valuable information about side effects and impact on other aspects of life, which would not be captured by a more narrowly focused measure of the target symptom. A better understanding is needed of the differences in what QOL instruments measure, since conclusions regarding the effectiveness of treatment may differ depending on which one is used to assess outcomes. Head-to-head comparisons of instruments within the same studies would increase precision for selecting QOL instruments for symptom management trials.
|
Value of Quality-of-Life Instruments in Symptom Management Trials in Cancer |
|---|
|
TopAbstractValue of Quality-of-Life...References |
|---|
Based on a review of all the symptom management trials funded by the National Cancer Institute (NCI) to date, Buchanan et al. (1) concluded that investigators in general have not provided compelling rationale for the use of global quality-of-life (QOL) measures or the selection of a particular QOL instrument. The purpose of this paper is to address the question of what value is added by measuring QOL in the context of symptom management trials. In these studies, QOL instruments would not primarily be used to measure the effect of treatment on the target symptom but instead would be used to determine the diffusion of the effect on other aspects of life. In other words, they would be used to assess the ripple effect of the change in symptoms produced by the intervention.
QOL instruments have been characterized most commonly in terms of the domains they measure, such as physical, psychological/spiritual, social, economic, and family domains (2). The subscales of an instrument normally identify the domains, sometimes with multiple subscales for a single domain. But the subscale names provide only partial information about what an instrument measures. For example, a physical domain subscale may contain items assessing symptoms, functioning, or adjustment to illness. It also may assess the status of an attribute, such as the level of pain, or judgments about the attribute, such as satisfaction with the level of pain or pain control (3). This is the case not only for the physical domain but also for the psychologic, social, and other domains.
Model of Health-Related Quality of Life of Wilson and Cleary
To better characterize what health-related QOL instruments measure, Wilson and Cleary (4) developed a model that identifies the conceptual approaches used by various instruments (Fig. 1). The arrows indicate the dominant causal associations. The main components are the five boxes in the middle of the figure. The first box, biological and physiological variables, focuses on the function of cells, organs, and organ systems. The second box is symptom status, which refers to physical, cognitive, and emotional symptoms perceived by the patient. The third component is functional status, which includes functioning in psychologic and social domains, as well as physical functioning. The next box, general health perceptions, refers to the integration of all of the health concepts that precede it, as perceived by the patient. The final box, overall QOL, refers to patients' own evaluation of their QOL, such as how happy or satisfied they are with life as a whole. This would include measures of life satisfaction and global QOL.
|
It should be noted that this first component of the model of Wilson and Cleary, biological and physiologic variables, does not actually represent a type of QOL measure because it is assessed using objective indicators rather than patient-reported outcomes (5). Nevertheless, because biological and physiologic variables provide the basis for the following four boxes in the model, it provides an appropriate starting point. Moving from the first component to second enlarges the focus from the cellular and organ level to that of the entire person, and consequently, the following four boxes all are measured in terms of patient-reported outcomes (5).
Components of the Model Measured by Quality-of-Life Instruments
The Functional Assessment of Cancer Therapy/Functional Assesment in Chronic Illness Therapy (FACT/FACIT) (6,7), the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-C30) (8,9), and the SF-36 Health Survey (SF-36) (10,11) are the most commonly used QOL instruments in clinical trials for chemotherapeutic agents in cancer. To identify what these three instruments measure, we categorized their items according to the major components of the model of Wilson and Cleary. First, the items of the FACT-General (FACT-G) were analyzed, grouped by the instrument's four subscale domains (physical, social, emotional, and functional). Figure 2 displays the percentage of items that measure symptoms, functioning, general health perceptions, and overall QOL in each domain. The analysis showed that for all four domains, the largest percentage of items focused on symptoms, such as "I have lack of energy" and "I have pain." The next largest percentage of items focused on functioning, such as "I am forced to spend time in bed."
|
The items of the EORTC QLQ-C30 and SF-36 also were analyzed according to the model of Wilson and Cleary, in the same manner as the FACT-G. As seen in Fig. 3, the analysis showed that all three instruments primarily measure symptoms and functioning, with the FACT and QLQ-C30 containing larger proportions of symptom items and the SF-36 larger proportions of functioning items. This emphasis on symptoms and functioning is consistent with the fact that the FACT and QLQ-C30 originally were developed for use in clinical trials of cancer therapeutic agents and the SF-36 was developed as a measure of health status.
|
Symptoms and Functioning
Table 1 provides examples of symptom and functioning items from the EORTC QLQ-C30 and the SF-36. In the context of a symptom management trial, the relevance of items such as these may not be immediately apparent. However, a QOL instrument can provide valuable information about side effects and impact on other aspects of life, which would not be captured by a more narrowly focused measure of the target symptom. For example, a recent randomized trial tested whether the cytoprotectant amifostine would reduce chemoradiotherapy-induced esophagitis in non–small-cell lung cancer patients (12). Previous small studies of amifostine had produced mixed results, using only the NCI Common Toxicity Criteria to evaluate esophagitis. In this study, the addition of the QOL measures provided a more comprehensive evaluation of amifostine, leading to a clearer idea of its positive and negative effects. The amifostine group experienced statistically significant improvements in pain and less deterioration, as shown by the EORTC QLQ-C30, as well as less dysfunction in swallowing. However, grade 3 or greater esophagitis was only slightly better in the amifostine group (30% with amifostine versus 34% without). In light of the negative effects of amifostine (higher rates of acute nausea and vomiting, cardiovascular toxicity, infection, and febrile neutropenia), it was concluded that further examination of amifostine should be limited to alternate routes of administration, such as subcutaneous, with lower doses.
|
One of the clearest examples of the value of a broad-based instrument in a randomized phase III trial is found in the International Randomized IFN versus STI571 study. This study compared imatinib versus interferon alpha plus low-dose cytarabine (IFN + LDAC) in newly diagnosed patients with chronic phase chronic myeloid leukemia (13). In this study, the FACT provided important information about symptoms and functioning in multiple domains, demonstrating the superiority of imatinib for first-line treatment. In the physical domain, the FACT demonstrated that imatinib produced fewer symptoms due to side effects and better physical functioning as early as the first month of treatment and across the 12 months of the study. These differences also were seen within the first month in patients who crossed over from IFN + LDAC to imatinib. Patients receiving IFN + LDAC reported more trouble sleeping, lack of energy, feeling weak, tiredness, trouble concentrating, nausea, and fever and chills, as measured by the FACT. By the second month of the trial, symptoms and functioning in the emotional and social domains were significantly better in the imatinib group. The differences between treatment arms were statistically significant as well as clinically meaningful and represented some of the largest treatment group differences in QOL reported in a clinical trial.
Another landmark study demonstrating the value of QOL assessment was a randomized phase III trial comparing azacytidine versus supportive care in patients with myelodysplastic syndrome (MDS) (14). At the time of the study, overall median survival for high-risk MDS ranged from 6 to 12 months. The only effective treatment was allogeneic bone marrow transplantation, which was an option for only a small percentage of patients. As the primary measure of QOL, the EORTC QLQ-C30 demonstrated improvement in symptoms and functioning that were both statistically significant and clinically important. Over the course of the study, fatigue and dyspnea, physical functioning, positive affect, and psychologic distress improved in the azacytidine arms. Those who crossed over from supportive care to azacytidine experienced similar improvements in symptoms and functioning. In contrast, for patients receiving supportive care only, QOL was generally stable or declined over time. Improvement in QOL paralleled the clinical findings of the study, with better treatment response and reduced frequency of and transformation to acute myelogenous leukemia or death, establishing azacytidine as an important treatment option for MDS.
General Health Perceptions
The term "general health perceptions" refers to a synthesis of all the various aspects of health into an overall evaluation. This component of the model of Wilson and Cleary is most commonly measured by a single global question asking people to rate their health in general on a response scale ranging from poor to excellent, as seen in the EORTC QLQ-C30 and the SF-36. In addition, the SF-36 also asks respondents to compare their general health now with their health a year ago, their anticipated health in the future, and the health of others they know. (The FACT-G does not contain items assessing general health perceptions.) The value of the general health rating in a symptom management trial was seen in a small study comparing two types of homeopathy for the treatment of menopausal symptoms in breast cancer survivors (15). In this trial, the symptom-specific measure for hot flashes showed a trend toward improvement, but it was not statistically significant. The general health scale of the SF-36, however, demonstrated statistically significant improvement for both groups receiving homeopathic treatment, as compared with the placebo group.
Overall Quality of Life
Of particular interest is the final component of the model of Wilson and Cleary. Since the model itself focuses on QOL, this final component could be considered to be the quintessential element of the model. It differs from the rest in that it provides the overall evaluation of QOL, a summation of all the components that come before it in the model. This can be measured through multiple-item scales assessing life satisfaction or happiness, or it can be measured by a single global QOL question. The Spitzer Uniscale (16) is an example of such a global question. In symptom management trials funded by NCI, the Uniscale is one of the most commonly used measures of QOL (1). The Uniscale consists of a single item that states, "Please rate your overall quality of life." Other examples of global QOL questions are found in the FACT-G and EORTC QLQ-C30. The item for the FACT-G is "I am content with the quality of my life right now," and the QLQ-C30 asks, "How would you rate your overall quality of life during the past week?" (The SF-36 does not include a global QOL question.)
These global QOL questions differ from total scores of QOL instruments in that respondents decide for themselves what to consider in rating the quality of their lives, based on what is important to them. This differs from a typical multi-item instrument in which items are summed to produce the total QOL score, with everything given equal value. In this case, the QOL score is based on whatever the authors of the instrument thought was important to include in their tool. This may or may not coincide with what is important to respondents when evaluating their QOL.
Patient-reported outcomes in general help to provide a better understanding of the impact of illness from the patient's viewpoint. However, overall QOL differs from measures of symptoms, functioning, and perceived health status by providing an evaluation of the whole of life, using the patient's own personal values. It is this incorporation of the patient's values that sets QOL assessment apart from measures of health status. This is a particularly salient consideration for interventions aimed at providing comfort, such as symptom trials. In addition, if the symptom trials are focused on palliation or the end of life, there are even greater reasons to evaluate the wider effectiveness of treatment in terms of overall QOL. This is because the ultimate purpose for health care interventions in these cases is to maximize QOL.
There is evidence that overall QOL is a distinct concept from the other components of the model of Wilson and Cleary, providing different information than symptoms, functioning, and perceived health status. In a study of 493 older patients, Covinsky et al. (17) demonstrated a lack of concordance between patient ratings of health and their global QOL. They found that 43% of those with the worst physical functioning rated their global QOL as good or better. Conversely, of those with the best physical functioning, 15% thought their QOL was only fair or poor. They also found a lack of concordance for psychologic health and QOL. Of those with the fewest psychologic symptoms, 21% rated their QOL as only fair or poor. In addition, patients consider different aspects of life when rating QOL than when rating health status. Based on a meta-analysis of 12 studies in chronic disease, Smith et al. (18) developed path models that demonstrated that patient ratings of their health and global QOL were influenced by different things. Perceived health status was most affected by physical functioning and to a lesser extent by emotional well being. On the other hand, global QOL was affected to the greatest extent by emotional well-being and less by physical functioning. A logical extension of these findings is that conclusions may differ depending on whether outcomes are measured in terms of health status or global QOL, particularly if there is lack of agreement between them. This provides an argument for the inclusion of overall QOL, in addition to measures of symptoms, functioning, and health status.
Because global QOL can be affected by things others than health status, some have questioned its contribution to symptom management trials. The concern is that QOL may be too far "downstream" to be sensitive enough to reflect important changes in the target symptoms. A phase II study of temozolomide in recurrent anaplastic astrocytoma (19) demonstrated that the effect of treatment on global QOL could be as great as the effect on symptoms. In this study, the EORTC Brain Cancer Module measured the target symptoms specific to astrocytoma, such as visual disorders, motor dysfunction, headaches, and seizures. The core instrument of the EORTC QLQ-C30 was used to measure the wider impact on QOL. For the purpose of this discussion, we will focus on the additional contribution made by the core instrument, over and above the Brain Cancer Module. Figure 4 presents the changes in QOL for patients with disease progression and for those who were progression free after 6 months of temozolomide. Regarding symptoms, the QLQ-C30 core instrument showed that the progression-free group experienced improvement in pain, fatigue, and insomnia. In terms of functioning, the progression-free group showed improvement in both the emotional functioning and social functioning scales, which were of even greater magnitude than the changes in symptoms. In global QOL, the progression-free group experienced statistically significant improvement, while the disease progression group declined. The changes in global QOL were as large or larger than the changes in the symptoms of pain, fatigue, insomnia, and dyspnea. Because median survival after recurrence is short and treatment only modestly effective, QOL was of prime importance in this study. The changes in QOL were both clinically and statistically significant, resulting in the recommendation for further evaluation of temozolomide in a phase III trial.
|
Conclusions and Recommendations
Over the past 30 years, the field of QOL assessment has become sophisticated and methodologically rigorous, with many well-established instruments available for use in cancer care. As seen in the studies presented, QOL instruments have made important contributions to therapeutic clinical studies. The value of any particular QOL instrument to a symptom management trial will be dependent on the match between the study's purpose and the information the instrument provides. The instrument's sensitivity, specificity, and interpretability all ultimately depend on the conceptual approach taken to measure QOL (3).
As shown by our item analysis, the instruments used most commonly in clinical trials of therapeutic agents focus primarily on symptoms and functioning. However, the content of QOL instruments will vary depending on the purposes for which they were originally developed. For example, an instrument focusing on rehabilitation needs of cancer patients, such as the Cancer Rehabilitation Evaluation System, contains content on disruption of daily activity caused by disease and treatment (20). The McGill Quality of Life Questionnaire, which was developed for patients with advanced disease, emphasizes meaning in life and existential concerns (21). The Quality of Life Index was developed as a measure of satisfaction with life in the context of cancer (22).
A better understanding is needed of the differences in what QOL instruments measure, as well as greater clarity regarding the distinctions between symptoms, functional status, general health status, and global QOL. Conclusions regarding the effectiveness of treatment may differ depending on which one is used to measure outcomes. To provide better empirical justification for the selection of instruments, head-to-head comparisons of instruments within the same studies are needed. This information would greatly increase precision for justifying the use of a particular instrument for symptom management trials as well as other clinical studies.
|
REFERENCES |
|---|
|
TopAbstractValue of Quality-of-Life...References |
|---|
(1) Buchanan D, O'Mara A, Keaghan J, Minasian L. Quality-of-life assessment in the symptom management trials of the National Cancer Institute-supported Community Clinical Oncology Program. J Clin Oncol (2005) 23:591–8.
(2) Ferrans C. Development of a conceptual model of quality of life. Sch Inq Nurs Pract (1996) 10:293–304.[Medline]
(3) Ferrans CE. Definitions and conceptual models of quality of life. In: Outcomes assessment in cancer—Lipscomb J, Gotay CC, Snyder C, eds. (2005) Cambridge (UK): Cambridge University Press. 14–30.
(4) Wilson IB, Cleary PD. Linking clinical variables with health-related quality of life: a conceptual model of patient outcomes. J Am Med Assoc (1995) 273:59–65.
(5) Ferrans C, Zerwic J, Wilbur J, Larson J. Conceptual model of health-related quality of life. J Nurs Scholarsh (2005) 37:336–42.[CrossRef][Web of Science][Medline]
(6) Cella D, Tulsky D, Gray G, Sarafian B, Linn E, Bonomi A, et al. The Functional Assessment of Cancer Therapy (FACT) scale: development and validation of the general version. J Clin Oncol (1993) 11:570–9.
(7) Cella D, Bonomi A, Lloyd S, Tulsky DS, Kaplan E, Bonomi P. Reliability and validity of the Functional Assessment of Cancer Therapy-Lung (FACT-L) quality of life instrument. Lung Cancer (1995) 12:199–220.[CrossRef][Web of Science][Medline]
(8) Aaronson N, Ahmedzai S, Bergman B, Bullinger M, Cull A, Duez NJ, 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) 85:365–76.
(9) Aaronson NK, Cull AM, Kaasa S, Sprangers MAG. The European Organization for Research and Treatment of Cancer (EORTC) modular approach for quality of life assessment in oncology: an update. In: Quality of life and pharmacoeconomics in clinical trials—Spilker B, ed. (1996) 2nd ed. Philadelphia (PA): Lippincott-Raven. 179–89.
(10) Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). Med Care (1992) 30:473–83.[Web of Science][Medline]
(11) Stewart AL, Hays RD, Ware JE. The MOS short-form general health survey. Med Care (1988) 26:724–35.[Web of Science][Medline]
(12) Movsas B, Scott C, Langer C, Werner-Wasik M, Nicolaou N, Komaki R, et al. Randomized trial of amifostine in locally advanced non-small-cell lung cancer patients receiving chemotherapy and hyperfractionated radiation: Radiation Therapy Oncology Group Trial 98-01. J Clin Oncol (2005) 23:2145–54.
(13) Hahn EA, Glendenning GA, Sorensen MV, Hudgens SA, Druker BJ, Guilhot F, et al. Quality of life in patients with newly diagnosed chronic phase chronic myeloid leukemia on imatinib versus interferon alpha plus low-dose cytarabine: results from the IRIS study. J Clin Oncol (2003) 21:2138–46.
(14) Kornblith AB, Herndon JE, Silverman LR, Demakas EP, Odchimar-Reissig R, Holland JF, et al. Impact of azacytidine on the quality of life of patients with myelodysplastic syndrome treated in a randomized phase III trial: a Cancer and Leukemia Group B study. J Clin Oncol (2002) 20:2441–52.
(15) Jacobs J, Herman P, Heron K, Olsen S, Vaughters L. Homeopathy for menopausal symptoms in breast cancer survivors: a preliminary randomized controlled trial. J Altern Complement Med (2005) 11:21–7.[CrossRef][Web of Science][Medline]
(16) Spitzer W, Dobson A, Hall J, Chesterman E, Levi J, Shepherd R, et al. Measuring the quality of life of cancer patients: a concise QL-index for use by physicians. J Chronic Dis (1981) 34:585–97.[CrossRef][Web of Science][Medline]
(17) Covinsky KE, Wu AW, Landefeld S, Connors AF, Phillips RS, Tsevat J, et al. Health status versus quality of life in older patients: does the distinction matter? Am J Med (1999) 106:435–40.[CrossRef][Web of Science][Medline]
(18) Smith KW, Avid NE, Assmann SF. Distinguishing between quality of life and health status in quality of life research: a meta-analysis. Qual Life Res (1999) 8:447–59.[CrossRef][Web of Science][Medline]
(19) Osoba D, Brada M, Yung W, Prados M. Health-related quality of life in patients with anaplastic astrocytoma during treatment with temozolomide. Eur J Cancer (2000) 36:1788–95.[CrossRef][Web of Science][Medline]
(20) Shag C, Ganz PA, Heinrich R. Cancer Rehabilitation Evaluation System—Short Form (CARES-SF). Cancer (1991) 68:1406–13.[CrossRef][Web of Science][Medline]
(21) Cohen S, Mount B, Strobel M, Bui F. The McGill Quality of Life Questionnaire: a measure of quality of life appropriate for people with advanced disease. Palliat Med (1995) 9:207–19.
(22) Ferrans CE. Development of a quality of life index for patients with cancer. Oncol Nurs Forum (1990) 17:15–9.[Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. A. Ganz and P. J. Goodwin Health-Related Quality of Life Measurement in Symptom Management Trials J Natl Cancer Inst Monographs, October 1, 2007; 2007(37): 47 - 52. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||