|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2002 American Society for Clinical Oncology The Minneapolis-Manchester Quality of Life Instrument: Reliability and Validity of the Adolescent FormByFrom the Division of Pediatric Oncology, City of Hope National Medical Center, Duarte, CA; Department of Child Health, Llandough Hospital, South Glamorgan, United Kingdom; Divisions of Health Services Research and Policy and Epidemiology and Clinical Research, University of Minnesota, Minneapolis, MN; Division of Pediatric Oncology, Children’s Hospital of Oakland, Oakland, CA; and Division of Pediatric Oncology, University of Washington, Seattle, WA. Address reprint requests to Smita Bhatia, MD, MPH, Children’s Oncology Group, PO Box 60012, Arcadia, CA 91006-6012; email: sbhatia{at}smtplink.coh.org
PURPOSE: With improvement in survival after childhood cancer, increasing emphasis is being placed on the impact of treatment and its sequelae on the health-related quality of life (HRQL) of survivors. The Minneapolis-Manchester Quality of Life Instrument (MMQL) is a standardized patient self-report instrument designed to assess HRQL in survivors of childhood cancer. The MMQL is being developed for three age groups to address the changing developmental need of different ages: MMQL-Youth (8 to 12 years), MMQL-Adolescent (13 to 20 years), and MMQL-Young Adult (21 to 45 years). This report focuses on the development and testing of the MMQL-Adolescent Form. PATIENTS AND METHODS: To validate the instrument, the MMQL-Adolescent Form was administered to 397 adolescents (129 healthy individuals, 110 patients with cancer undergoing therapy, and 158 subjects off therapy for cancer). Factor analysis was conducted to refine the instrument. Construct validity was conducted by comparing similar constructs in the MMQL-Adolescent Form and the Child Health Questionnaire (CHQ). Discriminate validity was determined by comparing healthy children with children with cancer either on or off therapy. Stability of the MMQL was tested by readministering the MMQL-Adolescent Form 2 weeks later. RESULTS: Internal consistency reliability was in the acceptable range for this instrument. The MMQL was able to discriminate between the three study populations. There were high correlations between the MMQL scales and similar CHQ domains. Test-retest reliability of the MMQL-Adolescent Form demonstrated that the instrument was extremely stable in all scales tested. CONCLUSION: Overall, the data provide evidence for the validity and reliability of the MMQL-Adolescent Form as a comprehensive, multidimensional self-report instrument for measuring HRQL among adolescent survivors of childhood cancer.
THE NUMBER OF CHILDREN surviving cancer has increased dramatically over the past 30 years.1 Increasing emphasis is now being placed on examining health status or health-related quality of life (HRQL) as a multidimensional, self-perceived construct, which includes aspects such as physical activity, psychological adjustment, social reintegration, and overall well-being. The World Health Organization’s definition of health is often used to define quality of life as a multidimensional construct, incorporating both objective and subjective data, including (but not limited to) the social, physical, and emotional functioning of the child and, when indicated, his/her family.2 The perception of health for a child or adolescent is different from adults, with health and illness defined as separate concepts by the younger age group rather than as a continuum.3 In addition, when developing an instrument, the differing developmental stages of a child and adolescent need to be addressed. Longitudinal assessments of HRQL are further complicated by the fact that children often cross developmental stages as they proceed through treatment and follow-up.4-6 Instruments need to be sensitive to these changes yet maintain constancy in the domains tapped. Another critical issue in HRQL with pediatric cancer patients is that of proxy respondent (parent or physician rating for a child considered too young to provide information). Studies have documented poor parent-child agreement with regard to emotional or subjective states,7-9 although concordance on functioning status, such as interference with daily activities, can be quite high.7 Therefore, although proxies may be needed with small children, they should be avoided with children old enough to speak for themselves. Among instruments measuring quality of life in the adult population, there are both generic (eg, SF-36)10 and cancer-specific (eg, Functional Assessment of Cancer Therapy,11 Functional Living Index,12 and the Cancer Rehabilitation Evaluation System13) scales. Unfortunately, instruments measuring quality of life in childhood and adolescence, and addressing all the concerns of cancer survivors, are scarce.14-24 The Minneapolis-Manchester Quality of Life (MMQL) instrument was developed to address the critical need for a practical and comprehensive, multidimensional, self-report instrument to assess HRQL in survivors of childhood cancer. Three versions of the MMQL address the changing developmental need of different ages: the MMQL-Youth Form (8 to 12 years; interview-based), the MMQL-Adolescent Form (13 to 20 years; self-administered), and the MMQL-Young Adult Form (21 to 45 years; self-administered). This report focuses on the development and testing of MMQL regarding its stability, internal consistency, and construct validity, for subjects between the ages of 13 and 20 years.
MMMQL: Instrument Development The MMQL-Adolescent Form, specific for adolescent survivors of childhood cancer, defines HRQL in terms of the impact of disease and treatment on an individual’s physical, social, psychologic, and cognitive functioning, and overall outlook on life. Higher scores on the MMQL indicate minimal negative impact and thus greater HRQL. The questionnaire was developed in three phases.
Phase I: Pretesting the MMQL-Adolescent Form Cronbach’s alpha25 was used to assess the internal consistency. Cronbach’s coefficient alpha for internal consistency reliability provides an indication of the degree of convergence between different items hypothesized to represent a unified construct. Items with a low corrected item-total correlation were not retained in the final questionnaire. Given that instrument development is an iterative process, more items were included in the instrument than were believed to be necessary, to ensure that the instrument contained a large enough item pool to sample sufficiently the domains being tested. All scales were constructed and scored using the Likert method of summated ratings, using a four- or five-point response scale.26
Phase II: Collection of Normative Data
Phase III: Reliability and Validity of the MMQL The seven scales of the MMQL-Adolescent Form consist of the following: (1) physical functioning (assesses functional status in the activities of daily living); (2) psychological functioning (encompasses emotional functioning including emotional distress and worry; (3) social functioning (assesses interpersonal relationships); (4) cognitive functioning (assesses disease- and treatment-related cognitive sequelae); (5) body image (assesses self-perception of body image and serves an indicator of self-esteem); (6) outlook on life (assesses satisfaction with current and future life situations); and (7) intimate relations (assesses interpersonal relationships, with a focus on intimacy). The stability of the instrument was tested by readministering the MMQL instrument to a subset of study participants after a 2-week interval (test-retest reliability). The validation strategy of the MMQL-Adolescent Form as a whole and of the scales of the MMQL consisted of several parts. The internal consistency of the MMQL and of each scale was assessed, thus supporting the instrument’s reliability and satisfying one of the major requirements for assessing validity.25 Discriminate validity of the MMQL instrument, and of each scale, was assessed by comparing subject groups differing in known health status and therefore known to differ in the underlying health construct being investigated.29,30 Three study populations were tested: a cohort of healthy participants, patients with cancer currently receiving therapy, and patients with cancer after completion of their therapy. It was expected that the MMQL instrument would distinguish between these three study populations, in that the healthy controls would have the highest MMQL scores and those off treatment would have higher scores than those on treatment. Construct validity was assessed using the Child Health Questionnaire-Child Form (CHQ-CF87).18 Construct validity measures the extent to which an HRQL measure relates to other measures in a manner consistent with theoretically driven hypotheses concerning the concepts (or constructs) being evaluated. The child-completed form CHQ-CF87 used here consists of 87 items. The concepts in the questionnaire are general health, physical functioning, bodily pain/discomfort, limitations in schoolwork or activities with friends because of physical problems or emotional/behavioral difficulties, behavior, mental health, self-esteem, limitations in family activities, and family cohesion. It was hypothesized that the MMQL scales would correlate substantially with the theoretically similar scales from the CHQ. Specific hypotheses for construct validity are listed in Table 1.
Procedure Nine institutions within the United States contributed patients for this study (see Appendix). The research protocol was approved by the institutional review boards at each of the participating institutions. Permission to contact patients and their parents was requested from the primary physicians. Informed consent and assent were obtained in person or by mail. In all cases, initial contact with the study subjects was made through the parents. For patients undergoing active treatment, consecutive patients were enrolled onto the study. For patients who were off therapy, a roster was made at each institution and a random sample drawn from it. Patients completed a self-administered questionnaire in the clinic or a questionnaire was mailed to them. The importance of self-report was emphasized in the instructions given to the study participants. To limit respondent burden, two separate arms of the study were created: validation and test-retest. Those placed on the test-retest arm were asked to complete the same questionnaire again, 2 weeks after completing it the first time. Those on the validation arm were asked to concomitantly complete the CHQ and the study questionnaire (MMQL).
Statistical Analysis Internal consistency reliability for all component scales was determined by calculating Cronbach’s coefficient alpha.25 Scales with reliabilities of more than 0.70 are recommended for the purpose of comparing groups of patients, whereas the more stringent reliability criterion of 0.90 is recommended for greater precision in analyzing individual patient scores.31 Cronbach’s internal consistency reliability coefficients were determined for the MMQL total scale and for the individual domains. Discriminate validity of the MMQL was determined by comparing the three study participant groups (normal, cancer–on therapy, and cancer–off therapy).32 Student’s t tests were conducted to determine which scales were able to differentiate between normative healthy controls and those with cancer (both on therapy and off therapy). Construct validity was assessed through correlation analyses between the MMQL-Adolescent Form and the CHQ. The correlational effect sizes are designated as small (0.10), medium (0.30), and large (0.50) on the basis of Cohen’s statistical guidelines.33
A total of 397 adolescents participated in this study. Table 2 lists the demographics of the study population. These included 129 subjects with no history of cancer, 110 subjects with cancer undergoing therapy at the time of the study, and 158 adolescents with cancer who were off therapy for 1 or more years. The median time off therapy was 6.3 years (range, 1.5 to 12.9 years). Patients with cancer were more likely to be older (median age at time of study, 16.6 years) than the healthy adolescent participants (median age, 15.8 years; P < .001). Participants with cancer were also more likely to be male (56%) when compared with those without cancer (43%, P = .04). Participants with cancer were less likely to be white (64%) when compared with those without cancer (85%). However, there was no difference in the distribution by race when the study population with cancer was compared to patients entered onto therapeutic studies by the Children’s Oncology Group. There was an overrepresentation of patients with acute leukemia in our study population, when compared with the distribution of cancer patients by diagnoses published by the Surveillance, Epidemiology, and End-Results program (47% v 25%).1
Internal Consistency Reliability Internal consistency reliability for all items ranged from 0.67 to 0.89 (overall alpha = 0.92). In addition, a weighted mean of the scales was also taken (alpha = 0.78). The results indicate that the items constituting the MMQL are consistent internally and appear to measure a unitary construct, as listed in Table 3.
Discriminate Validity Table 4 lists the results from tests of discriminate validity derived from comparing three groups of participants: those without cancer (normative controls), those with cancer and receiving therapy, and those with cancer and off therapy. Significant differences for study populations were demonstrated between those without cancer and with cancer, both on and off therapy (Table 4). Using the MMQL-Adolescent Form, normative controls had statistically significantly higher scores in the physical, cognitive, psychological, and social functioning domains. However, the MMQL-Adolescent Form could not discriminate between the three clinical groups in the body image, intimate relations, and outlook on life domains. The CHQ could successfully discriminate between the cancer patients on and off therapy for global health, physical functioning, bodily pain, self-esteem, general health perceptions, and family activities (data not shown).
Construct Validity The correlation between the MMQL-Adolescent Form and the CHQ scales is summarized in Table 5. As hypothesized (Table 1), the MMQL outlook on life scale correlated with the following CHQ scales: Global health, mental health, self-esteem, general health perceptions, and family activities. Again, as hypothesized, the MMQL physical functioning scale correlated with the following scales from the CHQ: global health, physical activities, and physical role limitations. Similarly, the MMQL psychological functioning scale correlated with the following CHQ scales: behavior, mental health, and general health perceptions. The MMQL cognitive functioning scale correlated well with the following CHQ scales: behavior, mental health, and self-esteem. The MMQL social functioning and body image scales correlated with the CHQ self-esteem scale. Finally, the MMQL intimate relationship scale correlated with the CHQ behavior and self-esteem scale. Several other significant correlations were observed between the MMQL and CHQ scales that were not hypothesized a priori, as shown in Table 5. However, the associations between these scales, when examined individually, can be explained. For example, physical functioning (MMQL) correlates significantly with the following CHQ scales: mental health, self-esteem, and general health perceptions, all of which could conceivably influence physical functioning.
Stability/Test-Retest Reliability As shown in Table 6, the MMQL-Adolescent Form demonstrated stability for all the scales examined when the instrument was readministered after 2 or more weeks.
The physical and psychologic consequences of childhood cancer and its treatment are likely to have an impact on the HRQL of cancer survivors. The importance of assessing HRQL in this population, with the ultimate goal of minimizing the negative consequences of childhood cancer and its treatment on survivors, is now recognized widely. Recently, both generic and disease-specific measures have been developed for children and adolescents that are multidimensional, and incorporate self-report. These include the RAND Health Insurance Experiment Child Form16 (proxy-administered for infants to adolescents and covering the following domains: physical, mental, social, and general health); Child Health and Illness Profile (covers 11- to 17-year-olds, is a self-report, and covers the following domains: activity, comfort, satisfaction, disorders, achievement, and resilience, and is being validated and refined)17; and the CHQ.18 The CHQ is available in parent- and child-completed versions, and the parent form (CHQ-PF50) has been validated on a representative sample from the general population. It is a generic assessment tool developed for children and adolescents. The Multiattribute Health Status Classification System19-21 (Health Utilities Index [HUI] Marks 2 and 3 [HUI2, 3]) includes a set of well-defined, generic, multiattribute health-system classification schemes that are compatible with multiattribute preference functions, allowing the computation of a single summary score of HRQL. The HUIs use 10 attributes to assess health status: sensation, vision, hearing, speech, ambulation/mobility, dexterity, emotion, cognition, self-care, pain, and fertility. The focus of the HUIs is on capacity rather than performance. Social interaction is not assessed directly in the HUIs. The limitation of the generic measures is that they may not adequately cover specific cancer concerns. Disease-specific measures have been developed for pediatric cancer patients and have attempted to include items that are more responsive to the impact of cancer and its treatment on these patients.22-24 The Pediatric Oncology Quality of Life Scale22 is a short measure (21 items) and is intended to be used by the parents of children from preschool age to adolescence. It covers three domains (physical functioning, emotional distress, and active medical treatment). The Perception of Illness Experience scale23 (34 items) can be used by the parents or patients (8 to 24 years of age), and the domains covered include physical appearance, interference with physical activity, integration in school, peer rejection, parental behavior, manipulation, and preoccupation with illness. The Perception of Illness Experience scale, although multidimensional, does not specifically tap emotional functioning or physical symptoms, two key domains that may be affected in this patient population. The Pediatric Quality of Life Inventory (PedsQL) is a modular instrument designed to measure HRQL in children and adolescents aged 2 to 18 years. The PedsQL 4.0 Generic Core Scales are multidimensional child self-report and parent proxy-report scales (physical, emotional, social, and school functioning) developed as generic core measure to be integrated with the PedsQL disease-specific modules. The PedsQL 3.0 Cancer Module is designed to measure pediatric cancer-specific HRQL, encompassing eight scales, including pain and hurt, nausea, procedural anxiety, treatment anxiety, worry, cognitive problems, perceived physical appearance, and communication.24 The existing pediatric cancer HRQL measures therefore have their strengths and limitations, the most obvious weakness being the scarcity of a pediatric patient self-report instrument that measures HRQL among cancer survivors. The purpose of this study was to present internal consistency reliability, discriminate validity, construct validity, and stability data for a new instrument (MMQL) developed to measure self-report of HRQL in survivors of childhood cancer. The study describes the necessary steps in the development of a comprehensive, disease-specific, self-report HRQL instrument.34 Overall, the data provide evidence for the validity and reliability of the MMQL-Adolescent Form as a comprehensive, multidimensional self-report instrument for measuring HRQL among adolescent survivors of childhood cancer. In addition, this instrument has been tested in a random sample of healthy children with no chronic illnesses, and thus provides normative data that can be used in future studies for comparing HRQL in various populations. High alpha coefficients for the MMQL-Adolescent Form as a whole (and for many of the individual scales) and high stability coefficients support the instrument’s reliability. The MMQL performed as expected in its ability to discriminate between the various clinical groups. The MMQL-Adolescent Form was able to discriminate between those with and without cancer in the physical functioning, cognitive functioning, psychological functioning, and social functioning domains. These findings are similar to reports on validity testing of other instruments such as CHQ, where disease-related and physical functioning scales, but not emotional scales, were sensitive to the presence or absence of disease.24,35,36 Application of the MMQL-Adolescent Form to a variety of cancer populations will help to further define the discriminatory properties of this instrument compared with other quality-of-life instruments. Overall, the pattern of correlations between the MMQL scales and the CHQ scales in the current study were consistent with the hypotheses and provide support for construct validity for the MMQL scales. As predicted, the correlations for physical, cognitive and psychological functioning, outlook on life, and body image were strong. The correlation for intimate relations was moderately strong. Moreover, the results indicated that the instrument was extremely stable for all the scales tested. Several limitations to our report need to be discussed. Longitudinal studies are needed to address the sensitivity of this instrument to small changes in HRQL, an issue that could not be addressed in this study. The study participants included patients with various cancer diagnoses, and thus were a heterogeneous group with potentially differing health concerns both on and off therapy. The results may be an underestimate of some of the issues because of the heterogeneity of these cancer groups. Future studies will need to address the utility of this measure among various cancer diagnoses. Nonetheless, the measure was able to identify differences between groups. Extensive review of the existing literature reveals that there continues to be a critical need for a valid instrument that would measure HRQL among survivors of childhood cancer.37-40 Using a thorough and systematic approach, we have created a valid and reliable self-report tool specific to childhood cancer survivors and obtained normative data from a healthy population. We are currently in the process of developing age-appropriate versions of this tool to address the differing developmental needs of younger individuals (8- to 12-year-olds) and older (21- to 45-year-olds) survivors of childhood cancer. Survival will always remain the ultimate goal in the management of children with cancer. Nevertheless, it is essential to assess the impact of cancer and its treatment on the HRQL of the child, and to incorporate such an assessment and appropriate intervention into the long-term management of a child with cancer. We hope that having a self-report cancer-specific instrument such as the MMQL will help in the decision-making process regarding selection of therapy, appropriate follow-up, and interventions where needed to prevent or manage sequelae that could negatively impact on a survivor’s HRQL.
APPENDIX
Supported in part by the Children’s Cancer Group, Chair’s Discretionary Research Funds. We thank Gwen Uman, PhD, for her assistance with statistics.
1. Ries LAG, Kosary CL, Hankey BF, et al (eds): SEER Cancer Statistics Review: 1996. Bethesda, MD, National Cancer Institute, 1999 2. World Health Organization: The First Ten Years of the World Health Organization. Geneva, World Health Organization, 1958 3. Bradlyn AS, Ritchey AK, Harris CV, et al: Quality of life research in pediatric oncology: Research methods and barriers. Cancer 78: 1333-1339, 1996[CrossRef][Medline] 4. Bradlyn AS, Harris CV, Spieth LA: Quality of life assessment in pediatric oncology: A retrospective review of phase III reports. Soc Sci Med 41: 1463-1465, 1995[CrossRef][Medline] 5. Bradlyn AS, Pollack BH: Quality-of-life research in Pediatric Oncology Group: 1991-1995. J Natl Cancer Inst Monogr 20: 49-53, 1996 6. Jenney MEM, Kane RL, Lurie N: Developing a measure of health outcomes in survivors of childhood cancer: A review of the issues. Med Pediatr Oncol 24: 145-153, 1995[Medline] 7. Pantell RH, Lewis CC: Measuring the impact of medical care on children. J Chronic Dis 40: 99S-115S, 1987 (suppl 1) 8. Weissman MM, Orvaschel H, Padian N: Children’s symptom and social functioning self-report scales: Comparison of mothers’ and children’s reports. J Nerv Ment Dis 168: 736-740, 1980[CrossRef][Medline] 9. Achenbach TM, McConaughy SH, Howell CT: Child adolescent behavioral and emotional problems: Implications of cross-informant correlations for situational specificity. Psychol Bull 101: 213-232, 1987[CrossRef][Medline] 10. Stewart AL, Hays RD, Ware JE: The MOS short-form general health survey: Reliability and validity in a patient population. Med Care 26: 724-735, 1988[Medline] 11. Weitner MA, Meyers CA, Gelke C, et al: The Functional Assessment of Cancer Therapy (FACT) scale: Development of a brain scale and revalidation of the general version (FACT-G) in patients with primary brain tumors. Cancer 75: 151-161, 1995 12. Morrow GR, Lindke P, Black P: Measurement of quality of life in patients: Psychometric analyses of the Functional Living Index–Cancer (FLIC). Qual Life Res 1: 287-296, 1992[CrossRef][Medline] 13. Schag CC, Heinrich RL, Aadland R, et al: Assessing problems of cancer patients: Psychometric properties of the Cancer Inventory of Problem Situations. Health Psychol 9: 83-102, 1990[CrossRef][Medline] 14. Achenbach TM, Edelbrock CS: Manual for the Child Behavior Checklist and the Revised Child Behavior Profile. Burlington, VT, University of Vermont, 1983 15. Lansky SB, List MA, Lansky LL, et al: The measurement of performance in childhood cancer patients. Cancer 60: 1651-1656, 1987[CrossRef][Medline] 16. Eisen M, Donald CA, Ware JE, et al: Conceptualization and measurement of health for children in the health insurance study. Santa Monica, CA, RAND Publications, 1980, No. R-2313 HEW
17. Starfield B, Bergner M, Ensminger M, et al: Adolescent health status measurement: Development of the Child Health and Illness Profile. Pediatrics 91: 430-435, 1993 18. Landgraf JM, Abetz L, Ware JE: The Child Health Questionnaire: A User’s Manual. Boston, MA, The Health Institute, 1996 19. Feeny DF, Barr RD, Torrance GW, et al: A comprehensive multiattribute system for classifying the health status of survivors of childhood cancer. J Clin Oncol 10: 923-928, 1992[Abstract] 20. Billson AL, Walker DA: Application of a modified multiattribute system in the assessment of health status in survivors of cancer. Am J Pediatr Hematol Oncol 15: 284-290, 1993[Medline] 21. Barr RD, Furlong W, Dawson S, et al: An assessment of global health status in survivors of acute lymphoblastic leukemia in childhood. Am J Pediatr Hematol Oncol 15: 284-290, 1993[Medline] 22. Goodwin DAJ, Boggs SR, Graham-Pole J: Development and validation of the Pediatric Oncology and Quality of Life Scale. Psychol Assess 6: 321-328, 1994[CrossRef] 23. Eiser C, Havermanns T, Craft A, et al: Development of a measure to assess the perceived illness experience after treatment for cancer. Arch Dis Child 72: 302-307, 1995[Abstract] 24. Varni JW, Burwinkle TM, Katz ER, et al: The PedsQLTM in pediatric cancer: Reliability and validity of the Pediatric Quality of Life Inventory generic core scales—Multidimensional Fatigue Scale, and Cancer Module. Cancer 94: 2090-2106, 2002[CrossRef][Medline] 25. Cronbach LJ: Coefficient alpha and the internal structure of tests. Psychometrika 16: 297-334, 1951[CrossRef] 26. Likert R: A technique for the measurement of attitudes. Arch Psychol 140: 5, 1932
27. Robison LL, Daigle A: Control selection using random digit dialing for cases of childhood cancer. Am J Epidemiol 120: 164-166, 1984 28. Lindeman RH, Merenda PF, Gold RZ: Introduction to Bivariate and Multivariate Analysis. Glenview, IL, Scott Foresman & Co, 1980, pp 245-282 29. Osoba D: Lessons learned from measuring health-related quality of life in oncology. J Clin Oncol 12: 608-616, 1994[Abstract]
30. Testa MA, Simonson DC: Assessment of quality-of-life outcomes. N Engl J Med 334: 835-840, 1996 31. Nunnally JC, Bernstein IR: Psychometric Theory, ed 3 . New York, NY, McGraw-Hill, 1994 32. McHorney CA, Ware JE, Raczek AE: The MOS 36-Item Short-Form Health Survey (SF-36): II. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care 31: 247-263, 1993[Medline] 33. Cohen J: Statistical Power Analysis for the Behavioral Sciences, ed 2 . Hillsdale, NJ, Erlbaum, 1988 34. Juniper EF, Guyatt GH, Jaeschke R: How to develop and validate a new health-related quality of life instrument, in Spilker B (ed): Quality of Life and Pharmacoeconomics in Clinical Trials. Philadelphia, PA, Lippincott-Raven, 1996, pp 49-56
35. 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 11: 570-579, 1993 36. King MT: The interpretation of scores from the EORTC quality of life questionnaire QLQ-C30. Qual Life Res 5: 555-567, 1996[CrossRef][Medline] 37. Feeny D, Furlong W, Mulhern RK, et al: A framework for assessing health-related quality of life among children with cancer. Int J Cancer Suppl 12: 2-9, 1999[CrossRef][Medline] 38. Monaco GA: Commentary on assessing health-related quality of life in children with cancer. Int J Cancer Suppl 12: 10, 1999[Medline] 39. Guyatt GH: Measuring health-related quality of life in childhood cancer: Lessons from the workshop. Int J Cancer Suppl 12: 143-146, 1999[CrossRef][Medline]
40. Eiser C, Morse R: A review of measures of quality of life for children with chronic illness. Arch Dis Child 84: 205-211, 2001 Submitted May 23, 2001; accepted August 26, 2002. This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2002 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
0.40 were retained in the initial item reduction. This iterative process resulted in the elimination of 47 items from the MMQL-Adolescent Form (on the basis of principal axis factoring with varimax rotation).
