Mirjam A. G. Sprangers

Methodological approaches for assessing response shift in longitudinal health-related quality-of-life research.

The impact of health state changes on an individuals quality of life (QOL) has gained increased attention in social and medical clinical research. An emerging construct of relevance to this line of investigation is response shift phenomenon. This construct refers to the changes in internal standards, in values, or in the conceptualization of QOL which are catalyzed by health state changes. In an effort to stimulate research on response shift, we present methodological considerations and promising assessment approaches for measuring it in observational and interventional clinical research. We describe and evaluate individualized methods, preference-based methods, successive comparison methods, design approaches, statistical approaches and qualitative approaches. The hierarchical structure of the construct is also discussed, with particular emphasis on how it might be elucidated by empirical assessment which uses the proposed methods and approaches. It is also recommended that criterion measures of change be included in future studies of response shift.

The European Organization for Research and Treatment of Cancer breast cancer-specific quality-of-life questionnaire module: first results from a three-country field study.

PURPOSE To construct a breast cancer-specific quality-of-life questionnaire (QLQ) module to be used in conjunction with the European Organization for Research and Treatment of Cancer (EORTC) QLQ-C30 and to test its reliability and validity cross-culturally. PATIENTS AND METHODS Module construction took place after the EORTC guidelines for module development. The module--the QLQ-BR23--consists of 23 items covering symptoms and side effects related to different treatment modalities, body image, sexuality, and future perspective. This module was tested in 170 Dutch, 168 Spanish, and 158 American cancer patients at two points in time. The timing for the Dutch and Spanish patients was before and during treatment with radiotherapy or chemotherapy. For the American patients, the questionnaire was administered at admission at the breast clinic and 3 months after the first assessment. RESULTS Multitrait scaling analysis confirmed the hypothesized structure of four of the five scales. Cronbachs alpha coefficients were, in general, lowest in Spain (range; .46 to .94) and highest in the United States (range; .70 to .91). On the basis of known-groups comparisons, selective scales distinguished clearly between patients differing in disease stage, previous surgery, performance status, and treatment modality, according to expectation. Additionally, selective scales detected change over time as a function of changes in performance status and treatment-induced change. CONCLUSION These results lend support to the clinical and cross-cultural validity of the QLQ-BR23 as a supplementary questionnaire for assessing specific quality-of-life issues relevant to patients with breast cancer.

Which chronic conditions are associated with better or poorer quality of life

The objective of the present study is to compare the QL of a wide range of chronic disease patients. Secondary analysis of eight existing data sets, including over 15,000 patients, was performed. The studies were conducted between 1993 and 1996 and included population-based samples, referred samples, consecutive samples, and/or consecutive samples. The SF-36 or SF-24 were employed as generic QL instruments. Patients who were older, female, had a low level of education, were not living with a partner, and had at least one comorbid condition, in general, reported the poorest level of QL. On the basis of rank ordering across the QL dimensions, three broad categories could be distinguished. Urogenital conditions, hearing impairments, psychiatric disorders, and dermatologic conditions were found to result in relatively favorable functioning. A group of disease clusters assuming an intermediate position encompassed cardiovascular conditions, cancer, endocrinologic conditions, visual impairments, and chronic respiratory diseases. Gastrointestinal conditions, cerebrovascular/neurologic conditions, renal diseases, and musculoskeletal conditions led to the most adverse sequelae. This categorization reflects the combined result of the diseases and comorbid conditions. If these results are replicated and validated in future studies, they can be considered in addition to information on the prevalence of the diseases, potential benefits of care, and current disease-specific expenditures. This combined information will help to better plan and allocate resources for research, training, and health care.

Is a single-item visual analogue scale as valid, reliable and responsive as multi-item scales in measuring quality of life?

Purpose: To compare the validity, reliability and responsiveness of a single, global quality of life question to multi-item scales. Method: Data were obtained from 83 consecutive patients with oesophageal adenocarcinoma undergoing either transhiatal or transthoracic oesophagectomy. Quality of life was measured at baseline, 5 weeks, 3 and 12 months post-operatively with a single-item Visual Analogue Scale (VAS) ranging from 0 to 100, the multi-item Medical Outcomes Study Short Form-20 (MOS SF-20) and Rotterdam Symptom Check-List (RSCL). Convergent and discriminant validity, test–retest reliability and both distribution-based and anchor-based responsiveness were evaluated. Major findings: At baseline and at 5 weeks, the VAS showed high correlations with the MOS SF-20 health perceptions scale (r = 0.70 and 0.72) and moderate to high correlations with all other subscales of the MOS SF-20 and RSCL (r = 0.29–0.70). The test–retest reliability intra-class correlation for the VAS was 0.87. At 5 weeks post-operatively, the distribution-based responsiveness was moderate for the VAS (standardised response mean: −0.47; effect size: −0.56), high for the physical subscales of the MOS SF-20 and RSCL (−1.08 to −1.51) and low for the psychological subscales (0.11 to −0.25). Five weeks post-operatively, anchor-based responsiveness was highest for the VAS (r = 0.54). Conclusion: The VAS is an instrument with good validity, excellent reliability, moderate distribution-based responsiveness and good anchor-based responsiveness compared to multi-item questionnaires. Its use is recommended in clinical trials to assess global quality of life.

The construction and testing of the EORTC colorectal cancer-specific quality of life questionnaire module (QLQ-CR38)

The objectives of the current study were to construct a colorectal cancer-specific quality of life (QL) questionnaire module to be used in conjunction with the European Organization for Research and Treatment of Cancer (EORTC) QLQ-C30 and to test its reliability and validity in The Netherlands. Module construction took place following the EORTC guidelines for module development. The module--the QLQ-CR38--consists of 38 items covering symptoms and side-effects related to different treatment modalities, body image, sexuality and future perspective. This module was tested among 117 colorectal cancer patients on several occasions. The timing was prior to treatment with radiotherapy or chemotherapy, during treatment and 3 months following the second assessment. For purposes of test-retest reliability, a subsample of patients completed the QLQ-CR38 1 week following the third assessment. Multitrait scaling analysis confirmed the hypothesised scale structure of the function scales but not of the symptom scales. Cronbachs alpha coefficients for seven of the nine scales exceeded the 0.70 criterion at one or both assessments. The test-retest reliability for all scales and one single item was 0.78 or higher. The stability of the two remaining single items was lower. On the basis of known-groups comparisons, selective scales distinguished clearly between patients differing in disease stage, initial and on-treatment performance status and the presence of a stoma. Additionally, selective scales detected change over time as a function of change in performance status and treatment-induced change. These results lend support to the clinical validity of the QLQ-CR38 as a supplementary questionnaire for assessing specific QL issues relevant to patients with colorectal cancer. Additional efforts to test the modules cross-cultural validity are needed.

The European Organization for Research and treatment of cancer approach to quality of life assessment: guidelines for developing questionnaire modules

The EORTC Study Group on Quality of Life (the Study Group) has adopted a modular approach to quality of life (QOL) assessment in cancer clinical trials. A ‘core’ instrument—the QLQ-C30—has been designed to cover a range of QOL issues relevant to a broad spectrum of cancer patients. It is intended that this QLQ-C30 be supplemented by more specific subscales (‘modules’) to assess aspects of QOL of particular importance to specific subgroups of patients. Since individual members of the Study Group were to be involved in module development, guidelines were required to standardize the module development process in order to ensure uniformly high quality across modules. These guidelines are presented in this paper. The term ‘module’ is defined, the composition of modules is outlined, and the criteria used to develop modules are specified. The module development process, consisting of four phases (generation of relevant QOL issues, operationalization of the QOL issues into a set of items, pretesting the module questionnaire, and large-scale field-testing) is described in detail. Further, issues related to cross-cultural instrument development, and the need for monitoring the module development process from within the Study Group are discussed. Finally, experiences with developing two site-specific modules (i.e., for head and neck, and breast cancer), are presented and the extent to which the guidelines meet practical requirements is discussed. The guidelines appear to provide a practical tool for module construction, that can facilitate the development of a comprehensive system for assessing the QOL of cancer patients internationally.

Colonic stenting versus emergency surgery for acute left-sided malignant colonic obstruction: a multicentre randomised trial

BACKGROUND Colonic stenting as a bridge to elective surgery is an alternative for emergency surgery in patients with acute malignant colonic obstruction, but its benefits are uncertain. We aimed to establish whether colonic stenting has better health outcomes than does emergency surgery. METHODS Patients with acute obstructive left-sided colorectal cancer were enrolled from 25 hospitals in the Netherlands and randomly assigned (1:1 ratio) to receive colonic stenting as a bridge to elective surgery or emergency surgery. The randomisation sequence was computer generated with permuted blocks and was stratified by centre; treatment allocation was concealed by use of a web-based application. Investigators and patients were unmasked to treatment assignment. The primary outcome was mean global health status during a 6-month follow-up, which was assessed with the QL2 subscale of the European Organisation for Research and Treatment of Cancer quality-of-life questionnaire (EORTC QLQ-C30). Analysis was by intention to treat. This study is registered, number ISRCTN46462267. FINDINGS Between March 9, 2007, and Aug 27, 2009, 98 patients were assigned to receive colonic stenting (n=47 patients) or emergency surgery (n=51). Two successive interim analyses showed increased 30-day morbidity in the colonic stenting group, with an absolute risk increase of 0.19 (95% CI -0.06 to 0.41) in analysis of the first 60 patients (14 of 28 patients receiving colonic stenting vs 10 of 32 receiving emergency surgery), and an absolute risk increase of 0.19 (-0.01 to 0.37) in analysis of the first 90 patients (23 of 47 patients vs 13 of 43). In accordance with the advice of the data safety monitoring committee, the study was suspended on Sept 18, 2009, and ended on March 12, 2010. At the final analysis of 98 patients, mean global health status during follow-up was 63.0 (SD 23.8) in the colonic stenting group and 61.4 (SD 21.9) in the emergency surgery group; after adjustment for baseline values, mean global health status did not differ between treatment groups (-4.7, 95% CI -14.8 to 5.5, p=0.36). No difference was recorded between treatment groups in 30-day mortality (absolute risk difference -0.01, 95% CI -0.14 to 0.12, p=0.89), overall mortality (-0.02, -0.17 to 0.14, p=0.84), morbidity (-0.08, -0.27 to 0.11, p=0.43), and stoma rates at latest follow-up (0.09, -0.10 to 0.27, p=0.35). However, the emergency surgery group had an increased stoma rate directly after initial intervention (0.23, 0.04 to 0.40, p=0.016) and a reduced frequency of stoma-related problems (between-group difference -12.0, -23.7 to -0.2, p=0.046). The most common serious adverse events were abscess (three in the colonic stenting group vs four in the emergency surgery group), perforations (six vs none), and anastomotic leakage (five vs one), and the most common adverse events were pneumonia (three vs one) and wound infection (one vs three). INTERPRETATION Colonic stenting has no decisive clinical advantages to emergency surgery. It could be used as an alternative treatment in as yet undefined subsets of patients, although with caution because of concerns about tumour spread caused by perforations. FUNDING None.

Cancer, fatigue and the return of patients to work—a prospective cohort study

Fatigue is a highly prevalent and debilitating symptom in cancer survivors. The aim of this study was to assess the impact of fatigue and other cancer-related symptoms on the return to work of cancer survivors. A prospective inception cohort study with 12 months of follow-up was initiated. At 6 months following the first day of sick leave, levels of fatigue, depression, sleep problems, physical complaints, cognitive dysfunction and psychological distress were assessed, in addition to clinical, sociodemographic and work-related factors. Data were obtained from one academic hospital and two general hospitals in the Netherlands. 235 patients who had a primary diagnosis of cancer and underwent treatment with curative intent were included. The rate of return to work was measured at 6, 12 and 18 months. Hazard ratios (HRs) for the duration of sick leave up to 18 months following the first day of sick leave were calculated. The rate of return to work increased from 24% at 6 months to 64% at 18 months following the first day of sick leave. Fatigue, diagnosis, treatment type, age, gender, depression, physical complaints and workload were all related to the time taken to return to work. Fatigue scores were also strongly related to diagnosis, physical complaints, and depression scores. Fatigue at 6 months predicted a longer sick leave with a hazard ratio of 0.71 (95% Confidence Interval (C.I.) 0.59-0.85), adjusted for diagnosis, treatment type, age and gender. In a multivariate Cox regression analysis, diagnosis, treatment, age, physical complaints and workload remained the only significant predictors of duration of the sick leave. 64% of cancer survivors returned to work within 18 months. Fatigue levels predicted the return to work. This was independent of the diagnosis and treatment, but not of other cancer-related symptoms. Better management of cancer-related symptoms is therefore needed to facilitate the return to work of cancer patients.

The European organization for research and treatment of cancer approach to developing questionnaire modules: an update and overview

The European Organization for Research and Treatment of Cancer (EORTC) Quality of Life Study Group has adopted a modular approach to quality of life (QoL) assessment in cancer clinical trials. The core instrument (the EORTC QLQ-C30) covers a range of QoL issues relevant to a broad spectrum of patients with cancer. The QLQ-C30 is designed to be supplemented by more specific subscales (‘modules’) to assess aspects of QoL of particular importance to specific subgroups of patients. Since individual members of the study group were to be involved in module development, guidelines were established. The primary aim of these guidelines was to standardize the module development process in order to ensure uniformly high quality across modules. This paper gives an update of the work completed to date. First, while the guidelines proved practical for module development, producing modules that exhibit adequate levels of psychometric and cross-cultural validity, experience pointed to three areas where the guidelines required more precision. These amendments will be provided and include (1) stricter monitoring of the developmental process from within the study group, (2) the explicit requirement of involvement of the study group and (3) a more precise definition of the criteria to be fulfilled before modules are allowed to be called ‘EORTC modules’. Second, an overview of the modules currently under development or available for general use is provided. These modules include those for body image, high-dose chemotherapy, leukaemia, myeloma, palliative care and the following cancers: bladder, brain, breast, colorectal, head and neck, lung, oesophageal, ophthalmic, ovarian, pancreas and prostate. Finally, the need for the coordination of efforts in module development, both from within and outside the EORTC, is discussed.

Adaptation to changing health. Response shift in quality-of-life research

A discussion of the impact that health changes have on ones perceived quality of life. It discusses the theoretical underpinnings of the response-shift construct, methodology for assessment in primary and secondary analyses, and application to treatment outcomes research.