Adam B Smith1, Kim Cocks2, Matthew Taylor3, David Parry4. 1. York Health Economics Consortium, University of York, Market Square, York YO10 5NH, UK; Research Innovation Office, University of York, York YO10 5DG, UK. Electronic address: adam.smith@york.ac.uk. 2. Department of Health Sciences, York Trials Unit, University of York, York YO10 5DD, UK. 3. York Health Economics Consortium, University of York, Market Square, York YO10 5NH, UK. 4. AstraZeneca Pharmaceuticals Ltd, Alderley Park, Macclesfield SK10 4TF, UK.
Abstract
OBJECTIVES: The study's aim was to assess the internal reliability for the nine domains of the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-C30) to evaluate homogeneity across clinical studies and whether sample characteristics predict coefficient heterogeneity. STUDY DESIGN AND SETTING: A systematic literature review was undertaken. Internal reliability was assessed against Cronbach α coefficient >0.70. Reliability generalization was undertaken using fixed- and random-effects models. A weighted least squares regression model was applied to determine whether baseline sample characteristics (language, percentage of women, sample size, sample means and standard deviations, and cancer type) predicted variation in α coefficients. RESULTS: A total of 33 studies were identified. Eight domains demonstrated good internal reliability (unweighted/weighted by sample variance). One domain, Cognitive Functioning, consistently performed poorly. In terms of moderating variables, none of the sample characteristic variables explained sample variance for the Physical or Role Functioning domains. For the other domains, language, percentage of women, and sample means and variances accounted for some of the heterogeneity observed. CONCLUSION: Most domains on the EORTC QLQ-C30 are reliable and may therefore be used to help inform decision-making processes, such as those involving individual patients.
OBJECTIVES: The study's aim was to assess the internal reliability for the nine domains of the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-C30) to evaluate homogeneity across clinical studies and whether sample characteristics predict coefficient heterogeneity. STUDY DESIGN AND SETTING: A systematic literature review was undertaken. Internal reliability was assessed against Cronbach α coefficient >0.70. Reliability generalization was undertaken using fixed- and random-effects models. A weighted least squares regression model was applied to determine whether baseline sample characteristics (language, percentage of women, sample size, sample means and standard deviations, and cancer type) predicted variation in α coefficients. RESULTS: A total of 33 studies were identified. Eight domains demonstrated good internal reliability (unweighted/weighted by sample variance). One domain, Cognitive Functioning, consistently performed poorly. In terms of moderating variables, none of the sample characteristic variables explained sample variance for the Physical or Role Functioning domains. For the other domains, language, percentage of women, and sample means and variances accounted for some of the heterogeneity observed. CONCLUSION: Most domains on the EORTC QLQ-C30 are reliable and may therefore be used to help inform decision-making processes, such as those involving individual patients.