Development of an item bank for the EORTC Role Functioning Computer Adaptive Test (EORTC RF-CAT). Issue 1 (December 2016)

• Record Type: Journal Article
• Authors: Gamper, Eva-Maria; Petersen, Morten; Aaronson, Neil; Costantini, Anna; Giesinger, Johannes; Holzner, Bernhard; Kemmler, Georg; Oberguggenberger, Anne; Singer, Susanne; Young, Teresa; Groenvold, Mogens
• Is Part Of: Health and quality of life outcomes. Volume 14:Issue 1(2016)
• Page Start: 1
• Page End: 11
• Publication Date: 2016-12
• Subjects: Cancer -- Oncology -- Health related quality of life -- Role functioning -- QLQ-C30 -- Computer-adaptive test; Outcome assessment (Medical care) -- Periodicals; Quality of life -- Periodicals; 362.1
• DOI: 10.1186/s12955-016-0475-x
• Languages: English
• ISSNs: 1477-7525

Abstract

Abstract Background Role functioning (RF) as a core construct of health-related quality of life (HRQOL) comprises aspects of occupational and social roles relevant for patients in all treatment phases as well as for survivors. The objective of the current study was to improve its assessment by developing a computer-adaptive test (CAT) for RF. This was part of a larger project whose objective is to develop a CAT version of the EORTC QLQ-C30 which is one of the most widely used HRQOL instruments in oncology. Methods In accordance with EORTC guidelines, the development of the RF-CAT comprised four phases. Phase I involved the conceptualization of RF. In Phase II, a provisional list of items was defined and revised by experts in the field. In phase III, feedback was obtained from cancer patients in various countries. Phase IV comprised field testing in an international sample, calibration of the item bank, and evaluation of the psychometric performance of the RF-CAT. Results Phases I-III yielded a list of 12 items eligible for phase IV field-testing. The field-testing sample included 1, 023 patients from Austria, Denmark, Italy, and the UK. Psychometric evaluation and item response theory analyses yielded 10 items with good psychometric properties. The resulting item bank exhibits excellent reliability (mean reliability = 0.85, median = 0.95). Using the RF-CAT may allow sample size savings from 11 % up to 50 % compared to using the QLQ-C30 RF scale. Conclusions The RF-CAT item bank improves the precision and efficiency with which RF can be assessed, promoting its integration into oncology research and clinical practice.