Psychometric testing of fatigue instruments for use with cancer patients

Background: Cancer treatment-related fatigue (CRF) is a common side effect of cancer treatment. A problem identified in most reviews of CRF is lack of sound approaches to measurement that are congruent with the conceptualizat ion of CRF as a self-perceived state. The diversity of instruments availabl e to measure fatigue and the lack of comprehensive testing of several promi sing instruments with cancer patients undergoing treatment provided the rat ionale for this study. The purpose of this article is to report the results of psychometric testing of several fatigue instruments in patients undergo ing cancer treatment. Objectives: The aims of this study were to determine the reliability, valid ity, and responsiveness of each instrument and to determine the ability of each instrument to capture CRF. Methods: Existing fatigue instruments with published psychometric informati on that indicated suitability for further testing were selected and include d the Profile of Mood States Short Form fatigue subscale (F_POMS-sf), Multi dimensional Assessment of Fatigue (MAF), Lee Fatigue Scale (LFS), and the M ultidimensional Fatigue Inventory (MFI). Data were collected at a universit y-based clinical cancer center and a freestanding comprehensive cancer cent er. Subjects completed all study instruments, which were presented in rando m order, at a time when CRF was expected to be high and again when it was e xpected to be low. A subset of subjects completed the instruments within 48 hours of one of the data collection points when CRF was expected to be rel atively unchanged to provide stability data. Results: Reliability estimates using Cronbach's alpha indicated that all in struments examined had good internal consistency. Test-retest correlations showed good stability for total scores on all the instruments, but some sub scales of the LFS and MFI had marginal stability. Factor analysis of all in struments indicated that only the LFS and the F_POMS-sf fully supported the ir construct validity. All of the instruments showed responsiveness to chan ges in CRF related to treatment. Conclusions: The results of the study provide researchers and clinicians wi th detailed comparisons of the performance of established fatigue measures in cancer patients undergoing treatment to use when selecting measures of C RF.