INTRODUCTION: Fatigue is one of the most important symptoms reported by patients with systemic lupus erythematosus (SLE) and a key concept of interest in SLE clinical trials. Despite this, fatigue remains poorly understood and sub-optimally measured by existing patient-reported outcome (PRO) instruments and scales. Here, we psychometrically evaluated the measurement properties of three PRO scales that purport to measure fatigue, using data from two SLE clinical trials. METHODS: Data were pooled from two completed phase 3 SLE trials: EMBODY1 (NCT01262365) and EMBODY2 (NCT01261793). FACIT-F, SF-36 Vitality and LupusQoL Fatigue data were selected for post hoc Rasch Measurement Theory psychometric analysis in two stages: (1) scale-to-sample targeting, thresholds for item response options, item fit statistics, and reliability; and (2) proposal and evaluation of pooled fatigue items based on the best-performing items. Responsiveness analyses on group-level (two effect size [ES] calculations and relative efficiency) and individual level (within person statistically significant difference), were conducted to compare original scales and pooled item sets. RESULTS: Scale-to-sample targeting was good for FACIT-F, but suboptimal for SF-36 Vitality and LupusQoL Fatigue. Thresholds for item response options were ordered for all three scales. Item misfit was found in all three scales (FACIT-F 10/13; SF-36 Vitality 4/4; LupusQoL Fatigue 1/4). Reliability statistics were good for FACIT-F (0.93) and LupusQoL Fatigue (0.80) but low for SF-36 Vitality (0.53). The pooled fatigue items improved some psychometric properties despite persisting misfit issues (2/10) and were more sensitive in detecting change at week 24 compared with un-pooled data (ES 0.41 vs. 0.26-0.25). CONCLUSIONS: FACIT-F, SF-36 Vitality, and LupusQoL Fatigue were found to have important limitations in the EMBODY1 and EMBODY2 SLE clinical trials. Findings from pooled fatigue items support the need for further research to improve conceptual underpinnings of fatigue PROs and make them fit for purpose for drug development.
INTRODUCTION:Fatigue is one of the most important symptoms reported by patients with systemic lupus erythematosus (SLE) and a key concept of interest in SLE clinical trials. Despite this, fatigue remains poorly understood and sub-optimally measured by existing patient-reported outcome (PRO) instruments and scales. Here, we psychometrically evaluated the measurement properties of three PRO scales that purport to measure fatigue, using data from two SLE clinical trials. METHODS: Data were pooled from two completed phase 3 SLE trials: EMBODY1 (NCT01262365) and EMBODY2 (NCT01261793). FACIT-F, SF-36 Vitality and LupusQoL Fatigue data were selected for post hoc Rasch Measurement Theory psychometric analysis in two stages: (1) scale-to-sample targeting, thresholds for item response options, item fit statistics, and reliability; and (2) proposal and evaluation of pooled fatigue items based on the best-performing items. Responsiveness analyses on group-level (two effect size [ES] calculations and relative efficiency) and individual level (within person statistically significant difference), were conducted to compare original scales and pooled item sets. RESULTS: Scale-to-sample targeting was good for FACIT-F, but suboptimal for SF-36 Vitality and LupusQoL Fatigue. Thresholds for item response options were ordered for all three scales. Item misfit was found in all three scales (FACIT-F 10/13; SF-36 Vitality 4/4; LupusQoL Fatigue 1/4). Reliability statistics were good for FACIT-F (0.93) and LupusQoL Fatigue (0.80) but low for SF-36 Vitality (0.53). The pooled fatigue items improved some psychometric properties despite persisting misfit issues (2/10) and were more sensitive in detecting change at week 24 compared with un-pooled data (ES 0.41 vs. 0.26-0.25). CONCLUSIONS: FACIT-F, SF-36 Vitality, and LupusQoL Fatigue were found to have important limitations in the EMBODY1 and EMBODY2 SLE clinical trials. Findings from pooled fatigue items support the need for further research to improve conceptual underpinnings of fatigue PROs and make them fit for purpose for drug development.
Authors: Ewan C Goligher; Jacques Pouchot; Rollin Brant; Raheem B Kherani; J Antonio Aviña-Zubieta; Diane Lacaille; Allen J Lehman; Stephanie Ensworth; Jacek Kopec; John M Esdaile; Matthew H Liang Journal: J Rheumatol Date: 2008-03-01 Impact factor: 4.666
Authors: Philip J Mease; Dennis A Revicki; Jacek Szechinski; Maria Greenwald; Alan Kivitz; Leonor Barile-Fabris; Jatinderpal Kalsi; Jennifer Eames; Marjatta Leirisalo-Repo Journal: J Rheumatol Date: 2007-11-15 Impact factor: 4.666
Authors: Megan E B Clowse; Daniel J Wallace; Richard A Furie; Michelle A Petri; Marilyn C Pike; Piotr Leszczyński; C Michael Neuwelt; Kathryn Hobbs; Mauro Keiserman; Liliana Duca; Kenneth C Kalunian; Catrinel Galateanu; Sabine Bongardt; Christian Stach; Carolyn Beaudot; Brian Kilgallen; Caroline Gordon Journal: Arthritis Rheumatol Date: 2017-02 Impact factor: 10.995
Authors: Munther Khamashta; Joan T Merrill; Victoria P Werth; Richard Furie; Kenneth Kalunian; Gabor G Illei; Jorn Drappa; Liangwei Wang; Warren Greth Journal: Ann Rheum Dis Date: 2016-03-23 Impact factor: 19.103
Authors: Thomas Morel; Stefan Cano; Susan J Bartlett; Caroline Gordon; Birgit Haier; Antoine Regnault; Matthias Schneider; Christian Stach; Sophie Cleanthous Journal: Rheumatology (Oxford) Date: 2022-08-03 Impact factor: 7.046