Reliability and sensitivity to change of the Bristol Rheumatoid Arthritis Fatigue scales.

Abstract

To examine the reliability (stability) and sensitivity of the Bristol Rheumatoid Arthritis Fatigue scales (BRAFs) and patient-reported outcome measures (PROMs) developed to capture the fatigue experience. The Multi-Dimensional Questionnaire (BRAF-MDQ) has a global score and four subscales (Physical Fatigue, Living with Fatigue, Cognitive Fatigue and Emotional Fatigue), while three numerical rating scales (BRAF-NRS) measure fatigue Severity, Effect and Coping. RA patients completed the BRAFs plus comparator PROMs. Reliability (study 1): 50 patients completed questionnaires twice. A same-day test-retest interval (minimum 60 min) ensured both time points related to the same 7 days, minimizing the capture of fatigue fluctuations. Reliability (study 2): 50 patients completed the same procedure with a re-worded BRAF-NRS Coping. Sensitivity to change (study 3): 42 patients being given clinically a single high dose of i.m. glucocorticoids completed questionnaires at weeks 0 and 2. The BRAF-MDQ, its subscales and the BRAF-NRS showed very strong reliability (r = 0.82-0.95). BRAF-NRS Coping had lower moderate reliability in both wording formats (r = 0.62, 0.60). The BRAF-MDQ, its subscales and the BRAF-NRS Severity and Effect were sensitive to change, with effect sizes (ESs) of 0.33-0.56. As hypothesized, the BRF-NRS Coping was not responsive to the pharmaceutical intervention (ES 0.05). Preliminary exploration suggests a minimum clinically important difference of 17.5% for improvement and 6.1% for fatigue worsening. The BRAF scales show good reliability and sensitivity to change. The lack of BRAF-NRS Coping responsiveness to medication supports the theory that coping with fatigue is a concept distinct from severity and effect that is worth measuring separately.