Psychometric evaluation of the Arabic version of the multidimensional assessment of fatigue scale (MAF) for use in patients with ankylosing spondylitis

Fatigue is a frequent symptom during ankylosing spondylitis (AS) often under estimated which needs to be measured properly with respect to its intensity by appropriate measures, such as the multidimensional assessment of fatigue (MAF). The aims of this study were to translate into the classic Arabic version of the MAF questionnaire and to validate its use for assessing fatigue in Moroccan patients with AS. The MAF contains 16 items with a global fatigue index (IGF). The MAF was translated and back-translated to arabic, pretested and reviewed by a committee following the Guillemin criteria (J Clin Epidemiol 46:1417?C1432, 1993). It was then validate on 110 Moroccan patients with AS. Reliability for the 3-day test?Cretest was assessed using internal consistency by Cronbach??s alpha coefficient and the intra-class correlation coefficient (ICC). External construct validity was assessed by correlation with pain, activity of disease and other keys variable. The reproducibility of the 15 items was satisfactory with a kappa statistics of agreement superior to 0.6. The ICC for IGF score reproducibility was good and reached 0.98 (IC 95%, 0.96?C0.99). The internal consistency was at 0.991 with Cronbach??s alpha coefficient. The construct validity showed a positive correlation between MAF and the axial (r?=?0.34) and peripheral (r?=?0.32) visual analogical scale, the Bath ankylosing spondylitis disease activity index (BASDAI) (r?=?0.77), the first item of BASDAI (r?=?0.85), the functional disability by the Bath ankylosing spondylitis functional index (r?=?0.64), the erythrocyte sedimentation rate (r?=?0.43) and the C reactive protein (r?=?0.30) (for all P?