Validity, reliability, and sensitivity to change of a French version of the arthritis impact measurement scales 2 (AIMS2) in patients with rheumatoid arthritis treated with methotrexate

Objective, To develop and validate a cross cultural version of the Arthritis Impact Measurement Scales 2 (AIMS2) to be used by French speaking populations. Methods, A French version of the AIMS:! was obtained using back translation, committee review, and pretesting. The French AIMS2 was studied in 127 patients with rheumatoid arthritis (RA) about to receive therapy with methotrexate (MTX). Construct validity of the questionnaire was assessed by factor analysis. Convergent validity was evaluated by correlation coefficients with joint counts, pain assessment, and sedimentation rate. Reliability was assessed by test-retest procedure at a 10-day interval, Cronbach's coefficients of internal consistency, and within scale factor analyses. Sensitivity to change after 12 and 24 weeks of therapy with MTX was assessed with computation of standardized response means (SRM) and paired t test comparisons. Results, Factor analyses of the French version clearly identified the same scales of the AIMS2, except for the walking and bending scale which loaded on several factors. Convergent validity of the physical and symptom components of the instrument was demonstrated by significant correlations with clinical and laboratory features. All the scales were reliable (intraclass correlation coefficients: 0.65 to 0.90; percentage of explained variance larger than 50% in all but one scale; Cronbach's alpha: 0.70 to 0.90). Sensitivity to change was demonstrated in 11 of the 12 scales (SRM: 0.30 to 0.77). Most of the improvement was noted by Week 12. Conclusions, This cross cultural adaptation of AIMS2 in French is valid, reliable, and responsive in patients with RA in whom MTX therapy is instituted. It would permit international comparison studies. This study provides evidence for construct validity and responsiveness of the original version of the AIMS2, not demonstrated previously.