Evaluation of a German version of the Bath Ankylosing Spondylitis Functional Index (BASFI) and Dougados Functional Index (D-FI).

Abstract

Transcultural adaptation of the two most widely used and accepted functional indices in ankylosing spondylitis, the Bath Ankylosing Spondylitis Functional Index (BASFI) and Dougados Functional Index (D-FI). The instruments were translated and tested for internal consistency (Cronbach's coefficient alpha), test-retest reliability (intraclass correlation coefficient, ICC), construct validity (testing for association with Schober's test, finger floor distance, occiput wall distance, frequency and duration of awakenings at night, and a visual analog scale pain), and responsiveness (standardized response mean, SRM). The study sample consisted of 72 patients of a randomized, controlled clinical trial receiving either Diclofenac or placebo. Visual assessment of distribution patterns revealed a ceiling effect of both instruments. Both questionnaires had a high internal consistency (Cronbach alpha: 0.81 [BASFI], 0.85 [D-FI] and a high test-retest reliability (ICC: 0.92 [BASFI], 0.89 [D-FI]). The limited responsiveness to Diclofenac treatment (SRM: 0.46 [BASFI], 0. 33 [D-FI]) may be related to the selected study sample. The BASFI was significantly correlated with all tested validation parameters. The D-FI was only significantly correlated with finger floor distance, occiput wall distance, and duration of awakenings at night. The transculturally adapted version of both functional indices are valid, reliable, and internally consistent. Because of psychometric advantages, the BASFI may be preferred in clinical trial settings.