Cross-Cultural and Construct Validity of the Animated Activity Questionnaire.

Abstract

The Animated Activity Questionnaire (AAQ) assesses activity limitations in patients with hip/knee osteoarthritis and consists of video animations; the patients choose the animation that best matches their own performance. The AAQ has shown good validity and reliability. This study aims to evaluate cross-cultural and construct validity of the AAQ. Cross-cultural validity was assessed using ordinal logistic regression analysis to evaluate differential item functioning (DIF) across 7 languages. Construct validity was assessed by testing correlations between the AAQ and a patient-reported outcome measure (PROM) and performance-based tests. Data from 1,239 patients were available. Compared to the Dutch language (n = 279), none of the 17 items showed DIF in English (n = 202) or French (n = 193), 1 item showed uniform DIF in Spanish (n = 99) and Norwegian (n = 62), and 2 items showed uniform DIF in Danish (n = 201). In all these languages, the occurrence of DIF did not influence the total score, which remained comparable with the original Dutch version. For Italian (n = 203) versus Dutch, however, 6 items showed uniform DIF, and 1 item showed nonuniform DIF, indicating some problems with the cross-cultural validity between these countries. With regard to construct validity, the correlations with PROM (0.74) and performance-based tests (0.36-0.68) were partly as expected (>0.60). The AAQ, an innovative tool to measure activity limitations that can be placed on the continuum between PROMs and performance-based tests, showed a good overall cross-cultural validity, and seems to have great potential for international use in research and daily clinical practice in many European countries.