Cross-cultural validity of the Animated Activity Questionnaire for patients with hip and knee osteoarthritis: a comparison between the Netherlands and Brazil

Abstract

Background: The Animated Activity Questionnaire (AAQ) was developed in the Netherlands to assess activity limitations in individuals with hip/knee osteoarthritis (HKOA). The AAQ is easy to implement and minimizes the disadvantages of questionnaires and performance-based tests by closely mimicking real-life situations. The AAQ has already been cross-culturally validated in six other countries.

Objective: To assess the cross-cultural validity, the construct validity, the reliability of the AAQ in a Brazilian sample of individuals with HKOA, and the influence of formal education on the construct validity of the AAQ.

Methods: The Brazilian sample (N = 200), mean age 64.4 years, completed the AAQ and the Western Ontario and McMaster Universities Index (WOMAC). A subgroup of participants performed physical function tests and completed the AAQ twice with a one-week interval. The Dutch sample (N = 279) was included to examine Differential Item Functioning (DIF) between the scores obtained in the Netherlands and Brazil. For this purpose, ordinal regression analyses were used to evaluate whether individuals with the same level of activity limitations from the two countries (the Dutch as the reference group) scored similarly in each AAQ item. To evaluate the construct validity, correlation coefficients were calculated between the AAQ, the WOMAC domains, and the performance-based tests. To evaluate reliability, the Cronbach's alpha coefficient, the intraclass correlation coefficient, and the standard error of measurement (SEM) were calculated.

Results: The AAQ showed significant correlations with all the WOMAC domains and performance-based tests (rho=0.46-0.77). The AAQ showed high internal consistency (Cronbach's alpha=0.94), excellent test-retest reliability (ICC_2,1 = 0.98), and small SEM (2.25). Comparing to the scores from the Netherlands, the AAQ showed DIF in two items, however, they did not impact on the total AAQ score (rho=0.99).

Conclusion: Overall, the AAQ showed adequate cross-cultural validity, construct validity, and reliability, which enables its use in Brazil and international/multicenter studies.

Keywords: Activity limitations measurement; Cross-cultural validity; Measurement properties; Osteoarthritis; Rehabilitation; Video animations.