CROSS-CULTURAL VALIDITY OF THE BRAZILIAN VERSION OF THE ABILHAND QUESTIONNAIRE FOR CHRONIC STROKE INDIVIDUALS, BASED ON RASCH ANALYSIS

Abstract

Objective: To evaluate the cross-cultural validity of the Brazilian version of the ABILHAND for stroke individuals. Subjects: A total of 107 community-dwelling chronic stroke survivors; mean age 58 years. Methods: Cross-cultural adaptation of the ABILHAND followed standardized procedures. Measurement properties of the adapted version were analysed using Rasch analysis. Cross-cultural validity was based on cultural invariance analyses. Results: The ABILHAND-Brazil demonstrated satisfactory performance as a rating scale. Only one item exhibited misfit to the Rasch model expectations. Principal component analysis of the residuals showed that the manual ability of the individuals encompassed different contents related to the degree of the paretic upper limb involvement in performing manual activities. Some minor local dependency was identified in 2 pairs of items (residual correlations >0.3). Furthermore, the adapted version exhibited high levels of reliability, no floor effects, and minimal ceiling effect. Analyses of cultural invariance showed that the ABILHAND-Original and ABILHAND-Brazil calibrations can be used interchangeably. Conclusion: The ABILHAND specific for stroke individuals demonstrated satisfactory measurement properties for use within both clinical and research contexts in Brazil, and cross-cultural validity for use in international/multicentric studies between Brazil, Belgium, and Italy.