Responsiveness and validity of three outcome measures of motor function after stroke rehabilitation.

Abstract

This study investigated and compared the responsiveness and validity of the Fugl-Meyer Assessment (FMA), the Action Research Arm Test (ARAT), and the Wolf Motor Function Test (WMFT) for patients after stroke rehabilitation. A total of 57 patients with stroke received 1 of 3 rehabilitation treatments for 3 weeks. At pretreatment and posttreatment, the 3 outcome measures, as well as the Functional Independence Measure (FIM) as the external criterion, were administered. The standardized response mean (SRM) and the Wilcoxon signed rank test were used to examine the responsiveness. Construct validity and predictive validity were examined by the Spearman correlation coefficient (rho). The responsiveness of the FMA, ARAT, and WMFT functional ability scores was large (SRM=0.95-1.42), whereas the WMFT performance time score was small (SRM=0.38). The responsiveness of the FMA was significantly larger than those of the ARAT and the WMFT-TIME, but not the WMFT functional ability scores. With respect to construct validity, correlations between the FMA and other measures were relatively high (rho=0.42-0.76). The FMA and the WMFT performance time scores at pretreatment had moderate predictive validity with the FIM scores at posttreatment (rho=0.42-0.47). In addition, the ARAT and the WMFT functional ability scores revealed a low predictive validity with the FIM (rho=0.17-0.26). The results support the FMA and the WMFT-FAS are suitable to detect changes over time for patients after stroke rehabilitation. While simultaneously considering the responsiveness and validity attributes, the FMA may be a relatively sound measure of motor function for stroke patients based on our results. Further research based on a larger sample is needed to replicate the findings.