Psychometric properties of the simple test for evaluating hand function in patients with stroke

Objective: The simple test for evaluating hand function (STEF) is widely used for clinical evaluation of upper extremity function in Japan. However, except for test-re-test reliability, its psychometric properties have not been investigated. The aim of this study is to explore its internal consistency, concurrent validity and responsiveness in patients with sub-acute stroke. Design: This was a prospective longitudinal study. Patients: Thirty-four inpatients who had suffered hemiparetic stroke within 60 days of participation were enrolled. Methods: To investigate its internal consistency and responsiveness, they were assessed with the STEF and Action Research Arm Test (ARAT) at admission and 3 weeks later. To explore its concurrent validity, the Fugl-Meyer Assessment (FMA), Motor Activity Log (MAL) and Functional Independence Measure (FIM (TM)) were also evaluated at admission. Results: The Cronbach's alpha for the STEF was 0.98-0.99, indicating excellent internal consistency. The STEF score strongly correlated with the ARAT, FMA and MAL scores and moderately with the FIM (TM) score. Effect sizes and the standardized response mean were 0.27 and 0.52 for the STEF and 0.30 and 0.95 for the ARAT, respectively. Conclusion: This study indicates that the STEF is reliable, valid and sensitive to changes when applied to patients with sub-acute stroke.