A Comparison of Psychometric Properties of the Smart Balance Master System and the Postural Assessment Scale for Stroke in People Who Have Had Mild Stroke

Abstract

Chien C-W, Hu M-H, Tang P-F, Sheu C-F, Hsieh C-L. A comparison of psychometric properties of the Smart Balance Master system and the Postural Assessment Scale for Stroke in people who have had mild stroke. Objective To compare the psychometric properties (including the test-retest reliability, responsiveness, and predictive validity) of the Smart Balance Master (SBM) system and the Postural Assessment Scale for Stroke patients (PASS) in patients with mild stroke. Design One repeated-measures design (at a 2-wk interval) was used to examine the test-retest reliability of the SBM and PASS, and another similar design was applied to investigate their responsiveness. Patients who participated in the responsiveness study were followed up approximately 1 year later, and the predictive validity of the SBM system and PASS were examined by assessing the patients’ comprehensive activities of daily living (ADL) function. Setting Three rehabilitation units in Taiwan. Participants Twenty patients with chronic stroke in the reliability study; 40 and 32 patients who had recently had a stroke in the responsiveness and predictive validity studies, respectively. Interventions Not applicable. Main Outcome Measures Three computerized tests of the SBM (the equilibrium score of the Sensory Organization Test, scores in rhythmic weight-shifting tests, and scores in the limits of stability test) and the PASS were used. The combination of the Barthel Index and Frenchay Activities Index was used to represent the comprehensive ADL function. Results For the SBM, all but the weight-shifting tests of the SBM had moderate to high reliability (intraclass correlation coefficient [ICC] range, .78−.91). The responsiveness of the equilibrium score and the limits of stability test were moderate (effect size [ d], .63) and small ( d range, .27−.33), respectively, whereas the responsiveness of the weight-shifting tests was limited ( d range, .04−.29). All but the weight-shifting tests of the SBM in the second evaluation had acceptable predictive validity for comprehensive ADL function ( r 2 range, .15−.17). The PASS showed high reliability (ICC=.84) and small responsiveness ( d=.41), and the PASS in the second evaluation had acceptable predictive validity ( r 2=.24). Conclusions The PASS and the equilibrium score and limits of stability scores of the SBM had acceptable test-retest reliability, responsiveness, and predictive validity in patients with mild stroke, but the psychometric properties of the weight-shifting tests of the SBM should be further examined before consideration of their usage in patients with stroke.