Validation of the stabilometer balance test

Background: A pronounced discrepancy exists between balance assessments for stroke survivors that are used for clinical purposes and those used for research. Clinical assessments like the Berg Balance Scale generally have stronger ecological validity, whereas research-based assessments like posturography are generally more reliable and precise. We developed a stabilometer balance test (SBT) that aims to couple measurement reliability and precision to clinical meaningfulness by means of a personalized and adaptive test procedure. Research question: To examine the validity, reliability, and measurement error of the stabilometer balance test in inpatient stroke patients. Methods: In this cross-sectional study, inpatient stroke patients (FAC > 2) were tested on a stabilometer with adjustable resistance to mediolateral movement. A modified staircase procedure was used to adapt task difficulty (i.e., rotational stiffness) on a trial-by-trial basis. The main outcome was the threshold stiffness at which a patient could just stay balanced. Threshold stiffness was correlated with the Berg Balance Scale and posturography measurements to determine concurrent validity (N = 86). Test-retest reliability (N = 23) was analyzed with the Intraclass Correlation Coefficient (ICC). Floor and ceiling effects were assessed. The minimal detectable change was determined at individual and group level. Results: Threshold rotational stiffness moderately correlated with the Berg Balance Scale (r=−0.559, p < 0.001), and the absolute path length of the center of pressure during posturography (r=0.348, p = 0.006). Test-retest reliability was good to excellent (ICC=0.869; 95%CI=0.696–0.944). There were no floor or ceiling effects. The minimal detectable change was sufficiently small to detect relevant changes in balance control both on individual and group level. Relevance: The SBT is both a valid and reliable balance assessment in stroke patients. It is at least as precise as current clinically preferred measures and does not suffer from ceiling effects. Therefore, it is suitable for use in clinical practice as well as research.