Rasch Validation and Comparison of the Mini-BESTest and S-BESTest in Individuals with Stroke

Abstract

Objective: The purpose of this study was to compare the psychometric properties of the Mini-Balance Evaluation Systems Test (Mini-BESTest) and S-BESTest and to evaluate which is more suitable for use in clinical settings for individuals with stroke. Methods: This multicenter retrospective cross-sectional study investigated 115 individuals with stroke (mean age, 70.8 y [SD = 11.2 y]) who were able to stand without physical assistance. All individuals were examined with the BESTest and with the Mini-BESTest and S-BESTest scored based on the BESTest results. The data were analyzed using a Rasch analysis (partial credit model). Results: The Mini-BESTest results revealed a correctly functioning rating scale, good fit of the data to the model (apart from 1 overfit item), good reliability for both persons and items (6 statistically detectable levels of balance ability), local dependence between 1 item pair, and essential unidimensionality. The S-BESTest results demonstrated disordered rating scale thresholds (1 response option required collapsing), good fit of the data to the model (apart from 1 underfit item), good reliability for both persons and items (5 statistically detectable levels of balance ability), local dependence between 2 item pairs, and essential unidimensionality. Conclusion: The analyses confirmed that the reliability of the S-BESTest was good and unidimensional and that the test provides several improved points, such as item redundancy and local independence of items. Nevertheless, the Mini-BESTest results supported previous findings as a whole and were better than those from the S-BESTest. Impact: Rasch analysis demonstrated that the Mini-BESTest was a better balance assessment scale than the S-BESTest for individuals with stroke based on its psychometric properties. The Mini-BESTest may serve as a useful scale for assessing balance in individuals with stroke, and a keyform plot and strata may help clinical decision-making in terms of interpreting scores and goal setting.