Concurrent validity, test-retest reliability and minimal detectable change of an adaptive balance test on a medio-lateral stabilometer in patients with stroke

Abstract

Background: Balance evaluation is a vital part of stroke rehabilitation. Recently, an adaptive stabilometer balance test has been developed which uses a modified staircase test procedure. If this test is able to accurately quantify balance performance in stroke patients with mild to moderate impairments it could be a valuable addition to currently used balance measures.

Aim: To evaluate the concurrent validity, test-retest reliability and minimal detectable change of an adaptive balance test on a medio-lateral stabilometer in stroke patients.

Methods: A cross-sectional validation design was used to carry out this study. Validity measurements consisted of a stabilometer balance test, a Berg Balance Scale and a posturography measurement. Participants of the test-retest reliability sample performed an additional stabilometer balance test. Validity was analyzed based on Pearson correlation. Test-retest was analyzed with Intraclass Correlation Coefficients. Minimal detectable change was calculated both at group level and at individual level.

Results: The validity sample consisted of 86 participants. Twenty-three participants participated in the reliability sample. A correlation of r=0.384 (p=0.002) and r=0.123 (p=0.339) was found between the stabilometer measurement and posturography for ‘sway’ and ‘curviness’, respectively. A correlation of r=-0.591 (P<0.001) was found between the stabilometer measurement and the Berg Balance Scale. The Intraclass Correlation Coefficient between both stabilometer measurements was 0.875. An ICC of 0.682 was found between the performance trails. Mean stabilometer balance test outcome (rotational stiffness) was 38.94Nm (±29.44). The minimal detectable change was calculated to be 20.996 and 4.378 at individual level and group level respectively.

Conclusion: The stabilometer balance test is a reliable measure, however it seems to measure a broader, more clinical concept of balance than just the theoretical construct. The stabilometer balance test is sensitive on group level, however it is not sensitive enough to be used on individual level.

Clinical Relevance: The stabilometer balance test can be used as balance evaluation tool within studies. It does not have a strong ceiling effect that some other balance measures do have. Furthermore the test outcome can be used as benchmark for further stabilometer balance training.