A novel approach to measuring wobble board performance in individuals with chronic ankle instability

Computerized wobble boards (WB) are used to objectively assess balance in healthy and chronic ankle instability individuals. As in field setting health professionals might not own WB, objective evaluations are not always feasible. Therefore, the aim of this tudy was to investigate the contribution of sagittal plane joints angular-displacement and anthropometrics to predict equations to estimate WB performance by portable two-dimensional motion analysis (2D-MA) and cross-validate the developed equations in chronic ankle instability individuals. Thirty-nine healthy and twenty chronic ankle instability individuals stood on a WB in single stance position. The balance test consisted of three 30s trials per limb keeping the platform flat at 0°. Trials were video recorded, and three time-segments joints angular-displacement analyzed with 2D-MA: segment 1 (T1) including 30s data, segment 2 (T2) from second 0 to 10, segment 3 (T3) only the first 5s. Mixed regression for multilevel models was used to estimate WB performance for each time-segment and to examine limb differences for the predicted WB performance in chronic ankle instability sample. The accuracy of the equations to detect injured limbs was calculated via area under the curve for receiver operating characteristic. Ankle and knee angular-displacement parameters, body height and lower limb length were the major predictors of WB performance for the extrapolated models (p < 0.05; R² = 0.83-0.56). The measured WB performance and T1 model showed significant (p < 0.05) performance differences between the injured and uninjured limbs. Receiver operating characteristic analysis showed an asymptotic significance of 0.03 for T1 equation with area under the curve of 0.70. The proposed models provide different methods to quantify the performance and accurately detect the injured limb in individuals with unilateral chronic ankle instability, when measuring balance via WB might not be feasible. App-makers may use the equations to provide an automatic all-in-one system to monitor the performance status and progress.