Between-session reliability of the star excursion balance test

Objective: To assess the learning effect, test retest reliability and measurement error associated with the SEBT. Design: Repeated-measures study. Setting: Controlled university laboratory environment. Participants: Twenty-two healthy recreational athletes (11 male age 22.3 +/- 3.7 years, 11 female age 22.8 +/- 3.1 years). Main Outcome Measures: Repeated-measures ANOVA assessed learning affects. Intraclass correlations coefficients, standard error of measurement and smallest detectable difference values were calculated to assess reliability and measurement error. Results: Results showed that excursion distances stabilised after four trials, therefore trials five to seven were analysed for reliability. Test retest reliability for all reach directions was high, with intraclass correlation coefficients ranging from 0.84 to 0.92. 95% confidence intervals, standard error of measurement and smallest detectable difference ranged from 77.84 to 94.00, 2.21-2.94% and 6.13-8.15%, respectively. Conclusion: These statistics will allow clinicians to evaluate whether changes in SEBT scores are due to change in an individual's performance or random error. The findings of this study show that the SEBT is a reliable measure of lower limb function in healthy recreational athletes. Changes in normalised scores of at least 6-8% are needed to feel confident that a real change in SEBT performance has occurred. (C) 2010 Elsevier Ltd. All rights reserved.