Reliability of observed step count in the Timed Up and Go test: A cross-sectional study

Background and aim The Timed Up and Go (TUG) test is one of the most widely used assessments. This study investigated the psychometric properties of the observed step count during this test performed by experienced physiotherapists, using video recordings of healthy adults. Methods In this cross-sectional study two physiotherapists independently observed video recordings of healthy participants completing the TUG twice, and obtained step count and the time taken to complete the TUG. For the intra-rater reliability, the step count observation was repeated after 10 months. ICC and Bland Altman plots were used to analyse the intra-and inter-rater reliability. Spearman correlation coefficient was used to investigate the correlation between step count and time taken. Results Thirty-six recordings from 31 healthy adults were analysed. The median step count across both raters and rating sessions was 12. The intra-rater reliability for step count was 0.924 (95% CI: 0.874 to 0.954), whereas the inter-rater reliability was 0.857 (95% CI: 0.730 to 0.919), the minimal detectable change (MDC) was 0.51. The correlation between the time taken and the average step count was rho = 0.71 ( p < 0.001). Conclusion The results indicate that the observed step count during the TUG in healthy adults using video recordings is reliable and is positively correlated with the time to complete the test. With an MDC of 0.51 steps, a change of more than one step confidently reflects a real change in performance, beyond measurement error. Further research should explore this outcome parameter in individuals with irregular and unpredictable gait patterns.