Absolute and Relative Reliability of the Timed ‘Up & Go’ Test and ‘30second Chair-Stand’ Test in Hospitalised Patients with Stroke

OBJECTIVE The timed 'Up & Go' test and '30second Chair-Stand' test are simple clinical outcome measures widely used to assess functional performance. The reliability of both tests in hospitalised stroke patients is unknown. The purpose was to investigate the relative and absolute reliability of both tests in patients admitted to an acute stroke unit. METHODS Sixty-two patients (men, n = 41) attended two test sessions separated by a one hours rest. Intraclass correlation coefficients (ICC2,1) were calculated to assess relative reliability. Absolute reliability was expressed as Standard Error of Measurement (with 95% certainty-SEM95) and Smallest Real Difference (SRD) and as percentage of their respective means if heteroscedasticity was observed in Bland Altman plots (SEM95% and SRD%). RESULTS ICC values for interrater reliability were 0.97 and 0.99 for the timed 'Up & Go' test and 0.88 and 0.94 for '30second Chair-Stand' test, respectively. ICC values for intrarater reliability were 0.95 and 0.96 for the timed 'Up & Go' test and 0.87 and 0.91 for '30second Chair-Stand' test, respectively. Heteroscedasticity was observed in the timed 'Up & Go' test. Interrater SEM95% ranged from 9.8% to 14.2% with corresponding SRD% of 13.9-20.1%. Intrarater SEM95% ranged from 15.8% to 18.7% with corresponding SRD% of 22.3-26.5%. For '30second Chair-Stand' test interrater SEM95 ranged between 1.5 and 1.9 repetitions with corresponding SRD of 2 and 3 and intrarater SEM95 ranged between 1.8 and 2.0 repetitions with corresponding SRD values of 3. CONCLUSION Excellent reliability was observed for the timed 'Up & Go' test and the '30second Chair-Stand' test in hospitalised stroke patients. The thresholds to detect a real change in performance were 18.7% for the timed 'Up & Go' test and 2.0 repetitions for the '30second Chair-Stand' in groups of patients and 26.5% and 3 repetitions in individual patients, respectively.