Reliability of the sit-to-stand test over dispersed test sessions

This study focused on the reliability, differences, and error of sit-to-stand (STS) test measurements obtained during 3 sessions separated by 6 weeks. Twenty-one community-dwelling elders (65-85 years) participated. The mean times required to complete five STS repetitions at each test session were 13.2, 12.0, and 10.9 seconds. The times were reliable (intraclass correlation coefficient = 0.82) but were significantly different (F = 6.889, p = 0.003) and demonstrated a linear trend over sessions (F = 10.157, p = 0.005). The technical error of the measurement for the times ranged from 1.7 to 2.8 seconds. Although the intraclass correlation coefficient showed the STS measurement to possess good reliability, the significant difference and trend over time suggests that subjects may experience some test familiarization with repeated performance. Given this possibility and the technical error of the measurement, clinicians should be cautious in equating small decreases in STS time to real improvement.