Test-retest reliability of the StepWatch Activity Monitor outputs in healthy adults.

BACKGROUND Activity Monitors give an objective measure of usual walking performance. This study aimed to examine the test-retest reliability of the StepWatch Activity Monitor outputs (mean steps/day; peak activity index; sustained activity indices of 1, 5, 20, 30, 60 minutes; steps at high, medium, and low stepping rates). METHODS Thirty healthy adults age 18 to 49 years wore the StepWatch for 2 3-day periods at least 1 week apart. RESULTS The intraclass correlation coefficients of the StepWatch outputs ranged from 0.44 to 0.91 over 3 days. The coefficient of variation ranged from 3.0% to 51.3% over the monitoring periods, with higher variation shown for shorter monitoring periods. The most reliable 5 outputs had 95% limits of agreement between 3-day periods that were less than 40%. These were mean steps/day ( ± 39.1%), highest step rate in 1 ( ± 17.3%) and 5 ( ± 37.4%) minutes, peak activity index ( ± 25.6%), and percentage of inactive time ( ± 9.52%). CONCLUSIONS Mean steps/day, highest step rate in 1 and 5 minutes, peak activity index, and percentage of inactive time have good test-retest reliability over a 3-day monitoring period, with lower reliability shown by the other StepWatch outputs. Monitoring over 1 or 2 days is less reliable.