Reliability of the Actigraph GT3X+Accelerometer in Adults under Free-Living Conditions

Background Reliability of the Actigraph GT3X+ accelerometer has not been determined under normal wear time criteria in a large sample of subjects and accelerometer units. The aim of this study was to assess contralateral hip difference and inter-instrument reliability of the Actigraph GT3X+ monitor in adults under long-term free-living conditions. Methods Eighty-seven adult subjects (28 men; mean (standard deviation) age 31.3 (12.2) years; body mass index 23.7 (3.1) kg/m(2)) concurrently wore two GT3X+ accelerometers (174 units in total) attached to contralateral hips for 21 days. Reliability was assessed using Bland-Altman plots, mixed model regression analyses and absolute measures of agreement for different lengths of data accumulation (single-day-, 7-day- and 21-day periods). Results There were no significant differences between contralateral hips (effect size <= 0.042; p >= .213). Inter-instrument reliability increased with increased length of data-accumulation. For a 7-day measurement period (n = 232 weeks), limits of agreement were +/- 68 cpm (vertical axis) and +/- 81.3 cpm (vector magnitude) for overall physical activity (PA) level, +/- 51 min for sedentary time, +/- 18.2 min for light PA, +/- 6.3 min for moderate PA, +/- 3.5 min for vigorous PA, and +/- 6.7 min for moderate-to-vigorous PA. Conclusions The Actigraph GT3X+ accelerometer is a reliable tool for measuring PA in adults under free-living conditions using normal data-reduction criteria. Contralateral hip differences are very small. We suggest accelerometers be attached to the right hip and data to be accumulated over several days of measurement.