VALIDITY AND RELIABILITY OF THE CSA ACCELEROMETER WORN ON THE ANKLE, HIP, AND WRIST

53 The validity and reliability of the CSA accelerometer was assessed in 60 young adults (30 men and 30 women, age 23.4 ± 2.9 yr, mean ± SD) during treadmill walking and jogging at 3.2, 6.4 and 9.7 km·hr-1 at 0% grade and at 6.4 km·hr-1 at 5% grade. Energy expenditure, measured by indirect calorimetry with open circuit spirometry (Sensormedics, Vmax 29), served as the criterion measure. Subjects wore four CSAs, one on the wrist and ankle, and one on each hip inserted into a pouch and secured by a waist strap. Intersession reliability was determined in 10 men and 10 women. Intraclass correlation coefficients for intersession reliability ranged from 0.37-0.68 for the wrist and ankle, and 0.57-0.83 for the hip. Interinstrument reliability ranged from R=0.55-0.72, with no significant differences between the right and left hip (paired t-test, p>0.05). However, significant differences in CSA counts were found between the hip, ankle and wrist at each velocity. Repeated measures ANOVA indicated that the CSA was sensitive to changes in speed at all sites (p<.001). Paired t-tests comparing mean CSA counts at 0% vs. 5% grade demonstrated no significant differences for the ankle or wrist, but significant differences for the hip (4772 @ 0% vs. 5000 @ 5%, p=.002), however, the mean difference was very small, indicating poor sensitivity to changes in grade. Regression analysis of VO2 vs CSA counts yielded the following adjusted R2 values: ankle, 0.37; wrist, 0.80, hip, 0.86. These data indicate that site selection for the CSA is an important consideration for assessing walking and jogging. When worn on the hip, it has moderate to high reliability and validity. With limitations, it can be used to estimate energy expenditure in large-scale studies.