ASSESSMENT OF PHYSICAL ACTIVITY WITH THE CSA ACCELEROMETER: LAB VS. FIELD MEASURES

54 Our purpose was to determine, in a field setting using the CSA accelerometer, ranges of physical activity corresponding to light, moderate and vigorous intensities, based on data in the 1996 Surgeon General's Report, and to compare field data with previously gathered laboratory data testing the same intensities of activity. Thirty non-obese adults (age 28.1 ± 4.7 yr, mean ± SD) were instructed, in randomized order, to“stroll”, “walk briskly” and “jog at a comfortable pace” for 5 min each around a flat, outdoor 400m track. Subjects wore a CSA monitor on their hip inserted into a pouch and secured by a waist strap. Data from minutes 3 and 4 were averaged for analysis. The velocity at which each subject walked and jogged was converted to a MET value using metabolic equations. Estimated field METs were then compared with METs measured by indirect calorimetry in the laboratory, while subjects walked and jogged on a treadmill at 3.2, 6.4 and 9.7 km·hr-1 at 0% grade. For both conditions, the relationship between CSA counts and METs was linear(lab: R2=0.89; field: R2=0.71, p<0.001). However, the CSA counts corresponding to 2 METs were over three times greater in the field compared to the lab (2803 vs 755 counts·min-1), while the field counts at 4 METs were 25% higher than those in the lab (4022 vs 3207 counts·min-1). However, at 7 METs the field counts were approximately 15% lower compared to the lab (5852 vs. 6885 counts·min-1). These data indicate that the CSA overestimates low to moderate physical activity and underestimates vigorous activity in the field compared to the laboratory. Evaluation of other types of activity is needed to determine its usefulness for assessing various intensities of physical activity in epidemiologic studies.