Actical Accelerometry Cut-Points for Quantifying Levels of Exertion: Comparing Normal and Overweight Adults

Weight, body fatness and ambulatory pattern all have the potential to affect accelerometer output and cause differences in output between overweight and normal-weight adults. The purpose of this study was to determine if Actical (Philips Respironics, Bend, OR) activity count cut-points for moderate and vigorous intensity exercise are different for overweight adults compared to normal-weight adults. Overweight adults with BMI >25 kg/m² (n=29) and Normal-Weight adults (n=25) walked at 3.2 and 4.8 km·h^-1 and ran at 6.4 km·h^-1 on a treadmill while simultaneously wearing an Actical accelerometer and obtaining measurements of oxygen uptake. Counts per minute (counts·min^-1) were determined at 3 METS (moderate) and 6 METS (vigorous) using ROC curves. The counts·min^-1 at 3 METs was 1726 and 1923 counts·min^-1 for Overweight and Normal-Weight groups, respectively. The cut-points at 6 METs were 4117 and 4032 counts·min^-1 for Overweight and Normal-Weight groups, respectively. The differences between groups were not statistically significant (p>0.73 for both). Correlations between BMI and counts·min^-1 were not significant (p>0.05) at any speed for the Normal-Weight group but were significant at 3.2 and 4.8 km·h^-1 for the Overweight group. Although there appears to be some relationship between activity counts·min^-1 and BMI, the results suggest that similar cut-points may be used for normal weight and overweight adults. However, the greater variability in counts at each speed and lower ROC curve areas for overweight adults suggest that it is harder to classify the activity intensity of overweight subjects compared to normal weight subjects.