Reliably measuring habitual free-living physical activity with the Actical® activity monitor
Date
2007
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Montana State University - Bozeman, College of Education, Health & Human Development
Abstract
The purpose of this study was to compare reliability measures between a hip- and wrist-worn Actical® accelerometer. A group of 40 (25 female, 15 male) Montana State University employees wore both monitors for 14 consecutive days. Intraclass correlation coefficients (ICC) were determined for three variables for each monitor wearing location for bouts of one, three, five, eight and ten consecutive minutes over the entire monitoring period and for each single day of monitoring. The three dependent variables were the raw activity count values (CNTMV, counts/day) at or above specified moderate intensity activity thresholds, and the corresponding summation of total time (TMV, mins/day) and activity energy expenditure (AEEMV, kcals/day) spent at or above a moderate intensity. As time bouts increased from one to ten minutes, ICC values decreased for both the hipand wrist-worn monitors. The number of days for the hip-worn monitor to reach an ICC of 0.80 ranged from 3 to 4, 2 to 5, and 3 to 6 for CNTMV, AEEMV, and TMV, respectively, while the wrist-worn monitor took 10 to 12, 5 to 8, and 3 to 5 days for the same respective variables.
No substantial differences in physical activity between weekdays and weekend days were observed. Subjects were divided post-hoc into an Active or Inactive group. Higher ICC values were observed with the hip-worn monitor in the Active group (5, 2 to 9, and 3 to 10 days to reach an ICC of 0.80 for CNTMV, AEEMV, and TMV, respectively), while higher ICC values were observed with the wrist-worn monitor in the Inactive group (3 to 4, 4 to 5, and 3 to 4 days to reach an ICC of 0.80 for CNTMV, AEEMV, and TMV, respectively). No substantial differences in physical activity were seen between weekdays and weekend days for the Active or Inactive group. These findings indicate that a hip-worn Actical® will reach a desired ICC value in fewer monitoring days than a wrist-worn Actical®, except in inactive populations where the wrist-worn Actical® required fewer monitoring days.
No substantial differences in physical activity between weekdays and weekend days were observed. Subjects were divided post-hoc into an Active or Inactive group. Higher ICC values were observed with the hip-worn monitor in the Active group (5, 2 to 9, and 3 to 10 days to reach an ICC of 0.80 for CNTMV, AEEMV, and TMV, respectively), while higher ICC values were observed with the wrist-worn monitor in the Inactive group (3 to 4, 4 to 5, and 3 to 4 days to reach an ICC of 0.80 for CNTMV, AEEMV, and TMV, respectively). No substantial differences in physical activity were seen between weekdays and weekend days for the Active or Inactive group. These findings indicate that a hip-worn Actical® will reach a desired ICC value in fewer monitoring days than a wrist-worn Actical®, except in inactive populations where the wrist-worn Actical® required fewer monitoring days.