AB Purpose: Wrist-worn accelerometer devices measure sleep in free-living settings. Few studies, however, have investigated whether these devices can also measure waking movement behavior (e.g., total movement volume, physical activity). The purpose of this study was to investigate the ability of a wrist-worn Actiwatch 2 sleep monitor to rank total movement volume and physical activity levels compared with a waist-worn ActiGraph GT1M accelerometer and self-reported leisure time physical activity, respectively. In addition, we compared temporally matched activity measured via the ActiGraph GT1M and Actiwatch 2 over the study week. Methods: A subset of women from the Healthy Women Study (n = 145; age, 73.3 +/- 1.7 yr) wore an Actiwatch 2 on their nondominant wrist and an ActiGraph GT1M on their dominant hip for seven consecutive days. Participants recorded their leisure time physical activity in a 7-d diary and completed the past year version of the Modifiable Activity Questionnaire. Analyses were conducted for all wake periods and separately for active periods when both devices were worn. Results: Spearman rank-order correlation coefficients for total movement volume between the Actiwatch 2 and ActiGraph GT1M were significant for wake periods (r = 0.47, P < 0.001) and, to a lesser extent, for active periods (r = 0.26, P < 0.01). However, the Actiwatch 2 did not rank participant's physical activity levels similarly to self-reported leisure time physical activity estimates ([kappa] <= 0.05, P > 0.05). Multilevel model analyses comparing temporally matched activity measured via the ActiGraph GT1M and Actiwatch 2 suggest that the two devices yielded similar levels of activity during wake periods (B = 0.90; SE, 0.008; P < 0.001) and during active periods (B = 0.81; SE, 0.01; P < 0.001). Conclusions: A wrist-worn Actiwatch 2 may be useful for ranking total movement volume and for assessing the pattern of activity over a day in older women. However, our data do not support using a wrist-worn Actiwatch 2 device for measuring physical activity.
© 2001-2024 Fundación Dialnet · Todos los derechos reservados