Validating the SenseWearŪ Armband as a Measure of Sedentary Behavior

Background/Purpose: The BodyMedia SenseWear® armband is a wearable monitor to assess metabolic physical activity and energy expenditure. Physical activity is an important health outcome variable, however recent literature has expressed the equally detrimental effects of prolonged sedentary behavior (SB). The SenseWear® armband also computes minutes of SB, but little in known about the validity of the armband for measuring SB. Therefore the purpose of this study was to examine the validity of the SenseWear® armband as a measure of SB in adults.

Method: A total of 24 adults (mean age 31.91 ± 11.34 yrs) were recruited to complete this study. Participants were required to wear the SenseWear armband, an Actigraph GT3X+ accelerometer, and to complete the Sedentary Behavior Record (SBR) for seven consecutive days. The armband estimates energy expenditure using complex algorithms with data from multiple sensors. Minutes of SB are determined from where MET values are < 1.5. Accelerometer sedentary time was computed where activity counts were < 150 counts per minute (Kozey-Keadle, Libertine, Lyden, Staudenmayer, & Freedson, 2011). Sedentary time is calculated from the SBR in 15 minute blocks, each day through summation of SB minutes. One participant was removed from data analysis as a result of non-compliance (insufficient wear time). Sleep time was not included in calculation of SB for any of the three measures. Convergent validity evidence was assessed with Repeated Measures ANOVA (α = 0.05). Known-group difference validity evidence was assessed using an independent t-test (α = 0.05) to examine differences in sedentary behavior between young and middle-aged adults. Older adults are typically more sedentary when compared with younger adults (Owen & Bauman, 1992).

Analysis/Results: There were no significant differences found in sedentary time between the SBR (mean ± SD; 522.53 ± 161.01), accelerometer (591.03 ± 148.58), and armband (610.22 ± 89.69) devices. A significant group difference was found (t₂₁=2.12; p = 0.04) in objective sedentary minutes measure with the armband between young (582.90 ± 66.13) and middle-aged adults (656.40 ± 101.00).

Conclusions: Our ANOVA results provide convergent validity evidence for measurement of SB time in adults with the SenseWear® armband device. As expected, there is a difference between minutes of SB between young and middle-aged adults which further supports known group difference validity evidence. These results suggest that the armband may be a useful measure of SB in adults.