Save the Date!
ActiGraph Digital Data Summit 2021November 4 - 5 | Sign Up for Event Updates Now
Using Accelerometers to Assess Temporal Changes in Older Adults’ Sedentary Time Following an Intervention
- Published on 2011
Introduction Accelerometer data provide reliable, valid, and stable measures of physical activity, and also allow for the estimation of sedentary time. Furthermore, as data are date and time-stamped, the outcomes of interventions targeting these behaviors can be explored beyond simple average of change.
Methods Stand Up For Your Health  was a quasi-experimental (pre-post) study that assessed the feasibility of an intervention to reduce and break up sedentary time in older adults. Participants (n=59; 75% female; age range 60-92 years) wore a GT1M accelerometer for 14 days during waking hours to derive sedentary time (<100 counts per minute (cpm); expressed as minutes and percentage of wear time) and breaks in sedentary time (> 100cpm, for at least one minute). The intervention consisted of one face-to-face goal setting consultation (delivered after seven days of accelerometer wear) and one individually-tailored mailing providing feedback on accelerometer-derived sedentary time. Intervention outcomes (participants’ changes in total sedentary time and in breaks) were assessed in three ways: pre- to post-intervention changes (via paired t-tests); day-by-day changes (via linear mixed models); and, hour-by-hour changes (via Wicoxon signed rank tests).
Results From pre- to post-intervention there was a significant reduction (baseline level [SD], mean change [95% CI]) in sedentary time (71.1% [8.9], -3.2% [-4.18 to -2.14], p<0.001), and a significant increase in the number of breaks in sedentary time per day (87.8 [14.0], 4.0 [1.48 to 6.58], p=0.003). Compared to the day prior to the intervention, participants had lower sedentary time (-4.6% [-6.96 to -2.36]) and more breaks (8.3 [2.5 to 14.1]) on the day immediately following the intervention. While significantly greater reductions in sedentary time were made in a consistent pattern during each hour after 10am, participants mostly broke up their sedentary time between 7pm and 9pm.
Discussion and Conclusion Examining the temporal patterns of the accelerometer data extended the understanding of the intervention effects beyond the pre-post mean change. These approaches have applicability to studies using other monitoring devices that capture time-stamped data.
References  Owen N, Ekelund U, Hamilton M, Gardiner P, Dunstan D. Sedentary behavior in adults: longitudinal, experimental, and intervention evidence. J Phys Act Health, 2010; 7(Suppl 3): S334-336.
ICAMPAM- Glasgow 2011