Martin Luther King Jr. Day
Our office will be closed Monday, Jan 18th in observance of Martin Luther King Jr. Day. We will reopen at regular business hours on Tuesday, Jan 19th.
An Novel Approach to Quantifying Intra-Day Physical Activity Pattern
- Presented on May 29, 2013
The importance of regular physical activity for maintaining optimal health is well documented. Accelerometer-based measures of activity have proven very useful in recent years for quantifying activity behavior in a wide variety of populations, including adults aged > 65 years. However, few studies have focused specifically on activity patterns within a 24-hour period.
Purpose To develop a new tool for quantitative analysis of intra-day physical activity, based on accelerometer data.
Methods Activity data were collected from a convenience sample of 44 generally healthy and sedentary adults in 3 age groups: Young (22.4 ± 0.5 yrs, mean ± SE; 5 Male, 6 Female), Middle-aged (53.5 ± 3.04; 5M, 8F), and Older (74.7 ± 1.3; 11M, 9F). Volunteers wore an accelerometer (Actigraph GT1M, Pensacola, FL) around the waist during waking hours for 10 days. A custom Matlab (Mathworks, Natick, MA) analysis program was used to track the accumulation of activity counts during the day and parse them into quartiles based on both time of day and percentage of total daily counts expended. Output variables, averaged across days for each participant, included absolute and relative (normalized to Awake Time) counts accumulated in each quartile. The percent of total activity expended halfway through each day (ActT1/2, %) also was calculated.
Results Consistent with the sedentary status of these individuals, total daily activity was low and similar across age groups (Young = 215 ± 65 counts∙d-1∙1000-1, Middle-aged = 237 ± 64, Older = 195 ± 10; p=0.39). All groups had expended more than half of their activity by the mid-point of their day (ActT1/2 Young = 53.8 ± 5.9%, Middle-aged = 58.5 ± 5.9, Older = 60 ± 10.6), with a trend for Older to accumulate relatively more activity by this time point compared with the Young (p = 0.11).
Conclusions The tool developed here provides a new way to use common accelerometer data to examine physical activity behavior within the course of a day. This approach may prove useful in studies investigating intra-day variations in activity, or the effects of interventions designed to change activity behavior. Support: NIH R01 AG21094, UMass Commonwealth Honors College