Skip to main content
ARS Home » Research » Publications at this Location » Publication #187219


item Paul, David
item Kramer, Matthew
item Rumpler, William
item Stote, Kim
item Clevidence, Beverly
item Harris, Gabriel
item Baer, David

Submitted to: Federation of American Societies for Experimental Biology Conference
Publication Type: Abstract Only
Publication Acceptance Date: 3/1/2006
Publication Date: 4/30/2006
Citation: Paul, D., Kramer, M.H., Rumpler, W.V., Stote, K.S., Clevidence, B.A., Harris, G.K., Baer, D.J. 2006. Variability in the physical activity of free-living adults. Federation of American Societies for Experimental Biology Conference, San Francisco, CA. FASEB J 20:A588.

Interpretive Summary: n/a

Technical Abstract: To determine daily and weekly variability in physical activity (PA), 47 adults wore activity monitors (Actigraph Model 7164) for four, seven day periods spread out over six months. Daily samples were considered acceptable for analysis if the estimated adherence was a minimum of 14 hrs/day. The within- subject day-to-day coefficient of variation (CV) was 31.5%, compared to 14.3% for week-to-week. The between-subject CV was 31.1%. Overall, the greatest amount of variance was explained by day-to-day effects (59.3%), when compared to subject-to-subject (36.7%) and week-to-week effects (4.1%). Despite the high daily variability in the PA of individual subjects, weekly PA is relatively constant. There is high variability between subjects, however. These results indicate that, in order to capture the habitual PA of a population, investigators must capture many days of data in a large number of subjects.