Epoch length and its effect on physical activity intensity

• Published in: Medicine and science in sports and exercise 2010-05, Vol.42 (5), p.928-934
• Main Authors: Edwardson, Charlotte L, Gorely, Trish
• Format: Article
• Language: English
• Subjects: Actigraphy; Adolescent; Child; Confidence Intervals; Data Collection - methods; England; Female; Humans; Male; Motor Activity; Physical Exertion - physiology; Space life sciences; Time Factors
• ISSN: 0195-9131; 1530-0315
• DOI: 10.1249/mss.0b013e3181c301f5

Researchers have begun to investigate the issue of epoch length on children's physical activity using small sample sizes, and to date, no studies have been conducted in adolescents. The purpose of this study was to investigate the effect of different epoch lengths (5, 15, 30, and 60 s) on derived levels of physical activity in both a child and an adolescent sample. Three hundred and eleven children age 7-11 yr and 234 adolescents age 12-16 yr were asked to wear an accelerometer during waking hours for 7 d. The epoch was set at 5 s, and when data were downloaded, the activity counts were then reintegrated into 15-, 30-, and 60-s epochs. A significant epoch effect was seen for time spent in vigorous physical activity, light physical activity, and rest in the child and adolescent samples and moderate-to-vigorous and moderate physical activities in the child sample only. The Bland-Altman analysis showed reasonable agreement on moderate-to-vigorous, moderate, vigorous, and light physical activities and rest between 5- and 15-, 5- and 30-, and 30- and 60-s epochs in the child sample and between 5- and 15-, 5- and 30-, and 15- and 30-s epochs in the adolescent sample. A short epoch is strongly recommended for child and adolescent samples to obtain a "real" picture of young people's physical activity behavior and to prevent accumulation of counts reflecting the average activity level when longer epochs are used. Activity prevalence studies using epoch lengths of 5 and 60 s in a child or an adolescent sample should not be compared nor should 15- and 60- and 30- and 60-s epochs in an adolescent sample.