Abstract
PURPOSE: To determine the annual rate at which physical activity changes in girls during middle school using both objective and self-report measures of physical activity. METHODS: Participants were sixth- and eighth-grade girls from the control schools in the Trial of Activity for Adolescent Girls (TAAG). Random cross-sectional samples initially were drawn from sixth-grade girls (n=786) and 2 years later from eighth-grade girls (n=1545). A cohort of 501 girls was in both the sixth- and the eighth-grade samples. The girls wore an accelerometer for 6 days and completed the 3-Day Physical Activity Recall. Data were summarized using 3.0-, 4.6-, and 6.5-metabolic equivalent cutpoints for accelerometry and self-reported physical activity. Analyses were performed using repeated-measures analysis of variance in PROC MIXED. RESULTS: More than 40% of the girls were white, approximately 20% were African American, and 20% were Hispanic. The annual percent decrease in physical activity in the cross-sectional sample was approximately 4% (-1.76 min moderate-to-vigorous physical activity/day), using accelerometer data. The percent decrease in physical activity based on self-report data was higher, 6% to 13%, depending on the physical activity variable. Declines tended to be larger in African American girls, but the ethnic differences were not statistically significant. CONCLUSIONS: Based on comparisons of cross-sectional samples of sixth- and eighth-grade girls, objectively measured physical activity declined at a rate of 4% per year.