Trends in domain-specific physical activity and sedentary behaviors among Chinese school children, 2004–2011

BACKGROUND: Dramatic increases in child overweight have occurred in China. A comprehensive look at trends in physical activity and sedentary behaviors among Chinese youth is needed. The study aimed to examine trends in domain-specific physical activity and sedentary behaviors, explore mean and distributional changes in predicted behaviors over time, and investigate how behaviors vary by residence. METHODS: Using 2004–2011 China Health and Nutrition Survey data, adjusted means for MET-hours/week from physical activity and hours/week from sedentary behaviors were determined for school children (6–18 years), stratifying by gender, age group, and residence. Physical activity domains included in-school physical activity, active leisure (out-of-school physical activity), active travel (walking or biking), and domestic activity (cooking, cleaning, and child care). For each physical activity domain, the MET-hours/week measure was determined from the total weekly time spent (hours) in domain-specific activities and corresponding MET-values using the Compendium of Energy Expenditures for Youth. Sedentary behaviors included television, computer use, homework, and other behaviors (board games, toys, extracurricular reading and writing). For each sedentary behavior, the hours/week measure was determined from total weekly time spent in specific sedentary behaviors. Residence groups included megacities (population ≥ 20million), cities/towns (300,000 ≤ population < 20million), and rural/suburban areas (population < 300,000). Repeated measure linear mixed and quantile regression models were used to predict adjusted means. RESULTS: Little change in physical activity behaviors occurred over time, with the exception of statistically significant trends toward increased domestic activity among male children (p < .05). Across all gender and age groups, statistically significant trends over time toward an average increase in computer use were seen (p < .01); these increases were largely driven by those ≥50th percentile on the distribution. Children living in megacities (versus rural areas) reported higher levels of physical activity, homework, and computer use. CONCLUSIONS: Intensified, systematic intervention and policy efforts promoting physical activity and reducing sedentary behaviors among children are needed.