Background and Purpose:
Sedentary behavior (SB) and physical inactivity are distinct constructs for which separate research and intervention paradigms may be warranted. To this end, we compared individual- and neighborhood-level risk factors of each among youth at risk of obesity.
Methods:
Data are from QUALITY, a cohort study of the natural history of obesity in Quebec, Canada. Baseline data were obtained in 2005-2008 when children were aged 8-10y (n=512 families). Activity level was measured using accelerometers at age 8-10y and again 2 years later at age 10-12y. At each time point, children were categorized as inactive if they did <60 min/day of moderate to vigorous physical activity (PA) and as excessively sedentary if they recorded <100 counts/min for
>
50% of the day. Children were required to have worn the device for at least 4 days and for at least 10 hours/day. Child-level factors included sex, sleep duration, and weekly frequency seeing friends; neighborhood-level factors included density of fast food outlets, convenience stores, and parks; school proximity, street connectivity, land use mix, disorder, social and material deprivation, and parental perceived safety. Separate logistic regression models were estimated for each of inactivity and excessive SB. We tested models using the identical set of baseline risk factors at both time points. Analyses were restricted to 413 children with complete data at age 8-10y, and to 283 children with complete data at age 10-12y. Models controlled for child’s obesity status, father and mother’s obesity status, and parental education.
Results:
At both time points, girls were 75% to 85% more likely to be inactive than boys, but were equally likely to be excessively sedentary as were boys. Also at both time points, each additional weekly outing with friends reduced the likelihood of being sedentary by 20%, but did not reduce the likelihood of being inactive. Only area-level disorder was associated with being excessively sedentary, and only in 10-12y olds; in contrast, several factors increased the likelihood of being inactive, including area deprivation at age 8-10y (OR: 1.7; 1.0-3.0) and perceived lack of safety at age 10-12y (OR: 2.8: 1.1-6.3). Moreover, the likelihood of being inactive decreased by 24% for each quintile increase in land use mix. Although obesity status in children was strongly associated with outcomes in all models, other determinants were unaffected by its inclusion in the models.
Conclusions:
Our findings suggest that physical inactivity and sedentary behavior are driven by largely distinct paradigms. Each of these may be impacted through increases in light PA. Although interventions need to target all spheres of influence, reducing physical inactivity may be more effectively mediated by features of the built environment, while leveraging social and peer groups may be more effective to reduce sedentary behaviors.
Do Neighborhoods with Highly Diverse Built Environment Exhibit Different Socio-Economic Profiles as Well? Evidence from Shanghai
