Perceived and objective neighborhood support for outside of school physical activity in South African children

BACKGROUND The neighborhood environment has the potential to influence children's participation in physical activity. However, children's outdoor play is controlled by parents to a great extent. This study aimed to investigate whether parents' perceptions of the neighborhood environment and the objectively measured neighborhood environment were associated with children's moderate-to-vigorous intensity physical activity (MVPA) outside of school hours; and to determine if these perceptions and objective measures of the neighborhood environment differ between high and low socio-economic status (SES) groups. METHODS In total, 258 parents of 9-11 year-old children, recruited from the South African sample of the International Study of Childhood Obesity, Lifestyle and the Environment (ISCOLE), completed a questionnaire concerning the family and neighborhood environment. Objective measures of the environment were also obtained using Geographic Information Systems (GIS). Children wore an Actigraph (GT3X+) accelerometer for 7 days to measure levels of MVPA. Multilevel regression models were used to determine the association between the neighborhood environment and MVPA out of school hours. RESULTS Parents' perceptions of the neighborhood physical activity facilities were positively associated with children's MVPA before school (β = 1.50 ± 0.51, p = 0.003). Objective measures of neighborhood safety and traffic risk were associated with children's after-school MVPA (β = -2.72 ± 1.35, p = 0.044 and β = -2.63 ± 1.26, p = 0.038, respectively). These associations were significant in the low SES group (β = -3.38 ± 1.65, p = 0.040 and β = -3.76 ± 1.61, p = 0.020, respectively), but unrelated to MVPA in the high SES group. CONCLUSIONS This study found that several of the objective measures of the neighborhood environment were significantly associated with children's outside-of-school MVPA, while most of the parents' perceptions of the neighborhood environment were unrelated.