Neighborhood environment and children's physical activity and body mass index: evidence from military personnel installation assignments.

BACKGROUND: The majority of existing studies use observed, rather than experimental or quasi-experimental, variation in individuals' neighborhood environments to study their influence on body weight and related behaviors.
PURPOSE: This study leverages the periodic relocation of military personnel to examine the relationship between neighborhood environment and children's physical activity (PA) and BMI in military families.
METHODS: This study utilizes data on 12- and 13-year-old children from the Military Teenagers Environments, Exercise, and Nutrition Study (N=903). Multivariate regression models are estimated, separately for families living on- and off-post, to examine the relationship between parents' perceptions of the neighborhood environment, measured using the Neighborhood Environment Walkability Scale-Youth Version (NEWS-Y), and children's self-reported PA and BMI.
RESULTS: Different features of the neighborhood environment were significant for off- versus on-post families. For children living off-post, a 1 standard deviation (SD) increase in the proximity-to-recreational-facilities subscale was associated with 16.5 additional minutes per week (p<0.05) of moderate PA (MPA), but street connectivity had a significant negative association with vigorous activity. For children living on-post, a 1 SD increase on the crime safety subscale was associated with 22.9 additional minutes per week (p<0.05) of MPA. None of the NEWS-Y subscales were associated with children's BMI.
CONCLUSIONS: Efforts to increase children's PA in military families should take into account that different aspects of the neighborhood environment matter for children living on- versus off-post.