Do Pedestrian Supportive Environments Assessed Using Virtual Audits Predict Childhood Obesity in the Fragile Families and Child Wellbeing Study?

Gina Lovasi, Columbia University
Steve Mooney, Columbia University
Nicolia Eldred-Skemp, Columbia University
Kathryn Neckerman, Columbia University
Andrew Rundle, Columbia University
Julien O. Teitler, Columbia University

The rising prevalence of childhood overweight has added to early elevations in cardiovascular risk factors. One strategy to prevent children from becoming overweight is to modify the local environment. The proposed analyses will use data from the Fragile Families and Child Wellbeing Study, which recruited 4,898 families from 20 large US cities in the years 1998-2000. We assembled Google Street View virtual audit data to characterize neighborhoods (1km buffers) around home address for respondents living in four of the 20 cities (Philadelphia, New York, Detroit, and San Jose). Previous research suggests that measures of the pedestrian environment can be reliably assessed from Google Street View, and this paper will describe the relationship of the following selected measures to child BMI z-scores at ages 5 and 9: abundance of shade trees, presence of convenient pedestrian crossings, conditions of available sidewalks. The statistical analyses will include descriptive tests and generalized estimating equations.