Despite growing research supporting the impact of the built environment on active school transport (AST), distance persists as the most powerful predictor of walking and biking to school. There is a need to better understand how environmental features interact with distance to affect AST, and whether the influence of environmental factors persist across different distance thresholds. Multilevel models using cluster-robust standard errors were used to examine for interactions between objectively measured macroscale environmental features and several reported distances from home to school (up to ¼, ¼ up to ½, ½ up to 1, 1+ miles) on the likelihood of parent reported AST for children grades 3-8 (n = 2751) at 35 schools who completed a Safe Routes to School Parent Survey about Walking and Biking to School (SRTS Parent Survey). An interaction between both intersection density and food-related land use with distance was observed. The likelihood of AST decreased as intersection density and distance increased (i.e., 31.0% reduced odds among those living within ¼ mile compared to 18.2% using ½-1-mile criterion). The likelihood of using AST were reduced as food-related land use and distance increased (i.e., 43.67% reduced odds among those living under ¼ mile compared to 19.83% reduced odds among those living ½-1 mile). Programs and infrastructure improvements focused on overcoming environmental barriers to promote AST may be most effective when targeting neighborhoods within ¼ mile of schools.
Keywords: active school transport; built environment; walking and biking to school.