Using built environment characteristics to predict walking for exercise

Table 2 Built environment characteristics used one at a time to predict walking for exercise

Built Environment Characteristic	Median (Interquartile range)	Walking or not C-statistic (95% CI)	Walking time Proportion of variation explained
Density (thousands of residential units per km²)	177 (105, 256)	0.52 (0.49, 0.55)	0.0 %
Connectivity (mean block size in km²)	0.13 (0.04, 0.44)	0.49 (0.46, 0.51)	0.6 %
Sidewalks^a (km of sidewalk-lined streets)	12 (1, 33)	0.51 (0.47, 0.54)	0.1 %
Destinations (count within 1 km buffer)
Grocery stores	1 (0, 5)	0.50 (0.46, 0.52)	0.0 %
Restaurants	3 (0, 11)	0.50 (0.48, 0.53)	0.1 %
Retail stores	8 (0, 26)	0.51 (0.48, 0.54)	0.0 %
Grocery-restaurant-retail complexes	0 (0, 1)	0.52 (0.50, 0.54)	0.0 %
Offices	6 (1, 23)	0.52 (0.49, 0.54)	0.2 %
Office complexes	0 (0, 1)	0.52 (0.49, 0.54)	0.0 %
Banks	1 (0, 7)	0.49 (0.46, 0.52)	0.1 %
Churches	4 (1, 12)	0.48 (0.44, 0.51)	0.1 %
Schools	4 (1, 9)	0.50 (0.47, 0.53)	0.2 %
School-church combinations	0 (0, 1)	0.50 (0.48, 0.53)	0.2 %
Fitness centers	1 (0, 5)	0.49 (0.46, 0.52)	0.1 %
Parks (percent of 1-km buffer covered)	2 (0, 5)	0.51 (0.49, 0.54)	0.5 %

Notes: Built environment characteristics were measured within a one-kilometer airline buffer; ^a Sidewalk data were available only for King County, and were investigated as predictors of walking in this subset

ISSN: 1476-072X