Introduction: Accurate conceptualizations of neighborhood environments are important in the design of policies and programs aiming to improve access to healthy food. Neighborhood environments are often defined by administrative units or buffers around points of interest. An individual may eat and shop for food within or outside these areas, which may not reflect accessibility of food establishments. This article examines the relevance of different definitions of food environments.
Methods: We collected data on trips to food establishments using a 1-week food and travel diary and global positioning system devices. Spatial-temporal clustering methods were applied to identify homes and food establishments visited by study participants.
Results: We identified 513 visits to food establishments (sit-down restaurants, fast-food/convenience stores, malls or stores, groceries/supermarkets) by 135 participants in 5 US cities. The average distance between the food establishments and homes was 2.6 miles (standard deviation, 3.7 miles). Only 34% of the visited food establishments were within participants' neighborhood census tract. Buffers of 1 or 2 miles around the home covered 55% to 65% of visited food establishments. There was a significant difference in the mean distances to food establishments types (P = .008). On average, participants traveled the longest distances to restaurants and the shortest distances to groceries/supermarkets.
Conclusion: Many definitions of the neighborhood food environment are misaligned with individual travel patterns, which may help explain the mixed findings in studies of neighborhood food environments. Neighborhood environments defined by actual travel activity may provide more insight on how the food environment influences dietary and food shopping choices.