Evaluating the relationship between community food environment and obesity: What assessment method should we use?

Background: Neighborhood food environment may affect eating behavior and associated obesity. Different measurements are used when evaluating the relationship between types of food outlets and overweight/obesity. No study has compared density to distance. The measurement techniques selected may impact findings concerning community environment inequities, interventions and policy efforts.

Study design: This study evaluated if the distance to or density within (1/2 mile, 1 mile, and 3 miles radius) for supermarkets, convenience stores and fast food outlets altered environment-obesity interactions. Demographics and health behaviors were obtained from 2007-2008 Nevada Behavioral Risk Factor Surveillance System (BRFSS) sample (n= 2361). North American Industry Classification System was used to categorize food establishments. Geographic Information Systems spatial analysis and weighted multiple logistic regression were conducted. Distance and density of food environment attributes were individually evaluated.

Results: Adjusted odds ratios for being overweight/obese were non-significant for all three distances to food outlets and 1/2 mile and 1 mile density. However, for density (OR 0.52, C.I.0.30-0.82), but not distance, being overweight/obese was significantly more likely when 0 or < 3 supermarkets were present, compared to ≥ 3 stores within 3 miles. Furthermore, similar results for spatial distance and density were found for various other demographics, health-related behaviors and outcomes variables. Only older, male, non-smokers and diabetic characteristics were significantly linked to food environment and overweight/obesity.

Conclusions: Findings suggest that overweight/obese associated to health behaviors and health conditions were weakly or not associated with food environment regardless of distance or density measurement.