Influence of supermarket distance on diet and weight status among pregnant women

Objective: We assessed 357 supermarket service areas in a 14 county area in central NC as part of the Pregnancy, Infection and Nutrition prospective cohort study. Methods: Supermarket location was obtained using ReferenceUSA Business database and mapped using ArcGIS software. Distance to closet and density of supermarkets were derived for each of 1781 PIN participants. We assessed service areas by creating 0.25 to 4 miles buffers around each supermarket. Results: Distance to the closest supermarket ranged from less than 1 to 12 miles, on average women lived within 2 miles of a supermarket. 12% of women did not have a supermarket within 4 miles, a cutpoint that we previously found associated with poor dietary intake. Households at <185% of poverty had 1 fewer supermarket within the 4 mile buffer and were 0.4 miles further to the closest supermarket, compared to households >185% poverty. Women who lived greater than 4 miles had significantly higher BMI. Distance of > 4 miles was associated with a higher BMI after controlling for income, education, age, race, marital status, number of children, smoking, and accounting for clustering of observations. Living > 4 miles was also associated with higher % of carbohydrate and lower % of protein, but not fat, in the diet. After controlling for confounders and accounting for clustering of data, the relationship between living > 4 miles and carbohydrate was attenuated (significant at p <0.06) but the relationship remained significant for lower protein intake. Conclusion: Distance to supermarkets may influence diet and weight.