But Where Can We Buy an Ounce of Prevention? High Sprawl, Low Access, and Fresh Fruit and Vegetable Consumption

Background & Purpose: Despite clear evidence on the crucial dietary role of fresh fruits and vegetables (FFV), the vast majority of Americans do not meet dietary guidelines for consumption, and those least likely to meet guidelines include people of color and people living in poverty. Conservative estimates indicate that approximately 9% of U.S. residents, or 25.8 million people, live in areas of severely restricted access to FFV. While social workers have long incorporated an ecological approach to addressing problems in living, historically, food environments have not been considered a social work issue. However, given that differential access to healthful, nutritious foods is an issue of social justice, social workers are called to respond.

Study Objectives & Research Questions: Prior studies demonstrating the effects of the local food environment on FFV consumption, particularly with poor and minority populations, often examine a single dimension of food environments, such as distance from home to grocery store or neighborhood grocery store density. These single-dimension measures may not capture important components of community level accessibility. Sprawl is a recently devised measure of the food environment that incorporates multiple components of an area’s overall food accessibility, including community level variables in addition to proximity and density. The purpose of this cross-sectional study utilizing secondary data sources was to quantitatively examine the role of sprawl on FFV consumption and consider the influence of poverty level and race on the relationship. Research Question 1: Is an area’s sprawl level related to individual FFV consumption? Research Question 2. How do the influence of race and poverty contribute to understanding the relationship between sprawl score and FFV consumption?

Methods:  To answer these research questions, the study utilized the Behavioral Risk Factor Surveillance System (BRFSS) merged with Ewing’s Sprawl Scale, a reliable and valid measure of sprawl drawn from the scientific literature (n= 122,265).  The study employed measures of association (correlation, chi square analyses, t-tests, and ANOVAs where appropriate) and linear regression analyses to examine the association between sprawl and FFV consumption, including the effects of covariates.

Results: Linear regression analyses examined FFV consumption and sprawl by using the continuous number of FFV consumed daily (FFV score) as the dependent variable, and sprawl as predictor variable. The covariates in the model included income, education, region, race, sex, BMI, worry over food purchasing, health status, marital status, age. The study found that residents of more sprawling areas consumed significantly fewer FFV servings daily then did residents of less sprawling areas. While the impact of sprawl on FFV was very small in this model, it was still considered significant (R² = .037, F(26, 31129151) = 45442.03, p = 000.)

Conclusions & Implications: Our multivariate analysis demonstrated that sprawl continues to adversely affect FFV consumption patterns, net of demographic and economic covariates. Sprawl influences FFV consumption and further contributes to the disparity in FFV consumption between disadvantaged and advantaged communities. Future research should incorporate more sophisticated mutlilevel modeling techniques to examine how multiple characteristics in individual environments may simultaneously influence FFV consumption.