Understanding the Associations of the Objective Food Environment, Food Purchasing Behaviors, and Dietary Intake of Fruits and Vegetables in a Community-based Sample in a Low-income, Minority Neighborhood in Austin, Texas
The relationship between geospatial access to food stores and food-purchasing decisions remains unclear. This study examined the association between geospatial access to specific types of food stores and participant-reported place of purchase for fruits and vegetables (F&V). ^ Data from a cross-sectional survey of a low-income, predominantly Hispanic adult population in Austin, Texas from 2016 to 2017 were used for this study. Geographic Information Systems (GIS) were used to calculate geospatial food environment variables: availability (dichotomous: yes/no) of ≥1 supermarkets, small grocery stores, convenience stores, and non-traditional stores (farmer/mobile markets, farm-stands) within 1000m Euclidean buffers of each participant’s home and network-distance to the nearest store. Participants’ self-reported use of specific store type for F&V purchasing and daily F&V consumption. Logistic regression was used to estimate: 1) the effect of store-specific geospatial access on using that store type for purchasing F&V and 2) the effect of store-specific F&V purchasing on reported daily F&V consumption. ^ The prevalence of reported use of specific store types among participants (n=775) for F&V purchases was as follows: 98.2% for supermarkets, 23.4% for small grocery stores, 3.0% for convenience stores, and 20.0% for non-traditional stores. Long (vs. short) distance to the nearest small grocery store was significantly and inversely associated with F&V purchasing at these stores, controlling for socio-demographic covariates (odds ratio (OR)=0.76; 95% confidence interval (95%CI): 0.63-0.93). Availability of ≥1 non-traditional stores within 1000m of home (equivalent to a ten-minute walk) was significantly and directly associated with purchasing F&V at these stores, controlling for socio-demographic covariates (OR=1.49; 95%CI:1.02-2.18). No significant associations were observed between supermarket or convenience store geospatial access and F&V purchasing. ^ Among participants, 32.0% reported “low daily F&V intake”, 49.2% reported “moderate daily F&V intake”, and 17.0% reported “high daily F&V intake”. Reported F&V purchasing at non-traditional stores was significantly and directly associated with increased daily F&V intake, controlling for socio-demographic covariates (OR=2.25; 95%CI:1.45-3.48). No significant associations were observed between purchasing F&V at supermarkets, small grocery stores, or convenience stores and F&V consumption. ^ Results from this study suggest that the role of geospatial access to food stores on F&V purchasing is store-type specific. While for some store types (small grocery and non-traditional stores) geospatial availability appears to influence consumers F&V purchasing decisions, for others (supermarkets and grocery stores) location may not be as important of a contributing factor. Thus, current emphasis on increasing geospatial access to supermarkets and convenience stores may not be as effective as increasing that of small grocery stores and non-traditional retailers in an effort to facilitate healthier food purchasing decisions in low-income, predominantly-Hispanic areas. However, further research is needed to support this conclusion.^
O'Neil, Molly M, "Understanding the Associations of the Objective Food Environment, Food Purchasing Behaviors, and Dietary Intake of Fruits and Vegetables in a Community-based Sample in a Low-income, Minority Neighborhood in Austin, Texas" (2018). Texas Medical Center Dissertations (via ProQuest). AAI10790883.