Using GIS to map the ‘food environment’

Yingru Li’s findings suggest that policymakers and planners need to improve community food environments in low-income, minority communities

When the average person hears about GIS, he or she is probably thinking about getting directions from one point to another without getting lost (that’s actually GPS). For Dr. Yingru Li, GIS—geographic information system—is a sophisticated tool for studying childhood obesity.

Dr. Li, now a faculty member in the Sociology Department at the University of Central Florida in Orlando, until recently was an assistant professor in Auburn University’s Department of Geology and Geography. She remains a member of the Obesity Working Group.

While she was at Auburn, Li focused on one county in Alabama’s Black Belt, a crescent-shaped swath that cuts across the state and includes some of the poorest counties in the U.S. The rate of obesity is consistently higher among those from lower socioeconomic backgrounds and from racial or ethnic minorities, such as the residents of the Black Belt. The region’s residents are overwhelmingly poor and African American.

She assessed the weight status of 613 African American students in kindergarten through fifth grade at four schools in the county, determining that 42.1 percent of the students were overweight (much higher than the national average of 30.6 percent). Li next looked at the “food environment”—nearby convenience stores, fast food outlets, grocery stores and full-service restaurants—because earlier studies suggest that children’s eating behaviors and weight might be influenced by the their surroundings, in particular the availability of healthy choices.

In fact, in the last two decades the increased availability of cheap, convenient, energy-dense food has been considered as contributing to soaring obesity rates in the U.S. Typically, convenience stores and fast food outlets sell more high-calorie and unhealthy food, while supermarkets and full-service restaurants offer more healthy and nutritious choices. Previous research has found that poor communities tend to have more fast food restaurants within walking distance than communities with higher socioeconomic status.

Li hypothesized that children living in healthier food environments are less likely to be overweight or obese, and that the associations between food environments and children’s weight status could change if socio-demographic characteristics and school effects were controlled.

Li identified 22 convenience stores, seven fast food outlets, five supermarkets and 11 full-service restaurants in the county. Food outlets and children’s home addresses were geocoded, and distances from stores to children’s homes were obtained using a system called ArcGIS. Store sizes were measured on Google Earth.

Then, she developed an overall index for measuring and assessing the food environment around each child. This included standardizing the probability that a child would patronize a particular food outlet. Four composite indexes were determined, one each for convenience stores, fast food outlets, supermarkets, and restaurants. Two multilevel regression models were created to further examine the association between food environments and children’s weight status.

The first model showed only the influence of the environment on children’s weight. As expected, the near availability of convenience stores, supermarkets, and full-service restaurants all were significantly association with children’s BMI. The model illustrated that children have a higher risk of being overweight or obese if their families patronize convenience stores more often, and are more likely to have a healthy weight if there is a supermarket available nearby for shopping. Unexpected finding were that having nearby fast food outlets was not related to children’s weight, and that children with a higher probability of patronizing full-service restaurants were more likely to be overweight or obese.

In a second model, adding socio-demographic factors and school effects caused significant changes in the relationship between children’s weight and the four types of food outlets.

Li says her findings suggest that policymakers and planners need to improve community food environments in low-income, minority communities. In addition, parents and school personnel should pay more attention to the kinds of food that are available in the neighborhood.