Examining the Impact of Community Food Environment on Obesity Prevalence in Appalachia from an Ecological Perspective

Introduction: Obesity is a critical health issue that is more pronounced in the Appalachian region. While the region has experienced some growth in the past 50 years, economic deprivation and isolation continue to impact health outcomes by limiting access to resources. The built food environment which refers to the presence or absence of various types of food stores has been shown to be related to rates of obesity but its relationship to rates obesity in Appalachia is understudied. The association between community food environments and obesity may be especially important for this region due to the prevalence of socioeconomic disadvantaged and rural land use. In this study, we examine how the food environments impact obesity in the Appalachian Region from an ecological perspective. We hypothesized that county-level change in numbers of grocery stores, supercenters, convenience stores, and fast food restaurants would be associated with obesity rates.

 

Method: To test these hypotheses, we used data collected as part of the Behavioral Risk Factor Surveillance System (BRFSS), the American Community Survey (ACS), and the Food Environment Atlas (FEA). Data analysis included 412 counties from all 13 states within the Appalachian region. To test our hypothesis we conducted descriptive spatial analysis using mapping, bivariate analysis, and parameter estimates in multivariate regression models. Maps were created in QGIS 2.12.3. Data pre-processing and analyses were conducted in R 3.2.2.

 

Results: The average age-adjusted obesity rate of Appalachian region counties was 33% in 2012, among which Breathitt County, KY was the highest (43%). From 2007 to 2011, while there was an increase of supercenters per capita in the Appalachian region (M(SD)= 7.47 (28.18)), there was an overall decrease of food availability, including the decrease of grocery stores per capita (M(SD)= -11.89 (28.88)), the number of fast food restaurants per capita (M(SD)= -3.28 (27.68)), and the number of convenience stores per capita (M(SD)= -4.77(20.61)). The change of grocery stores rate from 2007 to 2011 was significantly associated with the obesity rate in 2012 after controlling for obesity rate in 2006 = -.012(.004), p < .01). That is to say, for every one unit of decrease of per capita grocery stores at county level from 2007 to 2011, we could expect a 1.2 percent increase of the county-level obesity prevalence in the Appalachian region.

 

Implication: The results of this study highlighted the disparities in obesity prevalence in the Appalachian region and its association to the food environment. These finding suggest that macro-level interventions to attract the development of grocery stores in Appalachia maybe an important part of public health strategies to address obesity disparities in the region.  It also highlights the shift away from grocery stores to supercenters, a change that may impact individual’s food choices and behavior. The dynamic nature of the food environment, demonstrated in this study, and its relationship to obesity makes it an important area for further investigation and intervention development to address this critical health disparities in this region.