Date of Award


Document Type

Open Access Dissertation


Health Promotion, Education and Behavior


The Norman J. Arnold School of Public Health

First Advisor

Andrew T. Kaczynski


Youth obesity is a major public health concern due to an array of physical, social, and psychological health consequences. Residential location, often referred to as ‘place’, has continued to emerge as a key health determinant with studies showing that where a child lives impacts their health. Environments where it is easy for individuals to have low levels of physical activity – either by discouraging active behavior or promoting sedentary behavior – and easy for individuals to consume unhealthful foods – either by the limited availability of healthful foods or increased availability of unhealthy foods – have been coined ‘obesogenic’. Additional research is needed to improve measurement of obesogenic built environments and test associations with childhood obesity.

This study occurred in a southeastern US county (population:474,266) in 2013 and combined four unique datasets: 1) demographic, weight status, and addresses from all 3rd through 5th grade youth enrolled in a large southeastern school district (n=13,469), 2) detailed audit data on all public park facilities, 3) location of all food stores and restaurants , and 4) sociodemographic Census data. Global Moran’s Index and Anselin’s Local Moran’s I (LISA) were used to detect global and local spatial clustering of youth obesity, while residuals from a series of linear regression models were subsequently spatially analyzed and mapped to examine correlates of spatial clustering. Significant, positive global clustering (Index=0.04,p<0.001) was detected. In addition, LISA results showed that about 4.7% (n=635) and 7.9% (n=1,058) of the sample were identified as high and low obesity localized spatial clusters (p<0.01). Individual and neighborhood sociodemographic characteristics accounted for the majority of spatial clustering and differential patterns emerged by level of urbanization (e.g., urban, suburban, rural).

The second part of this study developed and tested an obesogenic built environment measure. Public parks (n=103) were identified and then scored using detailed audit data, while two commercial databases of food stores (n=395) and restaurants (n=717) were collected, categorized, and geocoded. Grocery stores that offered access to fresh produce were classified as ‘healthy’ while convenience stores, discount/drug stores, fast food restaurants, and fast casual restaurants with less access to fresh produce were classified as ‘less healthy’. Using GIS techniques, kernel density estimation procedures were used to create, normalize, and summarize separate raster (pixel) surfaces representing the nutrition and park environments. Using multilevel linear analyses, results showed that health promoting built environments, as indicated by availability of parks, presence of healthy food stores, and lack of unhealthy food outlets, were related to lower BMI z-score among youth (b=-0.25, p<0.05).

Identifying geographic areas that contain spatial clusters is a powerful tool for understanding the location of and contributing factors to patterns of childhood obesity. Environments that were classified as health promoting by providing greater access to public spaces to be active and places to consume healthier food options were related to lower youth obesity. This dissertation study integrated innovative methodology to analyze spatial patterns of youth obesity and develop and test a unique characterization of obesogenic built environments.