International obesity and socioeconomic status: relative vs. absolute

Existing literature highlights the developing burden of hypertension and obesity on the health infrastructures of middle-income countries; however, the influence of social and positional forces on these disease likelihoods have been poorly explored, potentially limiting intervention success. To aid future research and policy development, this dissertation examines data from six low-middle-income countries in the World Health Organization’s Study of AGEing and Global Health (SAGE) dataset. We explore four distinct statistical methodologies: Receiver Operator Characteristics (ROC), Spatial Analysis Techniques (Moran’s I, Hot Spot Mapping, and Geographically Weighted Logistic Regression), Bayesian Markov Chain Monte Carlo, and Hierarchical Structural Equation Modeling in order to evaluate the influence of geographic location, wealth, and social capital on disease prevalence and burden. These methodologies have been applied to chronic disease studies throughout developed, industrialized economies, but sparsely applied to low-middle income countries. Each method incorporates various aspects of "health in place" to examine the absolute and relative influences of being hypertensive or obese in a low-middle-income context. Results demonstrate that the Waist-Height Ratio metric operates equally well as Body Mass Index (BMI) and is a superior metric in public health applications. We find evidence of geospatial clustering of disease and a significant association between both absolute and relative wealth and increased body mass. Finally, we show that the influence of social cohesion on wealth and obesity is country dependent and suggest that social cohesion is an important factor for policy development and intervention planning. Given these results, we discuss the implications and potential applications of each methodology within the context of existing literature and policy for each country. We conclude with an examination of where the existing health frameworks may benefit from the inclusion of our methodologies in the development of data, statistical training, and policy formation to further develop chronic disease strategies.