Multi-Level Predictors of Obesity and Diet in Multi-Ethnic Populations

Title:

Multi-Level Predictors of Obesity and Diet in Multi-Ethnic Populations

Author:

Tali Elfassy

Date:

2016

Executive Summary:

The obesity epidemic in the United States is a major public health issue. Poor socio-economic conditions such as poverty or living under financial stress may adversely influence health outcomes such as obesity, and often disproportionately impact ethnic minority populations. Further, individuals living in poverty are often exposed to environments that promote unhealthy diet and obesity. Such obesogenic environments are characterized by increased exposure to foods high in sodium and low in potassium which may also be independently associated with obesity. Yet these associations have not been adequately explored, especially among ethnic minority populations. The current dissertation seeks to explore: 1) whether individual-level socio-economic status, as measured by exposure to poverty, is associated with trajectories of BMI over 20 years, using data from the Coronary Artery Risk Development In Young Adults Study; 2) whether neighborhood-level socio-economic environment is associated with individual-level measures of obesity and diet quality (urinary sodium and potassium), using data from the Heart Follow-Up Study; and 3) whether diet quality (sodium and potassium) is associated with obesity, using data from the Hispanic Community Health Study/Study of Latinos. From fully-adjusted linear mixed effects regression models, sustained poverty was found to be associated with faster BMI growth in White men and women, slower BMI growth in Black men, and no association in Black women. Using multi-level regression models, low vs. high neighborhood socio-economic status was found to be associated with higher BMI and lower urinary potassium excretion among women but not men. Finally, from fully-adjusted linear regression analyses, higher sodium and lower potassium was found to be associated with higher BMI and waist circumference. Taken together, the results highlight the multi-level and multi-factorial nature of the relationships among socio-economic status, dietary nutrients, and obesity among diverse populations.