Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Health Outcomes and Policy Research

Research Advisor

Shelley White-Means, Ph.D.


Song Hee Hong, Ph.D. Nancy Mele, Ph.D. Esra Ozdenerol, Ph.D. Gregory Taff, Ph.D.


Health disparities Jackson Heart Study Cardiovascular disease African Americans M3 Ecosocial Framework


According to findings in the field of genetics, it has now been established that the impact of race on health is mostly a function of a people's lived experience, not their genetic make-up. Due to this, the role of race in the study of health outcomes and health disparities has traditionally been specified improperly in statistical models due to confounding with ethnicity. Additionally, the role of geography or the impact of neighborhoods on health is often not specified properly.

Thus, the author proposes a multilevel, multicultural, and multi-temporal ecosocial framework that explains the impact of neighborhoods on health and the role of race via ethnicity. This conceptual framework builds on the work of social scientists while advancing researchers' understanding of the role of neighborhoods, culture, history, and socially assigned race as it relates to health outcomes and the existence of health disparities.

Data collected from 5,314 participants in the Jackson Heart Study (in Jackson, Mississippi) were used to analyze and test the conceptual framework. The outcomes of interest are cardiovascular disease risk factors and hospitalization due to ambulatory care sensitive conditions are tested in multilevel models. Dependent variables included: total physical activity scores, hypertension status, diabetes status, hypercholesterolemia status, and hospitalizations due to ambulatory care sensitive conditions.

Among the study sample in a cross-sectional analysis, factors such as higher income, car ownership, and having a larger social network were statistically significant predictors of higher physical activity scores. In the prospective analysis of hypertension status, none of the variables designed to test the conceptual framework was statistically significant; however, social status is implicated via the statistical significance of college education in more advantaged neighborhoods. In the prospective analysis of diabetes status, historical factors (unfair medical treatment), cultural factors (church attendance), and neighborhood factors (fast food restaurant availability) were statistically significant predictors of diabetes. Having perceptions of serious neighborhood problems were statistically significant predictors of hypercholesterolemia. In the cross-sectional analysis of hospitalizations due to ambulatory care sensitive conditions, only increased age was associated with the dependent variable (likely due to small sample size).

Based on the dissertation research and findings across the literature, the multilevel, multicultural, and multi-temporal ecosocial framework shows promise and deserves further investigation. If the proposed ecosocial framework is refined and proven to be valid in future research, it can possibly help transform the practice and delivery of health care. Medical care can be modified to first assess and then develop an ethnically congruent treatment plan informed by an individual's social identity and their view of the world of health and health care. Policy makers and health care administrators can pass policies that address issues of quality, perceptions, trust, preferences, and health stewardship for individuals and populations. Finally, health care systems around the world can be transformed to provide ethnically responsive and relevant care that is locally flexible, persistently proactive, and able to use evidence-based interventions to eliminate health disparities.