Date of Award


Degree Type


Degree Name

Doctor of Philosophy


School of Public Health

First Advisor

Dennis Thombs

Second Advisor

Subhash Aryal

Third Advisor

Emily Spence-Almaguer


Background: In the United States an estimated 10.9 million people aged 65 and older are living with diabetes mellitus. Previous research has found that demographic and socioeconomic factors, health status, health behaviors, and interventions are potential determinants of diabetes outcomes. Recent studies have found that limited health literacy may be a potential new determinant of diabetes outcomes. Limited health literacy is common in underrepresented and marginalized groups such as people with low economic status, low educational attainment, new immigrants, the elderly, racial and ethnic minorities, and patients with chronic conditions. Older adults are at higher risk of developing chronic complications from improper diabetes self-management and self-care.

Purpose: In this dissertation, we explore the association between health literacy and diabetes outcomes and self-management behaviors among older persons with an imputed health literacy score derived from demographic information.

Methods: This study is a secondary analysis of existing cross-sectional data from the National Health & Nutrition Examination Surveys (2009-2010), in the United States. The final sample consisted of 779 participants who were ≥55 years and older and diagnosed with DM by self-report or through bio-marker laboratory testing. The predictor and outcome variables were DAHL proxy health literacy score, categorized as adequate (HL>76), marginal (HL 63-75), and inadequate (HL<62); biomarker test (HbA1c, FBG, OGTT); and known indicators of proper diabetes management. The proxy HL score used in the present study was derived from four - iv - demographic variables (gender, age, education, and marital status). Hierarchical multiple regression, hierarchical logistic regressions, and linear regressions were performed to test the hypotheses and to determine the strength of the relationship between the proxy HL scores diabetes outcomes and self-management behaviors. Results: 1 in 3 participants were not aware of their DM positive status. Hierarchical regression analysis revealed that after controlling for the effect of covariates, health literacy score was not a significant predictor of glycohemoglobin score, (β=−.003; Sig F change; p = 0.66). Proxy health literacy score contributed to the prediction of fasting blood glucose in Block 2, and made a unique contribution (β=−.392; Sig F. change: p = .028) to the full model. The fully adjusted hierarchical regression models for HbA1c showed that HL did not add any variability to the model. The fully adjusted FBG model was not statistically significant. After controlling for covariates, we found that in separate hierarchical logistic regression, health literacy level, was not associated with predicting the odds of the eight indicators of proper diabetes management. Furthermore, the R2 change attributed solely to the addition of health literacy level did not exceed 2.7% for any of the logistic regression models.

Conclusion: The present study supports previous findings that found no association between HbA1c, diabetes self-management behaviors and health literacy. This study found that the characteristics of individuals who were not aware of their diabetes status was higher among subjects that were younger, more educated, higher socioeconomic status and not married. More studies are needed to examine factors associated with diabetes self-management behaviors that take into account individual health literacy, diabetes knowledge, and create targeted initiatives that decrease the risk factors associated with diabetes among the aging population.