Survival Analysis of Associations Between Health Insurance Status, Race, Yearly Household Income and Diabetes Incidence

Abstract

Objective We examined the effects of health insurance status, race, and income, on diabetes incidence in a racially and socioeconomically diverse sample of middle-aged American women.

Methods Using Kaplan-Meier survival curves and stratified Cox proportional hazards models, we evaluated the associations between insurance status, race, income, and diabetes incidence in a sample of 2,686 middle-aged women who participated in the Study of Women’s Health Across the Nation between the years of 1994 and 2008. Models were adjusted for covariates and stratified by age upon study entry.

Results Women with lower incomes tend to have higher risk for incident diabetes after adjusting for health insurance status, race, household income, BMI, and age upon study entry. Black/African American and Hispanic women may be at especially high risk for incident diabetes after adjusting for other covariates. Uninsured women were at significantly higher risk for diabetes incidence after adjusting for race, BMI, and age upon study entry, but insurance status is not found to affect diabetes incidence after additionally adjusting for household income.

Discussion Results for yearly household income and race reinforce findings from existing literature, and suggest the need for additional investigation of diabetes incidence among Hispanic women. Women without health insurance were at higher risk for diabetes when adjusted only for race, BMI, and age upon study entry. After adjusting the model additionally for household income, the association between insurance status and diabetes risk was non-significant, possibly due to confounding between household income and insurance status. Insurance status may be indicative of and confounded by several other socioeconomic, cultural, and environmental factors not explored in this study.