Risk of Diabetes Mellitus Among Medicaid Beneficiaries in Hawaii

INTRODUCTION: Medicaid is the largest primary health insurance for low-income populations in the United States, and it provides comprehensive benefits to cover treatment and services costs for chronic diseases, including diabetes. The standardized per capita spending on diabetes by Medicare beneficiaries enrolled in the fee-for-service program in Hawaii increased from 2012 to 2015. We examined the difference in odds of diabetes between Medicaid and non-Medicaid populations in major racial/ethnic groups in Hawaii. METHODS: We used data from 2013 through 2015 from the Hawaii Behavioral Risk Factor Surveillance System in this cross-sectional study to compare the difference in risk for self-reported diabetes between Medicaid (n = 1,889) and non-Medicaid (n = 17,207) beneficiaries. We used multivariate logistic regression models that could accommodate the complex sampling design to examine the difference in odds of diabetes between the 2 populations. RESULTS: In Hawaii, the Medicaid population was younger, was less educated, had more health impairments, and was more likely to be obese and Native Hawaiian/Other Pacific Islander (NH/OPI) than the non-Medicaid population. The unadjusted prevalence of diabetes in the Medicaid population in Hawaii was higher than that for the non-Medicaid population (10.3% vs 8.9%, P = .02). After adjusting for confounding variables, the odds of diabetes in the Medicaid population was still significantly higher than those in the non-Medicaid population (adjusted odds ratio [AOR] = 1.75; 95% confidence interval [CI], 1.33–2.31). Adjusted analysis stratified by race/ethnicity showed that non-Hispanic Asian (AOR = 2.23; 95% CI, 1.31–3.78) and NH/OPI (AOR = 3.17; 95% CI, 1.05–9.54) Medicaid beneficiaries had significantly higher odds of diabetes than their non-Medicaid counterparts. CONCLUSION: The odds of diabetes was significantly higher among the Hawaii Medicaid population than among the non-Medicaid population. Diabetes prevention programs should address the challenges and barriers that the Medicaid population faces. Our findings can be used to promote culturally competent diabetes education programs.