A previous study showed that 93% of adults with diabetes in the U.S. population have some form of health insurance, and 52% have multiple sources of insurance (1). An important but unanswered question is whether the source and nature of health insurance influences access to care, quality of care, and health status and outcomes for patients with diabetes. To investigate this issue, we analyzed data from the third National Health and Nutrition Examination Survey, which included a representative national sample of 721 adults aged ≥65 years with diagnosed diabetes who had Medicare health insurance coverage. Structured questionnaires and clinical and laboratory assessments were used to determine whether they had private health insurance and the frequencies of physician visits, screening for diabetes complications, and treatment and control of hyperglycemia, obesity, proteinuria, hypertension, and dyslipidemia.

Of all patients, 26% were not covered by private health insurance. These individuals were more likely to be of minority race and ethnicity, to have less education and lower income, and to be covered by Medicaid. They were similar to those with private insurance regarding health care access and utilization, including having a primary source of ambulatory medical care (97 vs. 98%), having a primary physician at this source (92 vs. 95%), attending four or more physician visits in the past year (81 vs. 73%), using diabetes therapy (78 vs. 74% using insulin or oral agents), taking at least three prescription medications (66 vs. 65%), and receiving blood pressure screening in the previous 6 months (91 vs. 94%). However, lower proportions took two or more insulin injections per day (31 vs. 56%), self-monitored their blood glucose (23 vs. 42% for insulin users), had an eye examination in the past year (63 vs. 75%), had their cholesterol checked (65 vs. 80%), or were treated for their hypertension (72 vs. 87%) or dyslipidemia (25 vs. 50%).

Despite these differences in indicators of health care access and use, there were few differences between those without and with private insurance in clinical indicators of health status, including blood glucose and blood pressure control, obesity, proteinuria, and lipid levels. These indicators of health status included HbA1c >7.0 (56 vs. 54%), microalbuminuria or clinical proteinuria (43 vs. 38%), BMI >30 kg/m2 (men 20 vs. 18%, women 34 vs. 40%), undiagnosed hypertension (≥140/90 mmHg, 16 vs. 8%), diagnosed but uncontrolled hypertension (43 vs. 45%), total cholesterol ≥240 mg/dl (35 vs. 33%), LDL cholesterol ≥130 mg/dl (56 vs. 46%), HDL cholesterol >45 mg/dl (47 vs. 58%), and triglycerides ≥200 mg/dl (36 vs. 41%). Adjustment for sociodemographic differences (age, sex, race, education, and income) and for Medicaid coverage status did not alter these findings.

In conclusion, there are some differences in sociodemographic characteristics and in health care access and use between diabetic patients who have and do not have private health insurance to supplement their Medicare coverage. However, there are few differences in clinical health status between these two groups, and the health status of patients with diabetes who are covered by Medicare does not appear to be improved by having private health insurance.

Harris MI: Racial and ethnic differences in health insurance coverage of adults with diabetes.
Diabetes Care

Address correspondence to Dr. Maureen I. Harris, 6707 Democracy Blvd., Room 695, Bethesda, MD 20892-5460. E-mail: mh63q@nih.gov.