The Relationship Between Insurance and Health Outcomes of Diabetes Mellitus Patients in Maryland: a Retrospective Archival Record Study

Background: Past studies examining the health outcomes of diabetes mellitus (DM) patients found that social determinants of health disparities were associated with variabilities in health outcomes. However, improving access to healthcare, such as health insurance, should mitigate negative health outcomes. The aim of the study was to explore the association between four types of health insurance, namely, Medicare Fee-For-Service (FFS), Medicare Managed Care (MC), Private FFS, and Private MC plans, and the health outcomes of DM patients, controlling for patients' social determinants of health.

Methods: This is a retrospective cross-sectional archival record study to explore the relationships between types of health insurance and health outcomes of DM patients who were at least 65 years old, or the elderly. Data was drawn from the 2012 Maryland Clinical Public Use Data and received an exempt status from our Institutional Review Board. Elderly Maryland residents with chronic DM were included in the study, resulting in a sample size of 43,519 individuals. Predictor variables were four types of insurance and health outcome variables were length of hospital stay (LOS), 30-day readmission, and end-stage renal disease (ESRD). Control variables included hospital characteristics, patient characteristics, and social determinants of health. Student's t-tests determined the statistical differences for the control variables between the types of insurance. Multiple hierarchical regression analysis was applied to test the association between insurance plans and LOS, while logistic regression analyses were applied to test the association between insurance plans with 30-day readmission and ESRD. Statistical significance was set at p < 0.05.

Results: t-test results indicated minimal statistical differences between the health statuses of patients enrolled in different insurance plans. After factoring out the control variables, regression analyses indicated that Medicare FFS patients had the worst outcome for LOS, 30-day readmission, and ESRD rates. Although patients on Medicare MC plans had lower LOS, 30-day readmission, and ESRD rates compared to those on Medicare FFS, patients enrolled in Private MC plans had the lowest odds of a 30-day readmission and patients enrolled in Private FFS had the lowest odds of an ESRD.

Conclusions: The data suggests that insurance plans were related to the health outcomes of elderly DM patients after considering their social determinants of health. Specifically, DM patients enrolled in managed care and private insurance plans had better health outcomes compared to those on Medicare FFS plans.