Effect of disease management programs on quality of care and disparities among Medicaid fee-for-service beneficiaries with diabetes

The prevalence of diabetes has more than doubled in the U.S. over the past 30 years and racial/ethnic disparities exist in diabetes prevalence and quality of care. Improvement of patient's ability to self-manage the disease is essential in reducing complications of diabetes and is a key principal of disease management programs (DM). DM programs also aim to empower patients to seek key components of comprehensive diabetes care including regular Hemoglobin A1C (HbA1C) testing, retinal exam, and LDL-C screening. The purpose of this study is: 1) to assess the impact of a disease management (DM) demonstration program on the receipt of comprehensive diabetes care among fee-for-service (FFS) Medicaid beneficiaries in California; and 2) to examine whether significant racial/ethnic disparities still exist in receipt of appropriate care after the implementation of the DM program. Claims data of California FFS Medicaid beneficiaries (ages 22 and older) with diabetes from two intervention counties and eight control counties between baseline years (2006 and 2007) and intervention year (2008) were analyzed. We used a difference-in-difference approach and logistic models to examine the receipt of semi-annual HbA1C test, annual retinal exam, and annual LDL-C test based on HEDIS measures between baseline and intervention years and between intervention and control groups. We further used this approach to compare receipt of these HEDIS measures among different racial/ethnic groups. We then calculated the predicted rates of the HEDIS measures and tested significance of the difference-in-difference using boot strapping methods for standard errors after controlling for age, sex, comorbidities, and county of residence. After the first year of the DM program, no statistically significant improvements were evident in the receipt of the three HEDIS measures. Furthermore, racial/ethnic disparities persisted in these rates of appropriate diabetes care. Specifically, observed lower rates of HbA1Ctesting, retinal exam and LDL-C testing among Latinos and African Americans vs. whites in the intervention group (e.g. 71% and 61% vs. 79% of HbA1C testing respectively) remained statistically similar pre- and post-intervention in comparison with the control group. Vendor-based DM programs may not be effective in improving the rates of appropriate diabetes care or addressing racial/ethnic disparities in the population after one year of implementation. Public and private efforts to improve diabetes patients' self-management skills should carefully examine whether vendor-based and patient-focused DM programs can improve quality of care in the short term. This is particularly important for Medicaid beneficiaries, a low-income and racial and ethnically diverse population.