The June issue of Health Affairs raised important questions about the current effort by states and the federal government to enroll individuals eligible for both Medicare and Medicaid into managed care, citing states’ and plans’ lack of experience in delivering integrated care to this vulnerable, complex population. Despite growing enrollment in Medicare Advantage and Medicaid managed care across states, researchers remind us that, according to 2011 MedPAC estimates, less than two percent of dual eligibles are enrolled in some type of integrated care program that coordinates at least some services across Medicare and Medicaid.

Policymakers stand to benefit from taking a closer look at the models of care and outcomes of plans that already offer this type of integration.  Avalere recently analyzed the outcomes of one such plan offered in Arizona.  The results suggest that the Mercy Care Plan’s model of care is successful in keeping people out of the hospital and lowering readmission rates relative to fee-for-service (FFS) dual eligible beneficiaries.

Background. Since its 1982 inception, Arizona’s Medicaid program has required all participants to enroll in a Medicaid managed care plan. In addition to participating in Arizona’s Medicaid managed care program, the Mercy Care Plan, fully administered by Aetna, also operates as a Medicare Advantage dual eligible special needs plan (DE-SNP), providing Medicare and Medicaid covered services including acute, long-term care, and pharmacy services to over 16,800 dual eligible beneficiaries. Approximately 23 percent of the Mercy Care dual eligible population receive long -term care, and 5 percent are developmentally disabled.

Arizona is unique in that it was the first to require all Medicaid beneficiaries enroll in a managed care plan in order to access benefits.  Moreover, all of Arizona’s Medicaid managed care plans must also serve as Medicare Advantage Dual Eligible Special Needs Plans (D-SNPs). This requirement has led to a high degree of dual eligible alignment relative to other states — one-third of all Arizona’s Medicaid managed duals are enrolled in plans that manage both their Medicare and Medicaid benefits. This has enabled plans like Mercy Care to gain special experience managing high-cost long-term care and senior populations under both programs.

The Mercy Care Model

Mercy Care uses interdisciplinary teams, including physicians, pharmacists, nurses and behavioral health specialists, to support intensive case management programs that are tailored to individual members.  Although benefits are financed from two different programs – Medicare and Medicaid – the plan has organized an integrated approach to simplify navigating the two systems.  Avalere compared Mercy Care Plan enrollees and Medicare FFS dual eligible beneficiaries across the country on four standardized measures: (1) access to preventive/ambulatory health services; (2) inpatient utilization (measured by hospitals days, discharges and length of stay); (3) emergency department utilization; and (4) all-cause readmissions.

Outcomes. When compared to Medicare FFS dual eligibles across the country, and adjusted to match the risk of the FFS dual eligibles, the total Mercy Care population had:

  • 43 percent fewer days spent in the hospital (per 1,000 months of beneficiary enrollment);
  • 31 percent fewer discharges (per 1,000 months of beneficiary enrollment);
  • 19 percent lower average length of stay;
  • 21 percent lower readmission rate;
  • 9 percent fewer ED visits (per 1,000 months of beneficiary enrollment); and
  • 3 percent higher proportion of members accessing preventive/ambulatory health services.

For each of the measures included in this analysis, results were computed in accordance with the 2012 Healthcare Effectiveness Data and Information Set (HEDIS) specifications. For the FFS population, Avalere used the CY 2008-2009 Medicare Standard Analytic Files (SAFs) including specific service dates (the services measured were provided in 2009; the 2008 data were used for risk adjustment). Mercy Care provided parallel results using their plan data.

This analysis was subject to some limitations. For example, while we captured both full and partial duals in the Medicare FFS dual eligible control group, the Mercy Care Plan only enrolls full duals. Although our results were risk-adjusted, because partial duals may have access to greater resources than full duals, the difference in populations may have had some impact on our findings. Also, to identify the Medicare FFS dual eligibles in the 2009 Standard Analytic Files, Avalere was limited to using indicators known to either undercount or overcount duals depending on the state. The analysts used a linear regression method to risk adjust the measures of inpatient utilization, ED visits and adults’ access to preventive/ambulatory health services. Alternative methods are frequently used to model these types of quality measures, particularly for patient-level risk adjustment. We explored these other model structures and found these alternatives did not provide any considerable gains in model fit or predictive power.

Despite these limitations, the study provides evidence of the promise of integrated care for dual eligibles. The analysis suggests that Mercy Care’s model of care resulted in decreased hospital utilization, as evidenced by lower inpatient and ED utilization and readmission rates, compared to the dual eligibles in Medicare FFS. Lower utilization in these areas translates to lower costs and may suggest an improvement in health outcomes for these beneficiaries. In addition, the Mercy Care population made slightly greater use of preventive services, worth noting because research has shown that dual eligibles demonstrate lower rates of preventive service use.

Given the disproportionate amount of spending already tied to dual eligibles, identifying ways to decrease costs while improving quality of care is critical. This study suggests that integrated approaches including the Mercy Care model should receive careful consideration among policymakers looking to improve health outcomes while reducing costs for dual eligible beneficiaries.

Aetna Inc. provided funding for this research. Avalere Health maintained full editorial control and the conclusions are those of the authors.