Essays on the Heterogeneous Effects of Medicare Advantage

University Of Michigan At Ann Arbor, Ann Arbor MI

Investigators

Abstract

PROJECT SUMMARY. Over the past several years, there has been a dramatic change in the structure of Medicare. While beneficiaries historically obtained coverage through a government-administered version of the program, i.e., ?traditional Medicare,? a rapidly growing number receive coverage through private Medicare Advantage (MA) plans. MA plans can differ substantially from traditional Medicare in terms of their benefits, provider choices, and other care management features. Much of the growth in MA has been the product of federal policy, and policymakers continue to debate what the future role of private plans in Medicare should be. To inform this discussion, researchers have generally examined the effect of MA on average beneficiary outcomes. However, optimal policymaking requires a more nuanced understanding of the effects of MA. This dissertation will provide a deeper understanding of how the effects of MA enrollment can vary by beneficiary group and plan. The first aim is to evaluate MA?s effect on utilization and quality of care among dual eligibles, i.e., beneficiaries who also receive Medicaid coverage. Dual eligibles are low income and tend to have significant health needs, both factors that may mediate the effect of MA. This analysis will rely on advanced econometric techniques, taking advantage of a discrete increase in average MA benchmark payments to identify potentially exogenous increases in enrollment. It will then relate this change in enrollment to beneficiary outcomes. Policymakers have attempted to nudge dual eligibles into MA in the past, and these results will help them determine whether to further promote or constrain MA enrollment among this population. The second aim is to identify the extent to which MA star rating measures of plan performance capture differences in plan quality, as intended, versus differences in enrollee characteristics. Relying on many of the same variables available to federal administrators, this analysis will first adjust star ratings for case-mix and compare the results to current star performance. Next, the study will consider whether any remaining variation in star ratings reflects additional differences across enrollee populations that are more difficult to observe, such as knowledge of health care. It will do so through a novel application of a ?switcher analysis,? an approach used in other contexts to identify selective enrollment into insurance. These results will help policymakers and beneficiaries gain a deeper understanding of the value of star ratings as a measure of plan quality. They will also help administrators decide whether to enhance the current star rating case-mix adjustment system. This research will evaluate private Medicare plans and therefore aligns well with AHRQ?s interest in ?examining innovative market approaches to care delivery and financing.? Notably, this research will focus on several AHRQ priority populations. For example, dual eligibles are low-income by definition and include a disproportionate share of minorities and individuals with multiple chronic illnesses.