Effects of Medicare Advantage Plan Rebates on Enrollment and Beneficiary Health

Tuesday, June 24, 2014: 3:40 PM
LAW B1 (Musick Law Building)

Author(s): Lauren Hersch Nicholas

Discussant: Joseph P Newhouse

Since 2006, the Centers for Medicare and Medicaid Services (CMS) have used a bid system to determine payments to Medicare Advantage plans.  CMS sets annual, county-level payment benchmarks, the maximum monthly amount they will reimburse plans, who “bid” the amount it costs them to provide standard Medicare benefits.  When bids are below the benchmarks, plans receive a rebate covering part of the difference between benchmarks and bids, which must be used to provide additional benefits or reduced premiums to enrollees.  Historically, the benchmarks have not been closely tied to local Fee-for-Service spending.  I use the variation in benchmark-induced benefit generosity across counties and over time to examine the impact of benefit availability on enrollment in Medicare Advantage and health outcomes of enrollees. 

            Using county-level payment and enrollment CMS data from 2006 – 2010, I show that a $1 increase in the county-level benchmark translates to a $0.68 increase (p < 0.001) in plan rebates used for supplemental benefits such as drug coverage in the donut hole using fixed effect and first difference regressions.  Higher rebates are associated with increases in enrollment in MA and enrollment by sicker beneficiaries; a $10 increase  in rebates was associated with a 0.20 percentage point increase in MA penetration (p < 0.001) and a 7.1% increase in the average MA enrollee’s risk score (p < 0.001). 

            I will use CMS and State Inpatient Data from several large, geographically diverse states to estimate the effect of benefit generosity on potentially preventable hospitalizations among Medicare beneficiaries.  To address concerns that unobserved enrollee health characteristics may drive MA benefit availability and health outcomes, I will instrument for plan rebates using the strong first stage provided by the county-level benchmarks (F = 246).  Findings from this study will inform potential reforms to Medicare Advantage plans as well as benefit design for the new plans being offered through state and federal Health Insurance Exchanges.