Monday, April 21
Grant Miller: Managerial Incentives in Public Service Delivery
Nicholas, Lauren. 2009. "Who Joins Medicare Managed Care? Voluntary Enrollment and Positive Selection." PSC Research Report No. 09-670. February 2009.
Objective. To test for biased selection into and out of Medicare managed care (MMC), and to demonstrate a novel way of obtaining health status information about MMC enrollees.
Data Sources. Medicare enrollment limited data set (LDS) and Healthcare Cost and Utilization Project State Inpatient databases from Arizona, Florida, New Jersey and New York from 1999-2004 supplemented with data from the Area Resource File (ARF) and the Census bureau.
Study Design. Fixed effects linear probability models are used to examine the probability of MMC enrollment conditional on beneficiary demographic, socioeconomic and health status characteristics for a random sample of elderly Medicare beneficiaries.
Data Extraction. The 5 percent LDS enrollment file was merged with ARF and Census data. E-codes and ICD-9 codes were used to identify Medicare beneficiaries hospitalized for accidents and marker conditions in the SID to generate a plausibly random sample.
Principal Findings. Results provide evidence of biased selection into MMC; those with several comorbid conditions or more serious health conditions are likely to remain in Fee-for-Service Medicare, while MMC plans attract enrollees with uncomplicated diabetes and hypertension. Younger beneficiaries and minorities are more likely to enroll in MMC. Minorities, the previously disabled, and dual-eligibles are more likely to disenroll from MMC mid-year.
Conclusions. Higher payments to MMC plans relative to Fee-for-Service spending may not effectively target the beneficiaries with the greatest health and financial needs.
Country of focus: United States.