Association for Public Policy Analysis & Management

Introduction: Research found that Medicare beneficiaries newly diagnosed with Alzheimer's disease and related dementias (ADRD) are more likely to switch from Medicare fee-for-service, known as Traditional Medicare (TM), to Medicare managed care plans, known as Medicare Advantage (MA) plans. This suggests that beneficiaries with ADRD may prefer MA plans because MA plans tend to perform better for those in need of care coordination while reducing unnecessary care, thereby achieving higher efficiency. However, since MA plans are paid on a capitated basis, they may design benefit designs in a way that can avoid high-cost beneficiaries, possibly leading to lower care satisfaction. However, little is known about health care utilization and care satisfaction between MA and TM beneficiaries with ADRD.

Objectives: We compared health care utilization and care satisfaction for MA and TM beneficiaries with ADRD and without ADRD.

Data: We used two datasets: the 2007-2012 Medicare Current Beneficiary Survey and the 2007-2012 Geographic Variation Public Use files from the Centers for Medicare and Medicaid Services.

Study Population: Our study population included 19,990 TM beneficiaries with ADRD (aged 65 and older with 12-months continuous enrollment), 691 MA beneficiaries with ADRD, 35,239 TM beneficiaries without ADRD, and 11,987 MA beneficiaries without ADRD. We identified ADRD cases through self-report in the survey.

Methods: Our outcomes were self-reported health care utilization (numbers of hospice visit, home health visit, inpatient hospital visit, outpatient hospital visit, medical provider visit, prescription drug use, institutional services use, facility services use, and dental services use, respectively) and care satisfaction (satisfaction on care costs, quality of care, and access to specialists). Our key independent variable was enrollment in MA plans. There is some evidence of endogenous plan choice. To control this endogenous plan choice, we performed a two-stage least squares regression model. We used county-level MA penetration rate as an instrumental variable. All models adjusted for individual-level demographic, socioeconomic, and health status characteristics as well as year and county-level fixed effects.

Results: We found that the county-level MA penetration rate was a strong instrument. For beneficiaries with ADRD, enrollment in MA plans was significantly associated with 21.21 fewer medical provider visits and 0.29 and 2.89 more institutional service uses and outpatient hospital visits, respectively. A similar trend was observed among Medicare beneficiaries without ADRD, but there were several differences. For beneficiaries without ADRD, enrollment in MA plans was significantly associated with 18.0, 1.14, and 0.12 fewer medical provider, outpatient hospital, and inpatient hospital visits, respectively, and 0.23 more dental services uses. No significant differences were detected in care satisfaction between MA and TM beneficiaries with ADRD. For beneficiaries without ADRD, however, enrollment in MA plans was significantly associated with lower satisfaction on access to specialists.

Conclusion: Our findings suggest that MA enrollment leads to a large reduction in medical provider visits for beneficiaries with ADRD without compromising care satisfaction. This implies that MA plans may have the potential to improve the efficiency of health care delivery for beneficiaries with ADRD.