Favorable Selection Among Dually Enrolled Beneficiaries In Private Medicare Plans

Medicare-Medicaid dual enrollees accounted for roughly a third of those programs' spending but less than 20 percent of enrollment in 2021. Policy makers have responded to dual enrollees' high levels of spending by encouraging their enrollment in private Medicare plans. However, assessing plans' impact on spending is complicated by favorable selection, in which healthier people prefer private plans to Medicare fee-for-service. We tested for selection into Medicare plans among dual enrollees, using linked Medicaid-Medicare claims data from the period 2017-22. We tracked people transitioning from Medicaid to dual enrollment and compared their pre-dual enrollment spending among Medicare plan choices. We found evidence of favorable selection. After adjustment for for beneficiary characteristics, a 1 percent increase in medical spending reduced beneficiaries' probability of enrolling in private plans by 1 percentage point (2.3 percent). The effects were driven by the highest-spending beneficiaries, who were 11 percentage points (25 percent) less likely to enroll in private plans. Selection appeared to be stronger among unintegrated plans (which cover Medicare but not Medicaid benefits), although higher spending on long-term services and supports reduced enrollment in all plan types. These findings highlight the need for researchers to control for underlying health status when evaluating health or spending outcomes in private plans.