Association of State Policies with Medicaid Disenrollment among Low-Income Medicare Beneficiaries

Background: Low-income Medicare beneficiaries who qualify for Medicaid often lack continuous enrollment in the program despite the substantial financial protections Medicaid confers. States have discretion to set policies that affect Medicaid eligibility, the ease of enrollment and recertification, and the value of coverage to enrollees, which may affect individuals’ ability and incentives to retain Medicaid. Although a large literature examines Medicaid coverage disruptions among non-elderly adults, no research has assessed the relationship between state policies and Medicaid disenrollment among Medicare beneficiaries, for whom the loss of Medicaid may be particularly disruptive given this population’s financial vulnerability, clinical risk factors, and age.

Objective: To assess patterns of Medicaid disenrollment among Medicare beneficiaries and examine the relationship between state policies and the loss Medicaid among Medicare beneficiaries who received either full or partial Medicaid (defining partial coverage as enrollment in one of the Medicare Savings Programs). Using contemporaneous enrollment data for the Part D Low-Income Subsidy (LIS)—a program with eligibility criteria that align closely with that of Medicaid for this population—we also examined whether Medicaid losses may have been unwarranted based on whether beneficiaries continued to receive the LIS following Medicaid disenrollment.

Design: We analyzed monthly Medicaid and LIS enrollment data in a 5% national sample of fee-for-service Medicare beneficiaries over the period 2012-2016. We assessed the proportion of Medicare beneficiaries who lost Medicaid for reasons other than death, the fraction who subsequently re-enrolled in Medicaid, and the proportion who continued to receive the LIS after disenrolling from Medicaid. We performed multivariate Cox proportional hazard regression models to assess the relationship between state policies and the likelihood of losing Medicaid adjusting for beneficiaries' health status.

Results: Approximately 18% of Medicare beneficiaries lost Medicaid for reasons other than death, 60% of individuals who disenrolled resumed this coverage within 2 years, and >50% continued to receive the LIS one year after disenrolling from Medicaid, suggesting that a substantial proportion of Medicaid losses may have been unwarranted. In multivariate survival analyses, we found that the likelihood of losing full or partial Medicaid was 22% lower in states that aligned Medicaid eligibility determinations with the federal Supplemental Security Income (SSI) program, 30% lower in states that raised or eliminated asset tests for partial Medicaid, and 32% lower in states whose Medicaid programs fully reimbursed providers for Medicare cost-sharing and deductibles that are waived for Medicaid recipients (vs. states that did not reimburse these amounts).

Conclusions: An appreciable number of Medicare beneficiaries lack continuous enrollment in Medicaid, despite the persistently low income and assets of this population. Many beneficiaries who disenrolled continued to receive prescription drug subsidies with eligibility requirements that closely align with Medicaid, suggesting that a large proportion of Medicaid coverage losses may have been unwarranted. Policies that make it easier for Medicare beneficiaries to maintain Medicaid eligibility (e.g., coordinating Medicaid and SSI eligibility determinations and lessening the stringency of asset tests) and that enhance access to care in Medicaid (via higher provider reimbursements) may reduce disenrollment.