States With Medically Needy Pathways: Differences in Long-Term and Temporary Medicaid Entry for Low-Income Medicare Beneficiaries

Abstract

Medically needy pathways may provide temporary catastrophic coverage for low-income Medicare beneficiaries who do not otherwise qualify for full Medicaid benefits. Between January 2009 and June 2010, states with medically needy pathways had a higher percentage of low-income beneficiaries join Medicaid than states without such programs (7.5% vs. 4.1%, p < .01). However, among new full Medicaid participants, living in a state with a medically needy pathway was associated with a 3.8 percentage point (adjusted 95% confidence interval [1.8, 5.8]) increase in the probability of switching to partial Medicaid and a 4.5 percentage point (adjusted 95% confidence interval [2.9, 6.2]) increase in the probability of exiting Medicaid within 12 months. The predicted risk of leaving Medicaid was greatest when new Medicaid participants used only hospital services, rather than nursing home services, in their first month of Medicaid benefits. Alternative strategies for protecting low-income Medicare beneficiaries' access to care could provide more stable coverage.

Keywords: Medicaid; Medicare; dual-eligible beneficiaries; low-income populations; state policy.

Figure A1

Selection of study cohort.

Figure 1

Percentage of Part D low-income subsidy beneficiaries who joined Medicaid, by state’s medically needy income threshold. Note. States without medically needy program represented as having a $0 medically needy income limit. Pearson’s correlation coefficient (ρ) is weighted by the number of Part D low-income subsidy beneficiaries in each state.

Figure 2

Percentage of new Medicaid participants who exited Medicaid within 12 months, by state’s medically needy income threshold. Note. States without medically needy program represented as having a $0 medically needy income limit. Due to data privacy restrictions, graph excludes states with fewer than 350 new Medicaid participants. Pearson’s correlation coefficient (ρ) is weighted by the number of new Medicaid participants in each state.

Figure 3

Predicted differences in Medicaid exit by type of health care use in month of Medicaid entry. Note. Results adjusted for beneficiaries’ age, race, gender, Part D LIS participation immediately prior to Medicaid entry, cumulative health care use prior to Medicaid entry, socioeconomic quintile, January observations, and observations after 2010.