Compared to Medicare-only beneficiaries, approximately 6.3 million dual eligibles in the U.S. are especially vulnerable and have high medical care costs [11
]. They are repeatedly reported as poor and underserved population [12
The dual eligibility program is designed to help low-income Medicare beneficiaries receive needed health care. As Medicare program's cost-sharing requirements – premiums, deductibles, and coinsurance – are often a financial burden to low-income beneficiaries and serve as a barrier to receiving needed care, federal and local governments have expanded over time Medicaid to certain eligibility groups whereby State Medicaid agencies are required to pay all or some of the Medicare out-of-pocket cost-sharing expenses for low-income Medicare beneficiaries that meet income and asset criteria [12
]. Collectively, these individuals are referred to as dual eligibles
Dual eligibles qualify for Medicare because they are aged 65 or over, disabled and receive Social Security Disabled Income (SSDI) assistance younger than age 65. Dual eligibles qualify for Medicaid because they are aged, blind, or disabled and meet the income and asset requirements for Supplement Security Income (SSI) assistance. Additionally, medically needy individuals qualify for Medicaid because they spend down a large portion of annual income and assets to pay for their medical or long-term care costs [13
]. Most of dual eligibles such as qualified Medicare beneficiaries (QMBs) and specified low-income Medicare beneficiaries (SLMB) are entitled to receive full Medicaid benefits. Others like qualifying individuals (OIs), however, are not entitled to full Medicaid benefits but subsidized for Medicare premiums and cost sharing [11
In this paper, we use the nationally representative data from Medical Expenditure Panel Survey (MEPS) to explore the relationship between dual eligibility for Medicare and Medicaid, and health care utilization. Understanding dual eligibles – population, their health care needs and health care usage – is the goal of this study for contributing to development of a relevant health policy. It seems useful to look at racial sub-groups of dual eligibles because racial differences may provide a clue to understanding heterogeneous effects of dual eligibility on health care use (and potentially unmet needs) of the poor and underserved population.
Dually eligible beneficiaries are better off than low-income Medicare-only enrollees who do not have dual coverage because duals are entitled to additional health benefits from Medicaid. We find that dual eligibility is positively correlated with the likelihood of using and the frequency of home health days but it is not significantly correlated with the frequency of office-based physician visits and hospital inpatient nights at the 5% level. The frequency of dental visits is inversely correlated with dual eligibility.
The significantly large effect of dual eligibility on the likelihood of using and the frequency of home healthcare receipt days is remarkable over the whole sample and particularly among Asian sample with dual eligibility. The large effect of dual eligibility on the home healthcare use could be explained by the level and scope differences in home health benefits between Medicare-only and Medicare-Medicaid dual eligibility programs. The traditional Medicare-only program does not provide home health benefits after 60-days. Medicaid, however, as a part of the dual eligibility program, takes almost full financial responsibility for the use of home health care.
Two competing theories may be involved in interpreting the dramatically increased utilization of home health care by dual eligibles: (i) overutilization by duals under the Medicare-Medicaid dual eligibility program in response to the additional home health benefits from Medicaid and (ii) underutilization by non-duals under the Medicare-only program because of its stringent home health benefits.
With similar generosity of Medicare-only and dual eligibility program for office-based physician services, the likelihood of using and the frequency of physician visits are not correlated with dual eligibility at the 5% level. Without differences in the level of benefits for hospital care, dual eligibility is not correlated with the frequency of hospital inpatient nights though it is statistically significant to increase the likelihood of using hospital care at the 5% level. The additional dental benefit provided by Medicaid is even inversely correlated with the frequency of dental visits though it is not statistically significant to decrease the likelihood of using dental services at the 5% level. These results suggest that overutilization theory does not seem to be persuasive. Dual eligibles do not always increase utilization of services in response to expanded benefits as shown in dental benefits newly covered by Medicaid.
Increasing the likelihood of using home healthcare and the high incidence rate ratio (IRR) of home health use by duals relative to Medicare-only beneficiaries (Table ) may be the result of delayed realization of their unmet needs under Medicare rather than the result of possible overutilization due to dual eligibility. The unmet need for home healthcare over the sample seems most urgent among all health care types examined by the study.
With respect to racial differences, dually eligible Afro-Americans use more office-based physician and dental services than white duals. Asian duals use more home health services than white duals at the 5% level. The dual eligibility program seems particularly beneficial to Afro-American duals. An epidemic like AIDS/HIV-positive, which is most prevalent among young Afro-American males, will not be treated under the traditional Medicare-only program while they are fully covered by Medicaid within the dual eligibility program [14
It is also true that there may exist inverse causality running from healthcare needs to dual eligibility: high levels of healthcare needs and usage may lead to qualifying for Medicare and Medicaid. Unravelling this complex relationship, further studies would need to control for potential confounding factors that may affect the relationship between dual eligibility and healthcare utilization.
In addition, it would be interesting to conduct an analysis separately for the disabled (e.g., Medicare beneficiaries under 65) and the elderly (e.g., Medicare beneficiaries 65 or older). Inherent differences between the two sub-groups may have affected or attenuated some of the relationships estimated by the study. To better understand the utilization differentials between the dually eligible and Medicare-only beneficiaries, further efforts may be required for examining the disease/individual-specific utilization patterns of the health care to consider potentially inherent heterogeneity in the characteristics of the dually eligible beneficiaries and their non-dual counterparts.