The effect of home care use on institutional care utilization and expenditures.
详细信息
文摘
Over the past several decades,developing home- and community-based services HCBS) alternatives to institutional care has been a priority for many state Medicaid programs,ostensibly in order to meet strong preferences for home care and to reduce costs. However,limited evidence exists on whether expanding home care reduces institutional care use or saves money overall. This study aims to estimate the causal effect of using home care services on the utilization and costs of institutional medical care,including nursing facilities and inpatient care,and to provide necessary evidence for public policy on long-term care resource allocation. This study uses Medicaid claims data of adults from the Cash and Counseling Demonstration and Evaluation program in three states to estimate the causal effect of paid home care services on the costs and utilization of nursing facilities and inpatient medical care,free from selection bias in the choice to receive home care. An instrumental variable approach was applied to address the potential bias caused by measurement error of home care,essential heterogeneity / endogenous health care use,or the reverse effect of institutional care use. The instrumental variable exploits the randomized assignment to traditional Medicaid versus Cash and Counseling,which indirectly increased consumers ability to receive home care. Two-part models and appropriate transformations were employed in order to address skewness and zero-mass distribution of the data. The results suggest that the use of Medicaid-financed home care services significantly reduced utilization and Medicaid expenditures on institutional care. Home care use has strong negative effects on the likelihoods of both incurring any inpatient care and any Medicaid-financed nursing facility services. Within the preferred specification,a $1000 increase in Medicaid home care expenditures avoided 1.61 days in nursing facilities and reduced annual Medicaid nursing facility costs by $173 across the full sample when endogeneity is addressed. The Medicaid offset is an underestimate of the total offset,as Medicare also pays for inpatient care and some nursing home stays,costs which would also be reduced by fewer nursing home and hospital stays. Therefore,Medicaid-financed home care appears to partially offset institutional care use,which usually generates much higher costs. Failure to address selection biases would underestimate this effect or even lead to the opposite conclusion that the two are net complements. More generous funding of home care might be a potential way to provide cost-effective care,but a rigorous overall evaluation must include the tradeoff in benefits,i.e. a rigorous assessment of preferences and health outcomes in the two care settings.