Medicare beneficiaries with multiple chronic conditions, certain types of serious conditions (e.g. heart disease, pulmonary disorders, mental disorders, cancer), and functional limitations have higher health and long-term care costs and more adverse outcomes than other beneficiaries.

One of the biggest opportunities for savings for Medicare, Medicaid, and beneficiaries themselves, is through reducing hospitalizations, readmissions, and institutional care, especially for these high-risk beneficiaries. Achieving these savings and serving this population will require innovative delivery models and a clear business case to convince organizations to implement those new models.

The Affordable Care Act set aside $10 billion for experiments in innovative care delivery and payment systems.  With these funds, the Centers for Medicare and Medicaid Innovation (CMMI) is launching and evaluating several initiatives, primarily Accountable Care Organizations (ACOs), bundled payment for care innovation, and primary care transformation.  These initiatives change financial incentives for health care providers so that while they bear some financial risk for the costs of providing care, they also stand to benefit from any savings produced.

Historically, it has taken additional legislative action to apply successful delivery models more broadly across the Medicare program. Now, the health care law has removed this barrier, giving the Secretary of Health and Human Services the ability to expand successful innovations that improve quality or lower costs.  While early results show improvements in quality and modest savings, most CMMI pilots and demonstrations to date are not specifically targeted on high-risk beneficiaries, where the biggest gains can be expected.

We propose a second stage of innovation that would focus on identifying promising delivery models specifically for high-risk beneficiaries in different settings including their homes, senior day centers, physician’s offices, hospitals, post-acute care, and nursing homes.  These models should place emphasis on lowering cost as well as improving the quality of life and care satisfaction for beneficiaries and their caregivers.

As with other ongoing experiments in CMMI, payment methods should be designed to provide an incentive for service providers to adopt and spread these models. Evaluation should take a system-wide view, not just a Medicare-centric focus on cost and quality of acute care.  Impacts on nursing home placement and long-term care services and supports should also be explicitly measured in the evaluation, as well as impacts on Medicaid, private plans, and beneficiary out-of-pocket costs.

Aligning Incentives

The problem of misaligned incentives is particularly acute for high-risk beneficiaries. Nearly one third of Medicare beneficiaries are enrolled in organizations with more aligned incentives such as Medicare Advantage plans or Accountable Care Organizations (ACOs).  These organizations take on more financial risk, such as capitated payments.

Yet, under the current system they do not benefit financially from reduced nursing home care and lower Medicaid expenditures. Nursing homes, on the other hand, do not benefit from reduced hospitalizations and Medicare savings, and actually receive higher reimbursement for short-term post-acute care beneficiaries than long-stay beneficiaries.

As a result, nursing homes and hospitals have little incentive to invest in strategies to keep high-risk beneficiaries in the community longer, or to improve the quality of care in nursing homes to reduce hospitalizations.  Hospitals that reduce institutional care, for example by preventing delirium in at-risk hospitalized patients, do not benefit from savings in long-term care despite the added staffing costs. Nursing homes typically lose Medicare revenues for post-acute care when residents are not hospitalized for acute illnesses. Further, the added costs of nursing home staffing and training to prevent or manage the conditions that typically require hospitalization are not compensated.

In caring for high-risk populations, non-traditional medical services may go a long way in preventing adverse outcomes. Traditional Medicare and most Medicaid programs do not cover the services of non-medical professionals or specially trained paraprofessionals that provide care or supportive services in patients’ homes such as social workers and handymen.  These services may prevent institutionalization, yielding long-term savings to Medicaid.

While not all at-risk Medicare beneficiaries are eligible for Medicaid, many low-income Medicare beneficiaries with functional limitations are likely to eventually qualify for Medicaid without assistance maintaining independent living. Programs such as CAPABLE or MIND at Home have shown significant improvements in well-being and reduced transitions to nursing homes, but they both entail the use of upfront expenditures for non-medical personnel not typically covered by Medicare.

Shifting these incentives will require changes in how both acute and long-term care are financed. Organizations considering taking on new risk will also need more empirical information on those effective innovations that cut across the entire care continuum and across all sources of financing care.

In particular, they will need evaluative data on the care quality and satisfaction, health outcomes, and cost savings of innovative delivery models. States such as Maryland with its new Medicare waiver based on achieving savings in Medicare, can also use evidence generated by delivery system innovations to design state-wide innovations.

Breaking Down Silos

A new round of CMMI innovations should identify high-risk beneficiaries who would benefit from these types of interventions, including but not restricted to the dual-eligible population. Tailored payment models need to be developed to generate a business case for providers to implement the models.

Shared savings may be sufficient incentive for organizations to shift care to a lower-cost setting such as Hospital at Home, which provides basic inpatient services in the individual’s home.  Nursing homes may need shared savings from reduced hospitalization to cover costs of training and supervising nursing home staff to prevent hospitalization.

Medicaid long-term care savings could also be shared with more comprehensive organizations such as ACOs or Medicare Advantage plans.  Organizations taking on more financial responsibility for both acute and long-term care, particularly small provider organizations, may need risk mitigation strategies such as reinsurance for unexpected high-cost cases.

What is most urgently needed to improve care and lower costs for the high-risk beneficiary population is an end to the silos that separate Medicare financing of acute care and Medicaid financing of long-term care.  An integrated strategy can and should align incentives across the entire care continuum, including long-term care.

The objective should be to incentivize care delivered in the least restrictive and least costly setting—in most cases that means care delivered at home and in the community rather than in a long-term institutional setting.  In the future, even more ambitious policies may be necessary. For example, a separate Medicare integrated program for high-risk beneficiaries, not just for dual-eligibles, could have its own set of benefits, delivery and payment models.

For now, CMMI has invited innovation awards, some of which focus on high-risk beneficiaries. But we believe, the agency should support a specific initiative coupling promising care delivery models for high-risk beneficiaries with a payment model encompassing all sources of acute and long-term care financing including Medicare, Medicaid, private plans, and beneficiaries.