Skip to main content

Improving Care for High-Need, High-Cost Medicare Patients

Clinical evidence suggests that frail and chronically ill Medicare beneficiaries who are not dually eligible for full Medicaid benefits could often greatly benefit from the integration of non-Medicare-covered social supports into the medical care model offered to them in the Medicare program. For instance, non-Medicare-covered support services such as in-home meal delivery, non-emergent transportation to medical appointments, and targeted case management services have demonstrated the propensity for reducing the need for avoidable hospitalizations.

These interventions can be particularly valuable for Medicare beneficiaries who are not dually eligible for full Medicaid benefits, reside in the community setting, have three or more chronic conditions, and have functional or cognitive impairment. A data analysis performed on behalf of the Bipartisan Policy Center projects that approximately 3.65 million Medicare beneficiaries meet the above criteria; the analysis also projects that these beneficiaries incur roughly $30,000 in annual Medicare costs per beneficiary, or more than twice the national average annual Medicare Fee-for-Service spending amount per beneficiary.

However, Medicare’s payment rules and regulations have created significant care integration barriers for Medicare Advantage (MA) plans and health care provider groups, such as Accountable Care Organizations (ACOs) and patient-centered medical homes, which would otherwise furnish and finance these non-Medicare-covered supports and services. These barriers were outlined in great detail in a preliminary report, issued in February 2017 by BPC.

Read Next

1. Improving Medicare’s Risk Adjustment Model to Account for Functional Impairment:
In conjunction with action on the recommendation to waive uniform benefit requirements, the Centers for Medicare and Medicaid Services (CMS) should examine potential modifications to the CMS-Hierarchical Condition Categories (CMS-HCC) risk adjustment model to ensure more accurate predictions of medical expenses for the highest- and lowest-cost Medicare beneficiaries. Under this scenario, CMS could enhance existing survey platforms used to measure frailty and/or develop a functional status assessment tool that would be implemented and used across MA and Medicare FFS to evaluate and document functional impairment of the Medicare beneficiary. If this approach proves feasible—and data gathered by CMS support the use of a frailty adjustment factor or other functional status measure in the risk model—the otherwise applicable risk scores could be adjusted upward for MA plans and ACOs with higher proportions of beneficiaries with functional impairments, and adjusted downward for MA plans and ACOs that treat fewer beneficiaries with functional impairments. To allow MA plans and ACOs to become more familiar with the impact that a frailty or functional impairment adjustment factor would have on risk scores, benchmarks, and bid pricing, CMS should phase in any finalized adjustment factor policy over multiple years.

2. Incentivizing the Provision of Non-Medicare-Covered Supports Through Quality Measurement:
In conjunction with action on the recommendation to waive uniform benefit requirements, CMS should develop MA Star Ratings Program measures and ACO Quality Measurement metrics that evaluate the incorporation of non-Medicare-covered health-related social supports and services (that can be reasonably financed within existing MA rebates and ACO shared savings amounts) into the MA and ACO care model. MA plans and ACOs with greater levels of social support and service integration should be rewarded with higher scores on these quality measures, while MA plans and ACOs with less comprehensive integration of these services should receive lower scores. Additional quality measure-focused approaches could include applying one measure of all-cause hospital readmissions for beneficiaries with multiple chronic conditions and functional or cognitive impairments, and a separate measure of all-cause hospital readmissions for all other enrolled or attributed beneficiaries. For MA plans, consistent with past Medicare Payment Advisory Commission recommendations, and to the extent feasible, CMS should improve the Star Ratings Program by examining options for assessing these and other quality measures at the plan benefit package level, rather than the contract level.

In this report, BPC also provides details of a data analysis relating to the costs of providing four illustrative non-Medicare-covered supports. While BPC’s recommendations provide the flexibility for MA plans, ACOs, and health care providers to prescribe, furnish, and finance the specific non-medical support intervention that works best for a particular chronically ill Medicare patient, for the purposes of illustration, the analysis projected the costs of the following four services: in-home meal delivery, minor home modifications, non-emergent medical transportation, and targeted case management. The analysis suggests that, if changes to the uniform benefit requirement were adopted, MA plans would be able to finance many if not all of the four illustrative benefits (when targeted to high-need, chronically ill populations) by making only minor reductions in the value of existing supplemental benefits that the MA plans currently offer to all MA enrollees. The analysis also indicates that while some of the four illustrative supports could be financed by ACOs through shared savings payments, the ACOs may need to target these services to smaller groups of very high-need patients in order to make free provision of these supports financially viable.

Through the policy changes included in the recommendations of this report, the Medicare program could create pathways for MA plans, ACOs, and other providers to better tailor care plans for frail and chronically ill Medicare patients, in a manner that integrates traditional medical care with non-Medicare-covered social supports. Although work needs to be done to develop analogous solutions to address the non-Medicare-covered social support needs and chronic care management issues for beneficiaries who are enrolled in Medicare Fee-for-Service, these recommendations present the opportunity for tangible, bipartisan fixes to policy problems that have impeded the evolution of person-centered care in the Medicare program.